CN118505048A

CN118505048A - Method and device for mining cloud clients on enterprise, electronic equipment and storage medium

Info

Publication number: CN118505048A
Application number: CN202410612991.2A
Authority: CN
Inventors: 陈俭; 杨聃; 刘杰; 冉会琼; 何意; 罗莉
Original assignee: China Telecom Corp Ltd
Current assignee: China Telecom Corp Ltd
Priority date: 2024-05-16
Filing date: 2024-05-16
Publication date: 2024-08-16

Abstract

The embodiment of the invention provides a method and device for mining cloud clients on enterprises, electronic equipment and storage media. The method comprises the following steps: constructing an enterprise cloud indication label of each target enterprise based on the multi-source database; quantifying an enterprise cloud indication label of each target enterprise based on enterprise cloud influence factors, and determining quantized values of each cloud influence factor of the target enterprise; determining the cloud probability of each target enterprise based on the cloud influence factor quantification value; and determining cloud clients on the enterprises based on the cloud probability of each target enterprise. According to the cloud computing method and the cloud computing device, cloud impact factors of various enterprises are quantified, and the cloud computing probability of the enterprises is determined, so that cloud computing enterprise clients are more accurately recommended, the client demand range is more clear, the cloud computing success rate of the enterprises is improved, invalid marketing disturbance is reduced, high-quality service is provided for the clients more conveniently, and client service perception is enhanced.

Description

Method and device for mining cloud clients on enterprise, electronic equipment and storage medium

Technical Field

The present invention relates to the field of big data analysis technologies, and in particular, to a method for mining cloud clients on an enterprise, a device for mining cloud clients on an enterprise, an electronic device, and a storage medium.

Background

The cloud on the enterprise means that the enterprise deploys a basic system, business and a platform of the enterprise to the cloud through the Internet, and the network is utilized to conveniently acquire services such as calculation, storage, data and application, so that the informatization construction cost of the enterprise is reduced, the innovation development ecology of the industrial Internet is constructed, the optimization of the whole manufacturing process, the whole industrial chain and the whole product life cycle is promoted, and the fusion development level of the manufacturing industry and the Internet is promoted.

Cloud computing is a centralized embodiment of information technology development and service mode innovation, and is a great innovation and a necessary trend of informatization development. The cloud computing system supports enterprise cloud loading, is beneficial to promoting enterprises to accelerate digital, networked and intelligent transformation, and improves innovation capability, business strength and development level; the method is beneficial to accelerating the development of software and information technology service industry, deepening the structural reform of a supply side, promoting the deep fusion of Internet, big data, artificial intelligence and entity economy and accelerating the construction of a modern economic system.

Under the direction of various policies, cloud on enterprises is imperative. In order to accurately and effectively push cloud clients on enterprises, the cloud clients on the enterprises are required to be excavated and recommended by means of big data. In the prior art, cloud service entering of enterprises in converged enterprise services is separated, cloud service requirements are led into a safe and reliable private network of an operator, the problem of safe cloud entering of vast middle and small enterprises is solved, and a safe and reliable cloud entering channel is provided for the enterprises.

However, the above solution only solves the channel problem of the security requirement of the cloud enterprise in a targeted manner, belongs to a method for improving the cloud perception of the service enterprise after the enterprise determines the cloud, but does not discuss and analyze how to judge which enterprises have the cloud demand more front, and cannot solve the problem of judging whether the enterprises have the cloud demand, so that the cloud client mining is incomplete, and the cloud work cannot be advanced from the root mining client demand only for the cloud client of the existing system.

Disclosure of Invention

Aiming at the defects in the prior art, the embodiment of the invention provides a method for excavating cloud clients on an enterprise, a device for excavating cloud clients on the enterprise, electronic equipment and a storage medium.

In a first aspect, an embodiment of the present invention provides a method for mining cloud clients on an enterprise, including:

constructing an enterprise cloud indication label of each target enterprise based on the multi-source database;

Quantifying an enterprise cloud indication label of each target enterprise based on enterprise cloud influence factors, and determining quantized values of each cloud influence factor of the target enterprise;

Determining the cloud probability of each target enterprise based on the cloud influence factor quantification value;

and determining cloud clients on the enterprises based on the cloud probability of each target enterprise.

As in the above method, optionally, the multi-source database comprises at least one of the following databases:

enterprise operator databases, enterprise business information databases, and enterprise public information databases.

In the above method, optionally, the constructing, based on the multi-source database, an on-enterprise cloud indication tag of each target enterprise includes:

acquiring communication data of a target enterprise based on an enterprise operator database;

After preprocessing the communication data, determining the communication tag characteristics of a target enterprise;

and screening out the cloud indication tags on the enterprises of the target enterprises from the communication tag features.

acquiring business data of a target enterprise based on an enterprise business information database;

after preprocessing the business data, determining the business label characteristics of a target enterprise;

and screening out the cloud indication label on the enterprise of the target enterprise from the business label characteristics.

constructing a cloud bidding indication sub-label based on bidding information in an enterprise public information database;

Constructing a cloud-top official network indication sub-label based on the enterprise official network information in the enterprise public information database;

And determining an enterprise cloud indication label based on the cloud advertisement bid indication sub-label and the cloud advertisement web indication sub-label.

As above, optionally, the constructing a cloud bidding indicator sub-tag based on bidding information in the enterprise public information database includes:

Acquiring enterprise bidding information from enterprise public information;

After preprocessing the enterprise bidding information, determining enterprise bidding feature text of a target enterprise;

extracting key entity information from the enterprise bidding feature text based on a natural language processing technology to form a bidding information data source table;

Text word segmentation is carried out on the bidding information data source table based on a preset enterprise cloud dictionary, and a plurality of phrase sequences corresponding to the bidding information data source table are determined;

and acquiring the cloud-up bidding indication sub-label of the target enterprise from the phrase sequences based on a keyword extraction algorithm.

As above, optionally, the constructing the sub-tag for cloud-top indication based on the corporate network information in the corporate public information database includes:

acquiring enterprise website data from an enterprise official network of a target enterprise;

After preprocessing the enterprise website data, determining enterprise website feature text of a target enterprise;

Based on a natural language processing technology, extracting main business information and product information of the target enterprise from the enterprise website feature text;

judging whether the main business information and the product information of the target enterprise are matched with keywords in the preset enterprise cloud dictionary;

Judging whether the target enterprise has a cloud service business opportunity or not based on a keyword matching result;

If the cloud service business information exists, constructing an upper cloud official network indication sub-label of the target enterprise according to the main business information and the cloud business opportunity information of the target enterprise.

As in the above method, optionally, the cloud influencing factors on the enterprise include:

Business operating factors and business economy factors.

In the above method, optionally, the quantifying the cloud indication tag on the enterprise of each target enterprise based on the cloud influence factors on the enterprise, and determining the quantified value of each cloud influence factor of the target enterprise includes:

screening influence factor indication labels corresponding to enterprise operation factors from enterprise cloud indication labels of the target enterprises;

Performing label assignment on each influence factor indication label;

and calculating the enterprise operation factor quantification value of the target enterprise according to the label value of each label in the label indicated by the influence factor corresponding to the enterprise operation factor.

screening influence factor indication labels corresponding to enterprise economic factors from the cloud indication labels of the enterprises of the target enterprises;

Performing label assignment on each influence factor indication label;

And calculating the enterprise economic factor quantification value of the target enterprise according to the label value of each label in the label indicated by the influence factor corresponding to the enterprise economic factor.

As in the above method, optionally, the cloud-on-enterprise influencing factors further include: business expansion factors for enterprises.

acquiring an influence factor indication tag corresponding to an enterprise service expansion factor from an enterprise cloud indication tag of the target enterprise;

Performing label assignment on each influence factor indication label;

Determining a business indication influence factor of a target enterprise according to the occurrence time interval of the last cloud business behavior event of the target enterprise;

and calculating an enterprise business expansion factor quantification value of the target enterprise according to the tag value of each tag in the tag of the influence factor indication corresponding to the enterprise operation factor and the business indication influence factor.

