CN116342191A - Intelligent bidirectional promotion method for service B2B E-commerce platform user - Google Patents

Abstract

The invention discloses an intelligent bidirectional popularization method for service B2B E-commerce platform users, which is characterized in that during the operation process of an e-commerce platform, product access data and behavior data of buyer users are collected and analyzed, and the flow, the customer quantity and the business opportunity quantity of product access in the platform are counted in real time; the best channels are screened by analyzing the flow, the client quantity and the business opportunity quantity, and the popularization of products in specific areas and the selection of specific channels are implemented by combining the popularization requirements; on the premise of not changing a platform structure system, the invention only depends on the existing operation data of the platform, and can analyze and judge the access preference of the buyer users in different areas and the heat channel of each area for directional popularization; meanwhile, product data of contract areas and the same industry can be gathered, dynamic aggregation popularization is implemented in the platform, repeated building cost of the system is reduced, the process of gathering data of materials and effects is avoided, and the popularization effect is achieved by more efficient implementation on the premise that the original electronic commerce platform is not changed.

Description

An intelligent two-way promotion method for serving B2B e-commerce platform users

technical field

The invention belongs to the field of electronic commerce, and in particular relates to an intelligent directional promotion method for serving B2B enterprise customers.

Background technique

With the rapid development of e-commerce, more and more merchants use e-commerce platforms to carry out trade activities. In order to allow target buyers to quickly find their own products among many commodities, many e-commerce platform merchants use the whole network The marketing method of promotion, although it is aimed at a large number of customers, the input cost is very high, and the target buyers cannot be accurately locked, resulting in a low ratio of input to output, so targeted promotion came into being. Targeted promotion is the precise promotion of information and the precise placement of advertising documents for target buyers and users. It can effectively save network resources, reduce business costs, and help customers improve promotion effects. As the market's demand for targeted promotion becomes stronger and stronger, in order to better serve B2B e-commerce platform customers, e-commerce platforms are actively seeking to achieve more accurate, more effective, and smarter targeted promotion technologies. However, the existing targeted promotion technology is to accumulate promotion effect data first, and then implement targeted promotion according to the promotion effect of each region. This method needs to accumulate effect data first and then analyze it. For B2B e-commerce platforms, the input cost is high , The cost of trial and error is high, and the progress of the effect is slow, and the actual accuracy and speed of the promotion effect are difficult to guarantee, which is not conducive to the development of the B2B e-commerce platform. An intelligent promotion decision-making method, device and electronic equipment (application number: 202210095965.8) uses promotion materials to promote products to be promoted to obtain promotion effect indicators; the promotion effect indicators include display volume, click volume, transaction order number and transaction amount; The input-output coefficient is obtained from the index; the optimal promotion material is determined according to the input-output coefficient. The present invention obtains the input-output coefficient according to the promotion effect index, and determines the optimal promotion material based on the input-output coefficient, can monitor the promotion effect of each promotion material in real time, and select the optimal promotion material for delivery, thereby The promotion efficiency is greatly improved; however, the success of the selection of promotion materials in the technology of this patent determines the screening accuracy of the final promotion effect index, and the uncertainty in the selection process of promotion materials leads to increased costs of intelligent promotion and slow results.

At the same time, the promotion of products on the platform by the B2B e-commerce platform is not limited to the promotion of external channels outside the platform, but also includes the promotion of products on the platform. The promotion of products within the platform relies on the data collection during the operation of the platform. However, the promotion of external channels increases the collection of materials and collection effect data, and the addition of a database system to the original platform architecture system will further bring about adjustments and changes in the platform architecture. .

To sum up, a targeted promotion method based on the same database and the same platform structure is needed to realize two-way promotion inside and outside the platform without repeating the system.

Contents of the invention

In order to solve the existing technical problems, the present invention provides an intelligent two-way promotion method for serving B2B e-commerce platform users. During the operation of the e-commerce platform, the product access data and behavior data of buyers are collected and analyzed to obtain buyer users. Industry, region, channel and keyword usage preferences, real-time statistics of product access traffic, customer volume and business opportunity volume in the platform; by analyzing traffic, customer volume and business opportunity volume, screen the areas where buyers and users gather and have the best drainage effect channels, combined with the analysis of the promotion needs of sellers and users, the promotion of products in specific areas and the selection of specific channels are implemented; the invention is based on the product access data, behavior data and product data continuously collected during the operation of the platform, without changing the platform structure system Under the premise, relying only on the existing operating data of the platform, it can not only analyze and judge the access preferences of buyers and users in different regions and the popularity channels of each region, but also implement external orientation for products on the platform in regions and channels with good promotion effects Promotion; at the same time, it can also aggregate product data in the same region and industry, implement dynamic aggregation and promotion in the platform, reduce the cost of repeatedly building systems, avoid the process of collecting material and effect data, and ensure that the original electronic Under the premise of a business platform, more efficient implementation can achieve the promotion effect.

The present invention provides an intelligent two-way promotion method for serving B2B e-commerce platform users, and the specific steps include:

Step 1: Based on the product access data and behavior data of buyers and users accessing the platform collected in real time by the promotion resource database, count the channels with high access popularity in each region; specifically: divide the data in the promotion resource database by region, and target the same region data, according to the channel statistics corresponding visits; select the channel whose visits exceed the threshold as a popular channel, and establish a mapping relationship between the region and the popular channel;

The data in the promotion resource database is uniquely identified by the product access data, and the visit volume, customer volume, and business opportunity volume corresponding to the product access data are counted in real time; the business opportunity volume includes inquiry volume, quotation request volume, and real-time message volume The total number of visits refers to the number of product visits by buyers; the number of customers refers to the number of buyers who visit products; the number of inquiries refers to the number of inquiries by buyers; Quotation request volume refers to the