The method optionally, performing label assignment on each influence factor indication label includes:

If the influence factor indicates that the tag is a category tag, determining a tag value of the category tag according to whether a category event corresponding to the category tag occurs or not;

And if the influence factor indicates that the label is a continuous label, carrying out box division processing on the continuous label based on a preset fraction, and determining the label value of the continuous label according to a box division result.

As above, optionally, the determining the cloud probability of each target enterprise based on the quantized value of the cloud impact factor includes:

and determining the cloud probability of each target enterprise based on the enterprise business factor quantitative value and the enterprise economic factor quantitative value of each target enterprise.

And determining the cloud probability of each target enterprise based on the enterprise business factor quantitative value, the enterprise economic factor quantitative value and the enterprise business expansion factor quantitative value of each target enterprise.

As above, optionally, the determining an enterprise cloud client based on the cloud probability of each target enterprise includes:

and screening cloud clients of the enterprises from the target enterprises according to a preset probability threshold and the cloud probability of each target enterprise.

The method as above, optionally, further comprising:

and recommending corresponding cloud services for each cloud client on the enterprise according to the cloud indication labels and the cloud service information on the cloud client on the enterprise.

In a second aspect, an embodiment of the present invention provides an apparatus for mining cloud clients on an enterprise, including:

the indication label construction module is used for constructing cloud indication labels on enterprises of all target enterprises based on the multi-source database;

The cloud influence factor quantification module is used for quantifying the cloud indication tags of the enterprises of each target enterprise based on the cloud influence factors of the enterprises and determining the quantification value of each cloud influence factor of the target enterprise;

the cloud probability determining module is used for determining the cloud probability of each target enterprise based on the cloud influence factor quantification value;

and the enterprise cloud client mining module is used for determining the enterprise cloud clients based on the cloud probability of each target enterprise.

In a third aspect, an embodiment of the present invention provides an electronic device, including:

The device comprises a memory and a processor, wherein the processor and the memory are communicated with each other through a bus; the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform a method of mining cloud customers on an enterprise as described in any of the first aspects above.

In a fourth aspect, an embodiment of the present invention provides a storage medium having stored thereon a computer program which, when executed by a processor, implements a method of mining cloud customers on an enterprise as set forth in any of the first aspects above.

According to the method for mining cloud clients on enterprises, provided by the embodiment of the invention, the cloud indication labels on the enterprises of all target enterprises are constructed based on the multi-source database; quantifying an enterprise cloud indication label of each target enterprise based on enterprise cloud influence factors, and determining quantized values of each cloud influence factor of the target enterprise; determining the cloud probability of each target enterprise based on the cloud influence factor quantification value; and determining cloud clients on the enterprises based on the cloud probability of each target enterprise. According to the embodiment of the invention, the cloud indication label of the enterprise is constructed based on the multi-source database, the cloud indication label portrait of the target enterprise is created, the cloud indication label portrait of the target enterprise is efficiently and comprehensively provided for clients of the enterprise, cloud influence factors of various enterprises are quantified, the cloud indication probability of the enterprise is determined, the cloud indication of the clients of the enterprise is more accurate, the client demand range is more clear, the cloud success rate of the enterprise is improved, invalid marketing disturbance is reduced, high-quality service is more conveniently provided for the clients, and the service perception of the clients is enhanced.

Drawings

FIG. 1 is a flow chart of steps of an embodiment of a method of the present invention for mining cloud customers on an enterprise;

FIG. 2 is a schematic diagram of a multi-source database composition in an embodiment of a method for mining cloud clients on an enterprise according to the present invention;

FIG. 3 is a flow chart of data acquisition of an enterprise network in an embodiment of a method of mining cloud clients on an enterprise in accordance with the present invention;

FIG. 4 is a flow chart of text recognition of an enterprise network in an embodiment of a method of mining cloud clients on an enterprise in accordance with the present invention;

FIG. 5 is a flow chart of the construction of an on-enterprise cloud indication library in an embodiment of a method of mining on-enterprise cloud customers of the present invention;

FIG. 6 is a view of a cloud target customer-oriented, scenerized, evaluated and quantified logic of an embodiment of a method of mining cloud customers on an enterprise in accordance with the present invention;

FIG. 7 is a block diagram of an embodiment of an apparatus for mining cloud customers on an enterprise in accordance with the present invention;

Fig. 8 is a block diagram of an embodiment of an electronic device of the present invention.

Detailed Description

In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.

Referring to fig. 1, a flowchart illustrating steps of an embodiment of a method for mining cloud clients on an enterprise according to the present invention may specifically include the following steps:

step S110, constructing enterprise cloud indication tags of all target enterprises based on a multi-source database;

Specifically, a large data acquisition technology is utilized to acquire a multi-dimensional external data source, the multi-dimensional external data source can be an operator data source, an industrial and commercial data source, a website information source and other public data sources, an enterprise multi-source database is constructed through the multi-dimensional external data source, and the multi-source database comprises various types of enterprise public information.

And screening all enterprises to determine target enterprises, for example, deleting enterprises which obviously have no cloud demand in an enterprise library to obtain the target enterprises, or screening the target enterprises according to screening conditions, wherein the screening conditions can be a certain area, a certain period of time, or screening according to other information, and the method is not limited.

The enterprise multisource big data is integrated and modeled through the multisource database, the cloud indication label on the enterprise of each target enterprise is built, and the cloud label portrait on the target enterprise is created through the cloud indication label on the enterprise, so that accurate insight of target clients of the cloud enterprise can be realized.

Step S120, quantifying an enterprise cloud indication label of each target enterprise based on the enterprise cloud influence factors, and determining quantized values of each cloud influence factor of the target enterprise;

specifically, an enterprise multisource database is constructed by acquiring and integrating different data source data, an enterprise cloud indication label set of a target enterprise is obtained based on the enterprise multisource database, cloud factors influencing the cloud of the enterprise exist through analysis of cloud enterprise clients discovery, such as economic factors, enterprise products and cloud service relevance factors, and the like, the cloud indication labels of the target enterprise are quantitatively modeled, each enterprise cloud indication label is assigned, and quantitative values of all influence factors in the target enterprise are obtained.

Step S130, based on the quantized values of the cloud impact factors, determining the cloud probability of each target enterprise;

Specifically, the cloud loading probability of the target enterprise is determined according to the cloud loading influence factor quantification value of each cloud loading influence factor in the current enterprise. For example, the weight of each cloud influence factor can be set, and the cloud influence factor quantized values of each cloud influence factor are weighted and summed to obtain the cloud probability of each target enterprise.

And step 140, determining cloud clients on the enterprises based on the cloud probability of each target enterprise.

Specifically, after the cloud probability of each target enterprise is determined, a probability threshold may be set, where if the cloud probability of the target enterprise is greater than the probability threshold, the target enterprise is used as an enterprise cloud client, and if the cloud probability of the target enterprise is lower than the probability threshold, the current enterprise cloud demand is lower. And excavating enterprise cloud clients with cloud requirements from a large number of target enterprises through the cloud probability.

In practical application, after the enterprise cloud indicating labels of all target enterprises are constructed in the multi-source database, an enterprise cloud prediction model can be constructed based on a machine learning algorithm, the enterprise cloud indicating labels are used as an input feature pool of the enterprise cloud prediction model, in the enterprise cloud prediction model, the enterprise cloud indicating labels are quantized by combining enterprise cloud influence factors, quantized values of all cloud influence factors are determined, quantized values of all cloud influence factors are obtained according to the quantized values of all cloud influence factors, and finally the cloud probability of each target enterprise is output.

According to the method for mining cloud clients on enterprises, provided by the embodiment of the invention, the cloud indication label on the enterprises is constructed based on the multi-source database, and the cloud label portrait on the target enterprises is created, so that efficient and comprehensive cloud client insight on the enterprises is realized, cloud influence factors on various enterprises are quantified, and the cloud probability on the enterprises is determined, so that the cloud clients on the enterprises are recommended more accurately, the client demand range is more clear, the cloud success rate of the enterprises is improved, invalid marketing disturbance is reduced, high-quality services are provided for the clients more conveniently, and the client service perception is enhanced.

Further, on the basis of the above embodiment, the multi-source database includes at least one of the following databases:

Specifically, fig. 2 is a schematic diagram illustrating the composition of a multi-source database in an embodiment of a method for mining cloud clients on an enterprise according to the present invention, where, as shown in fig. 2, the multi-source database may include at least one of an enterprise operator database, an enterprise business information database, and an enterprise public information database, for example, if the public information database is not queried, the enterprise operator database and the enterprise business information database may be used; if the operator used by the enterprise is an a operator and the operator needing to perform the cloud client mining on the enterprise is a B operator, the enterprise business information database and the enterprise public information database can be used so as to cover potential cloud clients on the enterprise, which are not the operators.

The enterprise operator database comprises communication data generated in the operation process of the operators, such as internet surfing behavior data, signaling position data, behavior consumption data and the like, the data can provide quasi-real-time enterprise communication service conditions, enterprise communication activity and other changes, and the quasi-real-time data can also reflect the current operation conditions of the enterprises. The enterprise operator database may also include operator data such as fixed internet protocol (Internet Protocol, IP) traffic data and Deep Packet Inspection (DPI) data.

The enterprise business information database comprises registration information in the business registration information, such as enterprise address information and enterprise network information; the enterprise business information, qualification certification, license, financial status and other data can embody the recent business direction and business condition of the enterprise.

Besides the enterprise operator database and the enterprise business information database, the data disclosed by the enterprise on the network side can be integrated into an enterprise public information database, such as data information of enterprise official websites, enterprise recruitment, enterprise public opinion and the like, and the data are actively disclosed by the enterprise, so that the actual operation and activity situation of the enterprise in operation and production can be embodied. The enterprise bidding information and the enterprise operation activity information published in the enterprise official network information in the part of data are particularly important, and the two parts of data source data are acquired, analyzed and integrated, so that cloud indication labels on related enterprises can be constructed.

The cloud indication label on the enterprise is built through the enterprise operator database, the enterprise business information database and the enterprise public information database, so that cloud-related business, cloud scale and enterprise informatization degree of the enterprise can be fully mined, the cloud demand on the enterprise can be described more accurately and reliably, and more targeted support is provided for subsequent cloud scheme establishment of enterprise clients.

The embodiment of the invention gives play to the advantages of double-layer identities of the communication operators and the cloud service operators, fully utilizes the enterprise operator database unique to the operators, integrates the enterprise business information database and the enterprise public information database, and creates a multisource database of enterprise clients in a full space. Meanwhile, aiming at enterprise clients of non-operators, the cloud-up indication labels of enterprises are built through the enterprise business information database and the enterprise public information database, and the recommended cloud-up clients are wider in coverage. By integrating internal and external data sources, the information island is broken through to form an operator-level enterprise end (B end) customer big data resource pool, the label system construction of the enterprise customer full life cycle is realized, accurate portraits can be carried out on the enterprise-level customer, and the enterprise customer is convenient for transverse and longitudinal integrated management.

On the basis of the foregoing embodiments, further, the constructing, based on the multi-source database, an on-enterprise cloud indication tag of each target enterprise includes:

Specifically, the enterprise operator database records enterprise communication data, such as: surfing behavior data, signaling position data, behavior consumption data and the like, wherein the data can provide changes such as near real-time enterprise communication service conditions, enterprise communication activity and the like, and the near real-time data can reflect the current business conditions of an enterprise. Communication data, including communication business data and behavior data, related to the target enterprise is fetched from the enterprise operator database, and the communication data is preprocessed, for example, cleaned and processed, and constructed as communication tag features for use.

And screening data labels closely related to the cloud demand based on communication label characteristics constructed in an enterprise operator database to obtain a data characteristic pool of the enterprise cloud prediction model, recording a data source list, and taking the data labels as enterprise cloud indication labels.

For example, an on-enterprise cloud indication tag determined from an enterprise operator database may be:

The method comprises the steps of product income amount, cloud product identification, cloud computer client identification, cloud host client identification, internet data center (INTERNET DATA CENTER, IDC) client identification, enterprise cloud disk activity jump identification, intelligent networking demand, industry video demand identification, cloud provider official network access times, cloud provider console access times, last access cloud provider console time, cloud server use identification and the like.

The embodiment of the invention fully plays the double-layer identity advantages of the communication operators and the cloud service operators, and builds the cloud indication label on the enterprise by applying the database base of the operators.

Specifically, the enterprise business information database comprises registration information, operation range, qualification certification, license, financial condition and other data in the business registration information, and can embody the recent business direction and operation condition of the enterprise. And (3) extracting the business data related to the target enterprise from the enterprise business information database, preprocessing the business data, such as cleaning and processing the data, and constructing and using the business label characteristics.

And screening data labels closely related to the cloud demand based on the industrial and commercial label characteristics constructed in the enterprise industrial and commercial information database to obtain a data characteristic pool of the enterprise cloud prediction model, recording a data source list, and taking the data labels as enterprise cloud indication labels.

For example, an on-business cloud indication tag determined from an on-business information database may be:

Unified social credit code, business name, business registration address, national economic industry code, national economic code name, camping service, registered capital (enterprise: ten thousand yuan), enterprise officer network self-research identification, informationized enterprise, software operation service (Software AS A SERVICE, saaS) identification, and the like.

In the embodiment of the invention, the cloud-up indication label of the enterprise is constructed based on the operator and business information base, the key feature data is extracted, the cloud-up client mining of the enterprise is completed, the efficient and comprehensive insight of the cloud-up client of the enterprise is realized, the operator and the cloud service provider are effectively assisted to push the cloud-up work, and the problem that the cloud-up client mining is incomplete due to the fact that only the client of the operator can be mined and other clients of the enterprise cannot be mined is solved through the business information.

Specifically, the enterprise can automatically disclose data, such as data information of enterprise official websites, enterprise recruitment, enterprise public opinion, and the like, on the network side besides the characteristics of the operator data and the enterprise business information data. The part of data is actively disclosed by the enterprise, and can embody the actual business activity condition of the enterprise in the business production. In the part of data, the enterprise bidding information and the enterprise official network information, such as the enterprise operation activity information published by the enterprise official network, are particularly important, and the two parts of data source data are acquired, analyzed and integrated by utilizing a big data technology, so that an upper cloud bidding indication sub-label and an upper cloud official network indication sub-label are respectively constructed, and the two are combined to be used as an upper cloud indication characteristic label of the enterprise.

The embodiment of the invention is based on an operator business information base, gathers and integrates the multisource information disclosed by the enterprise through a big data technology, combines the unique customer communication service data and behavior data analysis of the operator, extracts and processes key feature data, then completes cloud customer mining modeling on the enterprise through an algorithm, realizes high-efficiency and comprehensive cloud customer insight on the enterprise, realizes quantification on cloud-related service, cloud scale and information degree of the enterprise, is more accurate and reliable in describing cloud requirements on the enterprise, and provides more targeted support for subsequent cloud plan establishment of the enterprise customer.

On the basis of the above embodiments, further, the constructing a cloud bidding indicator sub-tag based on bidding information in the enterprise public information database includes:

Acquiring enterprise bidding information from enterprise public information;

Specifically, the bid data disclosed by the enterprise is a wind vane indicating the direction of development of the enterprise. Through timely and accurate acquisition of bidding information and analysis of various informationized construction projects, the cloud demand enterprises can be subjected to insight, and therefore potential cloud clients can be accurately excavated, and the method comprises the following specific steps:

step A1, collecting data sources: reliable enterprise public information data sources, such as government purchasing websites, business information platforms and the like, are selected, public data are acquired and arranged through a big data technology, enterprise informationized project bidding information data are collected, and timeliness, accuracy and comprehensiveness of data collection are ensured.

Step A2, data cleaning and pretreatment: and cleaning and preprocessing the collected enterprise bidding information to obtain enterprise bidding feature text.

The data cleaning and preprocessing comprises the following steps: and removing repeated data, processing missing values, normalizing field formats and the like to ensure the accuracy and consistency of the data. For example, for the case that there is a duplicate value for a data feature, one of the feature records is reserved, and the remaining duplicate feature values are directly deleted. And if the missing proportion of the features reaches a preset value (for example, 60 percent), the features are directly deleted, numerical features in the features with the missing proportion being smaller than the preset value (for example, 60 percent) are filled by an interpolation method, and character-type features are directly assigned with NAN values to obtain enterprise bidding feature texts.

Step A3, natural language processing (Natural Language Processing, NLP) technology application: processing enterprise bidding feature text by using an NLP technology, and converting text information into structured data which can be understood and processed by a computer, in particular purchasing profile, trading information and purchasing party information in a winning bid announcement, wherein winning bid party information; purchase profile in bid advertising, bid budget, buyer information, and the like. Meanwhile, the bidding text and the attachment information are processed, the specific content of the purchase is often displayed in the attachment in PDF and DOC formats, so that the corresponding attachment is required to be acquired and converted into text information, and a Named Entity Recognition (NER) technology is applied to recognize entity information in the text, such as enterprise names, places, dates and the like, and key entity information is extracted to form a bidding information data source table.

Step A4, dictionary construction and text word segmentation: and (3) pre-constructing an enterprise cloud dictionary, analyzing more than one hundred item service keywords such as a server, a host computer host, a data storage service, an internet data center (INTERNET DATA CENTER, IDC for short), a content delivery network (Content Delivery Network, CDN) and the like for expressing high-frequency keywords of the mainstream cloud service, forming the enterprise cloud dictionary, and facilitating accurate positioning of specific contents of bidding projects.

The word segmentation technique is used to segment the data in the bidding information data source table into word or phrase sequences so that the text data can be better processed and analyzed.

Step A5, generating labels from the keywords: based on the previous processing result, using TF-IDF (termfrequency _oversecultfrequency) keyword extraction algorithm, extracting the most representative and important project keywords from word or phrase sequences in the bidding information data source table, combining the project, business keywords and NER entity results to form the final enterprise cloud bidding indication sub-label.

For example, the cloud bid indication sub-label may be:

the number of the bid-winning informationized projects, the purchase amount of cloud storage resources, the purchase identification of a cloud host, the number of bid-winning projects, the time length of the last bid-winning clouding project, the total amount of bid-winning clouding projects in the last year of enterprises, the cloud project identification of bid-winning government class and the like.

The embodiment of the invention acquires the bidding information content of the client, analyzes the construction content by analyzing the bidding items of the client and processes and extracts the key business information by means of a natural language processing technology, thereby positioning the specific requirements of the client cloud business.

On the basis of the foregoing embodiments, further, the constructing a sub-label for cloud-top official website indication based on the corporate intranet information in the corporate public information database includes:

Specifically, based on enterprise record website information in an enterprise business information database or an enterprise public information database, text information in an enterprise domain name is captured through technical means such as web crawlers and big data analysis, and accurate business information is added to an enterprise through methods such as cloud business dictionary matching, natural language processing and the like, so that a full view of the enterprise information is constructed, later cloud recommendation is facilitated, and the method specifically comprises the following steps:

Step B1, acquiring website data: web crawler technology is used for capturing web page data of the enterprise network, and text information in relevant pages or specific labels is captured and stored in a file by analyzing a hypertext markup language (Hyper Text Markup Language, HTML) structure.

Step B2, data cleaning and pretreatment: and cleaning and preprocessing the captured webpage data, removing the HTML label, processing special characters, unifying text formats and the like, and obtaining clean text data serving as enterprise website characteristic text of a target enterprise.

FIG. 3 is a flowchart of acquiring enterprise network data in an embodiment of a method for mining cloud clients on an enterprise, as shown in FIG. 3, firstly acquiring an initial official website address list of the enterprise, judging whether the website content is allowed to be acquired, if yes, determining that the official website list can be acquired, adding a virtual request head, performing browser simulated access, acquiring network web page source code information, analyzing the network web page data of the enterprise network, obtaining related data according to requirements, downloading the related content, cleaning the data according to the requirements, forming regular data, and storing the regular data in a local database.

And B3, analyzing and extracting the text: the hosting business information and product information are extracted from the enterprise web site feature text using natural language processing techniques, such as text parsing and information extraction techniques. Algorithms such as keyword matching, entity recognition, etc. may be used to locate and extract relevant information.

Step B4, keyword matching: firstly, determining word libraries containing keywords related to cloud services, such as 'cloud computing', 'cloud service', 'cloud storage', 'software development', 'programming', 'API', and the like, in an enterprise cloud dictionary, and then performing similarity matching according to the keywords in the dictionary and an enterprise text analysis result, so as to judge whether the enterprise relates to the cloud services.

Fig. 4 is a flowchart of text recognition of an enterprise network in an embodiment of a method for mining cloud clients on an enterprise, as shown in fig. 4, inputting a text paragraph acquired by the enterprise network, and segmenting the text of the enterprise network by Jieba, wherein Jieba is an open-source Chinese word segmentation tool. And (3) performing similarity matching with a special word stock (keywords in an enterprise cloud dictionary) by stopping word stock segmentation optimization, if the similarity is greater than 70%, marking the enterprise identifier as 1, otherwise marking the enterprise identifier as 0.

Step B5, cloud demand identification: based on the keyword matching result, judging whether the enterprise has a business opportunity related to cloud service, classifying the enterprise by using a rule engine or a machine learning model, and judging whether the product of the enterprise is related to the cloud service and the potential degree of the demand. For example, an enterprise performs various activities related to cloud services, which indicates that the enterprise has cloud requirements, and the cloud service requirements are calculated and quantified through a rule model and a machine learning algorithm mode.

And generating corresponding labels according to the extracted main service information and cloud service business opportunity information. The keywords can be used as labels, and the labels can be generated by classifying according to the service types and the demand potential.

For example, the cloud-top sign sub-label constructed based on the corporate network information may be:

cloud agent identification, cloud provider identification, cloud planning identification on enterprises, informationized enterprise identification, cloud-related recruitment information identification, whether a machine room is built by itself, whether a server is owned by itself and the like.

According to the embodiment of the invention, a multidimensional external data source is obtained through a big data acquisition technology, a machine learning algorithm and a natural language processing technology are applied, and a DPI analysis technology unique to an operator is combined to image a demand customer, so that deep analysis of a customer service pain point is realized, big data modeling is further completed, and an upper cloud support service is accurately provided for guiding the demand.

Further, on the basis of the above embodiments, the cloud-on-enterprise influencing factors include:

Business operating factors and business economy factors.

Specifically, by analyzing cloud-up enterprise clients, factors affecting cloud-up of an enterprise can be classified into three types, one factor is a business environment factor, namely, cloud-up conditions of an enterprise occur due to national policy or government requirements, and one factor is an enterprise operation factor, because the enterprise has actual business requirements, cloud-up needs to be performed to obtain cloud resources, and in addition, enterprise economic factors are basic factors of whether the enterprise is in consideration of cloud-up.

The Cloud resource management method is characterized in that the Cloud resource management method aims at the Cloud influence factor of a Cloud-up 'just-needed' enterprise (Cloud HARD NEEDS, CHN), and the Cloud-up 'just-needed' enterprise refers to an enterprise which is not Cloud-up, but Cloud resources are closely related to actual business production of the enterprise, and the Cloud resource management method is an indispensable foundation in the actual operation process or in external projects.

In addition, there are cloud-up "improved" enterprise clients (Cloud Improve Demand, CID), which refer to enterprise clients that have some cloud resources, but cloud resource update and cloud service expansion factors may need to increase the number of cloud resources or improve the cloud resource performance. For the cloud-up 'improved' enterprise clients, in addition to the three cloud-up influencing factors, enterprise business expansion factors also need to be considered. The enterprise business expansion factor refers to that because of the expansion of the enterprise business, the original cloud resources cannot support the development of the enterprise business and the expansion of the cloud resources is needed.

The cloud indication labels on the enterprises are divided according to the four factors influencing the cloud on the enterprises, wherein the enterprise environment factor index serves as an administrative policy factor aiming at the industry, so that the labels with the policy requirements can be quantified. The other three factors have complex influence, and analysis and quantization can be performed by constructing a quantization function.

The embodiment of the invention takes enterprise business information, operator data and enterprise public information as the basis, fully considers the influence degree of enterprise economic factors, enterprise operating factors and enterprise service expansion factors on enterprise cloud, and deep excavates cloud-up 'improved' enterprise clients and 'just needed' potential cloud-up enterprise clients, thereby observing cloud-up target clients, improving the cloud-up success rate of the enterprise, reducing invalid marketing harassment, providing high-quality service for the clients more conveniently, enhancing customer service perception, trampling cloud support mission of cloud platform service providers, steadily and solidly propelling the work of providing technical support service for the cloud-up enterprises, and helping the clients to realize digital transformation and service innovation.

Based on the foregoing embodiments, further, the quantifying the cloud indication label on the enterprise of each target enterprise based on the cloud impact factors on the enterprise, and determining the quantified values of the cloud impact factors of the target enterprise includes:

Performing label assignment on each influence factor indication label;

Specifically, the influence factor indication label corresponding to the enterprise operation factor is screened out from the cloud indication label of the enterprise of the target enterprise and is used as an enterprise operation business indication (ENTERPRISE BUSINESS OPERATIONS, EBO).

The enterprise business indicator EBO mainly comprises the indicating data related to the cloud resource demand degree of the enterprise in the actual production and management process. By analyzing the use mode of related cloud resources after handling the cloud resources by the cloud-up enterprise clients, the cloud resources handled by the enterprise are mainly used for meeting the cloud resource requirements of the enterprise in the self informatization mode and two actual production and operation scenes of cloud resource requirements when external operation is carried out except for policy requirements. Therefore, quantitative analysis can be performed on the two main scenes, and the cloud resource demand degree of enterprise business can be well displayed.

On the basis of the analysis, corresponding indication labels are respectively screened from an enterprise cloud indication label library of a target enterprise aiming at two scene characteristics, wherein the indication labels reflecting own requirements are as follows: whether the enterprise is a software company, accesses the cloud provider officer network frequency, whether to teach the company and other labels, etc.; indication labels related to the external operation cloud resource requirements of enterprises, for example: whether SAAS suppliers, whether cloud resource projects are bid, cloud-related recruitment information, and the like.

The index tag analysis for two scenarios finds that the index tag is composed of a category type tag and a continuous type tag, the category type tag being for example: whether the enterprise is a software company, a SAAS vendor, etc. contains information on whether a category-related event has occurred, continuous tags such as: and information related to the numerical value such as the cloud recruitment quantity, the bid amount in the cloud resource project and the like.

In order to reduce information loss, the influence degree of two scenes on the enterprise cloud resource requirement is fully obtained, the indication labels are divided into category labels and continuous labels according to label types, quantization formulas are respectively constructed, and influence degree is calculated.

And for the category type label, marking as a, defining that if the corresponding category related event occurs, the label is assigned as1, and if the corresponding category related event does not occur, the label is assigned as 0. And then summing all the category labels, carrying out normalization processing, recording p category labels, recording the comprehensive influence of all the category labels on cloud on an enterprise as EBO1, and obtaining a category label influence degree quantification formula:

Wherein, eba ₁ is the quantitative value of the comprehensive influence of the type tag on the cloud on the enterprise, a _i is the tag value (0 or 1) of the ith type tag in the enterprise business sign eba, and p is the number of type tags in the enterprise business sign eba.

And b is marked for continuous labels such as indication labels of cloud recruitment quantity, bid amount in cloud resource projects and the like. The indication labels and the enterprise cloud business operation situation show positive correlation, so that the indication is directly mapped forward according to the value, continuous labels can be subjected to box separation processing based on a preset fractional number for conveniently fusing different relevant influence factors, after a classified indication set after box separation is obtained, each indication box is respectively assigned from small to large according to the value, and a corresponding value is assigned when the indication falls in a relevant interval, so that the indication box is used as a quantitative value of the influence degree of relevant characteristics.

For example, if the preset quantile is the quantile, the quantile is used for carrying out the box-dividing processing on the continuous labels, after the four-classification label set after the box division is obtained, each label box-dividing is respectively assigned with the value from small to large: 0.25,0.5,0.75,1, if the label falls in the relevant interval, a corresponding value is assigned, the value is taken as an influence degree quantization value of relevant characteristics, b is marked, the number of the indications is q, and a continuous label influence degree quantization formula is obtained based on the analysis:

Wherein, EBO ₂ is the comprehensive influence quantification value of the continuous type label on the cloud on the enterprise, b _i is the label value of the ith continuous type label in the enterprise business indication EBO, and q is the number of continuous type labels in the enterprise business indication EBO.

Based on the quantitative calculation result of the enterprise business indication, carrying out summation normalization to finally obtain an influence degree quantitative formula of the enterprise business indication EBO:

Wherein, EBO is the quantized value of the enterprise operation factors.

In the embodiment of the invention, the enterprise operation business indication is constructed by applying the cloud indication label on the enterprise, the indication factors influencing the cloud on the enterprise are creatively converged and converged, the influence degree of the cloud on the enterprise is evaluated according to different aspects of the influence on the cloud on the enterprise, the quantitative calculation of the indication is realized, and the success rate of the cloud on the enterprise is improved.

Performing label assignment on each influence factor indication label;

Specifically, the influence factor indication label corresponding to the enterprise economic factor is screened out from the cloud indication label of the enterprise of the target enterprise and is used as the enterprise economic capability indication (ENTERPRISE ECONOMIC CAPACITY, EEC).

The enterprise economic capability indicator EEC represents enterprise economic strength, and includes indicators of whether the enterprise has recruits for people in the last half year, whether the enterprise has bid-winning projects, registered funds, social security payment numbers and the like. In order to fully consider the influence of the enterprise economic capability indication EEC on the cloud on the enterprise, the comprehensive quantification is performed by constructing an enterprise economic capability function.

Like the business operation indicator EBO, the business economic capability indicator EEC is also composed of two types of labels, one is a category label, and is marked as c, for example, "whether the business recruits people in the last half year" and "whether the business has a bid item in the last half year", and for this part of the indicator labels, whether an event occurs can be quantified, so that if the related event occurs, the label is assigned with 1, and if not, the label is assigned with 0. And then summing all the labels and carrying out normalization processing, and recording n related labels, wherein the related influence is recorded as EEC1, so as to obtain a quantitative formula of the influence degree of the category type label:

wherein EEC1 is the comprehensive influence quantification value of the type label on the cloud of the enterprise, c _i is the label value of the ith type label in the EEC indicated by the economic capability of the enterprise, and n is the number of the type labels in the EEC indicated by the economic capability of the enterprise.

For another continuous type of indication of the economic impact of the business, such as registered funds, social security contributions, etc., d is noted. By analyzing the correlation between the correlation indication and the enterprise economic capability, the value of the continuity label is in positive correlation with the enterprise economic capability, the larger the label value is, the stronger the enterprise economic capability is, and the smaller the label is, the weaker the enterprise economic capability is. Therefore, the labels can quantify the economic capacity of enterprises according to the value, in order to more intuitively quantify the related labels, the continuous labels are firstly classified based on quartiles, and after four classified labels after the classification are obtained, each label is respectively assigned with the value from small to large: 0.25,0.5,0.75,1, if the label falls in the relevant interval, a corresponding value is assigned, the value is taken as a quantized value of influence degree of the label on the economic capacity of an enterprise, the quantized value is marked as d, the number of the labels is m, and a continuous label influence degree quantized formula is obtained based on the analysis:

Wherein EEC ₂ is the quantitative value of the comprehensive influence of the continuous type label on the cloud of the enterprise, d _i is the label value of the ith continuous type label in the EEC for indicating the economic capability of the enterprise, and m is the number of continuous type labels in the EEC for indicating the economic capability of the enterprise.

Based on the quantized calculation result of the enterprise economic capability indication, carrying out summation normalization to finally obtain an influence degree quantization formula of the enterprise economic capability indication EEC:

EEC is an enterprise economic factor quantification value.

In the embodiment of the invention, the enterprise economic capability indication is constructed by applying the cloud indication label on the enterprise, the indication factors influencing the cloud on the enterprise are creatively converged and converged, the influence degree of the cloud on the enterprise is evaluated according to different aspects of the influence on the cloud on the enterprise, the quantitative calculation of the indication is realized, and the success rate of the cloud on the enterprise is improved.

Performing label assignment on each influence factor indication label;

Specifically, the influence factor indication label corresponding to the business expansion factor of the target enterprise is screened out from the cloud indication label of the target enterprise and is used as the business expansion indication CSBE (Cloud Service Business Extension) of the enterprise.

The enterprise cloud service expansion indication CSBE is a quantitative value of cloud resource expansion demand degree of the cloud clients on the cloud, which is based on the behavior of the enterprise in the use process of the cloud resources of the operator and the condition of the enterprise accessing related cloud products, and mainly refers to that the cloud resources originally purchased by the cloud clients on the cloud cannot meet the service demand due to service expansion, so that the cloud resources are required to be expanded. The indication is mainly constructed based on the analysis of the behavior of the enterprise in the use process of the cloud resources of the operator and the condition of the enterprise accessing related cloud products, such as the number of times of accessing cloud disk pages, the number of websites of different cloud resource suppliers, the order amount of the cloud products and the like. The same as the business operations indicator EBO and the business economic capability indicator EEC, including category type tags and continuous type tags, category type tags such as: the cloud disk purchase page is accessed, and the continuous tags such as the cloud product order amount and the like.

After analysis of the indication, the enterprise's willingness to transact cloud business is found to have strong correlation with the enterprise's time of occurrence in addition to the specific content of occurrence. For example, the number of times that an enterprise accesses a cloud disk ordering page has an influence on enterprise cloud resource ordering, besides the frequency of access has a strong correlation with the latest access time, the closer the access time interval statistics time window is, the greater the possibility that the enterprise handles cloud products, if the access time of the cloud disk ordering page is far from the statistics time window, the influence degree can be suddenly reduced, so that in order to more comprehensively and accurately quantify enterprise service expansion instructions, the influence of time factors needs to be taken into consideration.

The time interval of the last cloud service behavior event is recorded as J, the service indication influence factor of the time interval factor on the service behavior indication is beta, and because of the negative correlation between the time interval and the service behavior, the larger the value of the time interval J is, the smaller the influence of the correlation indication on the enterprise cloud service handling is, so that the negative correlation exists between the time interval J and the service behavior, and the quantitative assignment is carried out on the time interval J for better quantifying the influence degree of the correlation indicator, as follows:

j is less than or equal to 7 days, and the beta value is 1;

j is greater than 7 days and less than or equal to 15 days, and then the beta value is 0.75;

J is greater than 15 days and less than or equal to 30 days, and then the beta value is 0.5;

J is greater than 30 days and less than or equal to 45 days, and the beta value is 0.25;

j is greater than 45 days, then the beta value is 0.

Calculating influence factor comprehensive evaluation indexes in a given time interval of enterprise cloud service expansion factors according to an enterprise economic capacity factor comprehensive evaluation index quantification method, meanwhile adding a service index influence factor beta as a weight to obtain a product, taking the product as a quantification value of a single index, and finally summing all the indexes and normalizing the indexes, wherein the calculation formula is as follows:

wherein CSBE is an enterprise service expansion factor quantization value, e _i is a tag value of the ith type tag in the enterprise cloud service expansion indication CSBE, and p is the number of the CSBE type tags in the enterprise cloud service expansion indication.

According to the embodiment of the invention, creatively converging and converging the index factors influencing the cloud on the enterprise, subdividing the index factors according to different aspects of influence on the cloud on the enterprise, dividing the index factors into three index sets of an enterprise economic capacity factor index set, an enterprise business factor index set and an enterprise cloud business expansion factor index set, and simultaneously respectively constructing an enterprise economic capacity factor comprehensive evaluation index, an enterprise business comprehensive evaluation index and an enterprise cloud business expansion factor comprehensive evaluation index on the basis of respectively analyzing the influence degree of the influence index in the three sets on the cloud on the enterprise, and evaluating the influence degree of the three types of indexes on the cloud on the enterprise, thereby realizing quantitative calculation of the three types of indexes.

On the basis of the foregoing embodiments, further, the determining, based on the quantized values of the cloud impact factors, the cloud probability of each target enterprise includes:

Specifically, in the process of actually confirming cloud target enterprise client scenes, a plurality of reasons for influencing cloud requirements of enterprises are found, and the relations are complicated, so that in order to better improve the accuracy of locking the cloud target clients, targeted carding analysis is needed for cloud requirement scenes of the enterprises, so that the main influence scenes are obtained through insight, and the focusing of the requirement scenes is realized.

In order to realize the focusing of a demand scene, constructing an index set based on a main influencing factor scene, and finding out three main reasons influencing the cloud of an enterprise by carrying out statistical analysis on cloud-up enterprise clients, wherein one of the three main reasons is related to a business environment policy, the government has policy requirements on the cloud of related industries or the cloud of the enterprise can give higher subsidy, the clients are environment influencing, the influence cloud factors are single and are only related to the industry attributes of the enterprise clients and have no too many business trend factors, and target clients can be directly locked according to the industry attribute indexes; the other two types have strong correlation with the business requirements of enterprises, and multiple influencing factors are needed to be comprehensively considered, wherein one type is an enterprise which is not in Cloud, but Cloud resources are closely related to the actual business production of the enterprise, and the Cloud resources of the part of enterprises are essential bases in the actual business process or in external projects, and belong to 'just-needed' CRD (Cloud RIGID DEMAND) clients for the Cloud requirements; one type is enterprise clients with partial cloud resources, but cloud resource updating and cloud service expansion factors may need to increase the number of cloud resources or improve the performance of the cloud resources, and belongs to cloud resource 'improved' CID (Cloud Improve Demand) clients.

Under comprehensive analysis, the first scene target enterprise client can be directly and accurately locked through the policy requirement, and the actual meaning of prediction is not great, so the embodiment of the invention mainly focuses on the last two demand scenes.

The method is combined with a constructed enterprise cloud influence factor index set (an enterprise business operation business index EBO, an enterprise economic capability index EEC and an enterprise cloud business extension index CSBE) and two enterprise cloud demand scenes, and the influence factor index set is divided into two types of index libraries, namely an enterprise cloud business index library related to the actual use of an enterprise and an enterprise business operation business index library related to the actual business operation of the enterprise.

Fig. 5 is a flow chart for constructing an enterprise cloud indication library in an embodiment of a method for mining cloud clients on an enterprise, as shown in fig. 5, the enterprise information data, the communication data, the business data and the operation data are obtained from an enterprise operator database, an enterprise business information database and an enterprise public information database, enterprise cloud indication tags of each target enterprise are constructed, the enterprise cloud indication tags are grouped according to cloud services and enterprise operation, an enterprise cloud service trace feature library and an enterprise operation service feature library are obtained, the enterprise operation service feature library and enterprise cloud service indication are constructed, and finally the enterprise cloud indication library is obtained.

According to the cloud demand scenes, the influence factor indicators are analyzed and screened, wherein the enterprise owner of the 'just-needed' type is influenced by the economic capacity of a company and the correlation of the actual business operation of the enterprise, so that the enterprise business operation related indicators and the enterprise basic related information of the users are collected to obtain an enterprise cloud business indicator set, and then the indicators in the indicator set are mapped and divided into the enterprise economic capacity indicator set and the enterprise actual business operation indicator set. According to the quantization method of the enterprise actual operation business indication and the enterprise economic capability indication, the cloud probability of the cloud-up 'just-needed' CHN enterprise client is calculated in a quantization mode, and the calculation formula is as follows:

wherein CHN is the cloud probability of the 'just-needed' enterprise client, EBO is the enterprise business factor quantitative value of the 'just-needed' enterprise client, and EEC is the enterprise economic factor quantitative value of the 'just-needed' enterprise client.

And calculating based on the formula, so as to obtain the cloud probability value of the cloud 'just-needed' enterprise client.

According to the embodiment of the invention, the demand scenes influencing the cloud on the enterprise are creatively analyzed by using the scene-based thinking, the enterprise is divided into the business environment factor influence scenes and the cloud demand guiding scenes according to different requirements of the enterprise on cloud resource demands, meanwhile, the cloud demand guiding scenes are further subdivided, the cloud 'just-needed' scene enterprise guest group is creatively provided, the cloud probability index model is constructed and evaluated, the cloud probability index model of the enterprise client in the corresponding scene is quantized, and a new service popularization mode is opened up for the targeted client screening and scene marketing of the service marketing department.

Specifically, based on cloud-up "improved" CID enterprise scene analysis, the cloud-up probability of the relevant enterprise is closely related to the enterprise cloud service expansion indication factors besides being influenced by the economic capacity of the enterprise and the correlation of the actual business operation and service of the enterprise, so that the influence of the three indications needs to be fully considered for the mining of the target users of the part of the enterprise. Meanwhile, correlation analysis finds that three influence indicators have great difference on the influence of an upper cloud 'improved' enterprise, and the influence degrees are ranked according to the sizes as follows:

"Enterprise cloud service extension sign" > "Enterprise economic Capacity sign" > "Enterprise operation service sign"

Based on the analysis result, adding influence degree weights to different types of indications, and marking asTherefore, after the cloud probability of the cloud-up 'improved' enterprise client is calculated, the quantization result of the corresponding influence index set is weighted and normalized, and the calculation formula is as follows:

wherein the method comprises the steps of And is also provided withCID is the cloud probability of the 'improved' enterprise client, CSBE is the enterprise business expansion factor quantification value of the 'improved' enterprise client, EEC is the enterprise economic factor quantification value of the 'improved' enterprise client, and EBO is the enterprise business factor quantification value of the 'improved' enterprise client.

And calculating according to the formula to obtain the cloud business handling probability value of the cloud-up 'improved' enterprise.

According to the embodiment of the invention, the demand scenes influencing the cloud on the enterprise are creatively analyzed by using the scene thinking, the enterprise is divided into the business environment factor influence scenes and the cloud demand guiding scenes according to different requirements of the enterprise on cloud resource demands, meanwhile, the cloud demand guiding scenes are further subdivided, the cloud 'just-needed' scene enterprise guest group and the cloud 'improved' scene enterprise guest group are creatively provided, the cloud probability index model is respectively constructed and evaluated for the two scenes, the cloud probability of enterprise clients in the corresponding scenes is quantified, and a new service popularization mode is developed for the targeted client screening and scene marketing of the service marketing department. Both "retrofit" and "on demand" potential cloud clients can be mined and include clients other than the operator, with more comprehensive coverage of the recommended cloud clients.

Fig. 6 is a logic diagram for evaluating and quantifying a cloud target customer in a cloud customer on an enterprise in an embodiment of a method for mining an enterprise of the present invention, as shown in fig. 6, an enterprise economic capability indication EEC, an enterprise business indication EBO and an enterprise cloud business extension indication CSBE are constructed through a cloud customer on an enterprise indication library, a cloud customer on-demand type on-demand probability value is determined through the enterprise economic capability indication EEC and the enterprise business indication EBO, and an cloud business handling probability value is determined through the enterprise economic capability indication EEC, the enterprise business indication EBO and the enterprise cloud business extension indication CSBE.

According to the embodiment of the invention, the cloud demand enterprises are judged based on the communication big data and the cloud service handling enterprise data of the operators, so that the cloud demand enterprises are more accurately recommended, the client demand range is more clear, the cloud of the enterprises is purposefully promoted, reliable, safe and high-performance cloud computing service is provided for the enterprise clients, and the enterprises are helped to realize digital transformation and service innovation.

On the basis of the foregoing embodiments, further, the determining, based on the cloud probability of each target enterprise, an enterprise cloud client includes:

Specifically, in the actual business side application quantization result process, in addition to considering achieving the business goal by achieving the cloud scale on the enterprise, the marketing cost and the efficiency problem need to be considered, so that a proper preset probability threshold needs to be drawn according to the actual marketing cost to control the number of cloud goal clients on the recommendation.

The enterprise clients with the cloud demand are accurately insight through constructing a big data indication quantification mode, all enterprise clients are marked with the cloud probability value, and the cloud demand quantification of the enterprise is achieved. And providing data support for a business director to determine an enterprise cloud business recommendation target, and providing a marketing customer group by means of issuing a marketing target enterprise list with an enterprise cloud probability value. If the highest probability value of the recommended clients is 0.8, then 20% clients with probability values greater than 0.64 are preferentially selected according to the two-eight principle to conduct trial marketing application (the probability threshold value of the recommended clients of the upper cloud=1-the highest probability of the output of the model is 20%), and then the number of the recommended cloud target clients is controlled according to the actual effect reduction threshold screening condition, so that low-cost and high-conversion of business marketing is achieved.

On the basis of the above embodiments, further comprising:

Specifically, cloud indication tags on enterprises are clustered and clustered, cloud requirements on the enterprises are confirmed according to clustering characteristics, and a customized solution of cloud is formulated according to cloud service information and based on the requirements of the enterprises.

And clustering target enterprises according to different business scenes by constructing a clustering algorithm, recommending targeted cloud products by adopting different marketing strategies for different enterprise groups in the process of recommending cloud products, and realizing the re-promotion of marketing benefits. If the customer has the characteristics of external service, winning informationized project, accessing the cloud control console and the like of the SaaS application, the customer may be a software integrator, the customer has clear and just needs a cloud server, cloud storage resources and the like, and later service recommendation is taken as an important point. The customer grouping analysis is used for providing basis support for marketing resource allocation and cloud solution establishment in the specific marketing process of enterprise customer marketers.

According to the embodiment of the invention, a multidimensional external data source is obtained through a big data acquisition technology, a machine learning algorithm and a natural language processing technology are applied, and a DPI analysis technology unique to an operator is combined to image a demand customer, so that deep analysis of customer service pain points is realized, big data modeling is further completed, the demand is used as a specific demand for guiding and positioning customer cloud services, and corresponding cloud services are handled by a recommending enterprise according to the demand, so that cloud recommendation, namely a cloud service scheme containing the demand is realized.

The method for mining cloud clients on enterprises provided by the embodiment of the invention has the following beneficial effects:

(1) By applying the communication data characteristic pair cloud client mining method, a multi-source database consisting of three data sources of enterprise business information data, operator data and enterprise public information data is constructed, and quantification of enterprise cloud-related business, enterprise cloud scale and enterprise informatization degree is realized through the constructed multi-source enterprise cloud indication label, so that the description of cloud requirements on the enterprise is more accurate and reliable, and more targeted support is provided for subsequent cloud scheme establishment of the enterprise client.

(2) The method comprises the steps of converging and converging index factors influencing cloud on an enterprise, subdividing the index factors according to different aspects of cloud influence on the enterprise, dividing the index factors into three index sets of enterprise economic capacity factor index sets, enterprise business factor index sets and enterprise cloud business expansion factor index sets, respectively analyzing influence degrees of the index on the cloud on the enterprise in the three sets, respectively constructing enterprise economic capacity factor comprehensive evaluation index, enterprise business comprehensive evaluation index and enterprise cloud business expansion factor comprehensive evaluation index, evaluating influence degrees of the three types of indexes on the cloud on the enterprise, and realizing quantitative calculation of the three types of indexes.

(3) The cloud demand guiding method is characterized in that a demand scene affecting cloud on an enterprise is analyzed by using a scene thinking, the enterprise is divided into a business environment factor affecting scene and a cloud demand guiding scene according to different requirements of the enterprise on cloud resource demands, meanwhile, the cloud demand guiding scene is further subdivided, cloud 'just needed' scene enterprise guest groups and cloud 'improved' scene enterprise guest groups are creatively provided, an evaluation cloud probability index model is respectively constructed for the two scenes, cloud probability of enterprise clients under corresponding scenes is quantified, and a new service popularization mode is opened up for target client screening and scene marketing by a service marketing department.

It should be noted that, for simplicity of description, the method embodiments are shown as a series of acts, but it should be understood by those skilled in the art that the embodiments are not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts are not necessarily required by the embodiments of the invention.

Referring to fig. 7, a block diagram of an embodiment of an apparatus for mining cloud clients on an enterprise according to the present invention may specifically include the following modules:

an indication label construction module 710, configured to construct an on-enterprise cloud indication label of each target enterprise based on the multi-source database;

The cloud loading influence factor quantification module 720 is configured to quantify an enterprise cloud loading indication label of each target enterprise based on the enterprise cloud loading influence factors, and determine quantification values of each cloud loading influence factor of the target enterprise;

the cloud probability determining module 730 is configured to determine a cloud probability of each target enterprise based on the quantized value of the cloud influencing factor;

and an enterprise cloud client mining module 740, configured to determine an enterprise cloud client based on the cloud probability of each target enterprise.

As in the apparatus above, optionally, the multi-source database comprises at least one of the following databases:

As in the above apparatus, optionally, the indication tag construction module 710 is specifically configured to:

As in the above apparatus, optionally, the indication tag construction module 710 is configured to: the method is particularly used for constructing the cloud bidding indication sub-label based on bidding information in the enterprise public information database:

Acquiring enterprise bidding information from enterprise public information;

As mentioned above, optionally, the indication label construction module 710 is specifically configured to, when constructing the sub-label of the cloud-top indication based on the corporate network information in the corporate public information database:

As above, optionally, the cloud-on-enterprise influencing factors include:

Business operating factors and business economy factors.

As mentioned above, optionally, the cloud impact factor quantization module 720 is specifically configured to:

Performing label assignment on each influence factor indication label;

As above, optionally, the cloud-on-enterprise influencing factors further include: business expansion factors for enterprises.

Performing label assignment on each influence factor indication label;

As mentioned above, optionally, the cloud impact factor quantization module 720 is configured to, when performing label assignment on each of the impact factor indication labels, specifically:

As mentioned above, optionally, the cloud probability determining module 730 is specifically configured to:

As mentioned above, optionally, the on-enterprise cloud client mining module 740 is specifically configured to:

The apparatus as above, optionally, further comprising:

And the recommending module is used for recommending the corresponding cloud service for each cloud client on the enterprise according to the cloud indication label and the cloud service information on the enterprise of each cloud client on the enterprise.

For the device embodiment, since the device embodiment is substantially similar to the method embodiment, the description is relatively simple, and the relevant points only need to be referred to the part of the description of the method embodiment, which is not repeated herein.

Referring to fig. 8, there is shown a block diagram of an embodiment of an electronic device of the present invention, the device comprising: a processor 810, a memory 820, and a bus 830;

Wherein processor 810 and memory 820 communicate with each other through bus 830;

the processor 810 is configured to invoke program instructions in the memory 820 to perform the methods provided by the method embodiments described above, including, for example: constructing an enterprise cloud indication label of each target enterprise based on the multi-source database; quantifying an enterprise cloud indication label of each target enterprise based on enterprise cloud influence factors, and determining quantized values of each cloud influence factor of the target enterprise; determining the cloud probability of each target enterprise based on the cloud influence factor quantification value; and determining cloud clients on the enterprises based on the cloud probability of each target enterprise.

Embodiments of the present invention disclose a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the methods provided by the method embodiments described above, for example comprising: constructing an enterprise cloud indication label of each target enterprise based on the multi-source database; quantifying an enterprise cloud indication label of each target enterprise based on enterprise cloud influence factors, and determining quantized values of each cloud influence factor of the target enterprise; determining the cloud probability of each target enterprise based on the cloud influence factor quantification value; and determining cloud clients on the enterprises based on the cloud probability of each target enterprise.

Embodiments of the present invention provide a non-transitory computer readable storage medium storing computer instructions that cause a computer to perform the methods provided by the above-described method embodiments, for example, including: constructing an enterprise cloud indication label of each target enterprise based on the multi-source database; quantifying an enterprise cloud indication label of each target enterprise based on enterprise cloud influence factors, and determining quantized values of each cloud influence factor of the target enterprise; determining the cloud probability of each target enterprise based on the cloud influence factor quantification value; and determining cloud clients on the enterprises based on the cloud probability of each target enterprise.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.

It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or terminal device that comprises the element.

The above detailed description of a method for mining cloud clients on an enterprise, a device for mining cloud clients on an enterprise, an electronic device and a storage medium provided by the invention applies specific examples to illustrate the principles and embodiments of the invention, and the above examples are only used for helping to understand the method and core ideas of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.

Claims

1. A method for mining cloud customers on an enterprise, comprising:

2. The method of claim 1, wherein the multi-source database comprises at least one of the following databases:

3. The method of claim 2, wherein constructing the cloud-on-enterprise indication tags for each target enterprise based on the multi-source database comprises:

4. The method of claim 2, wherein constructing the cloud-on-enterprise indication tags for each target enterprise based on the multi-source database comprises:

5. The method of claim 2, wherein constructing the cloud-on-enterprise indication tags for each target enterprise based on the multi-source database comprises:

6. The method of claim 5, wherein constructing a cloud bid indication sub-label based on bid information in an enterprise public information database comprises:

Acquiring enterprise bidding information from enterprise public information;

7. The method of claim 6, wherein constructing the cloud-top sign sub-tag based on the corporate network information in the corporate public information database comprises:

8. The method of claim 1, wherein the cloud-on-enterprise influencing factors comprise:

Business operating factors and business economy factors.

9. The method of claim 8, wherein quantifying the cloud-on-enterprise indicator tags for each of the target enterprises based on the cloud-on-enterprise influencing factors, determining respective cloud-on-influencing factor quantified values for the target enterprises, comprises:

Performing label assignment on each influence factor indication label;

10. The method of claim 9, wherein quantifying the cloud-on-enterprise indicator tags for each of the target enterprises based on the cloud-on-enterprise influencing factors, determining respective cloud-on-influencing factor quantified values for the target enterprises comprises:

Performing label assignment on each influence factor indication label;

11. The method of claim 10, wherein the cloud-on-enterprise influencing factors further comprise: business expansion factors for enterprises.

12. The method of claim 11, wherein quantifying the cloud-on-enterprise indicator tags for each of the target enterprises based on the cloud-on-enterprise influencing factors, determining respective cloud-on-influencing factor quantified values for the target enterprises comprises:

Performing label assignment on each influence factor indication label;

13. The method of claim 9, 10 or 12, wherein said assigning a label to each of said influencing factor indicator labels comprises:

14. The method of claim 10, wherein the determining the cloud-up probability for each of the target enterprises based on the cloud-up impact factor quantified values comprises:

15. The method of claim 12, wherein the determining the cloud-up probability for each of the target enterprises based on the cloud-up impact factor quantified values comprises:

16. The method of claim 1, wherein the determining an enterprise cloud client based on the cloud probability of each of the target enterprises comprises:

17. The method as recited in claim 1, further comprising:

18. An apparatus for mining cloud clients on an enterprise, comprising:

19. An electronic device, comprising:

The device comprises a memory and a processor, wherein the processor and the memory are communicated with each other through a bus; the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any of claims 1-17.

20. A computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the method of any one of claims 1 to 17.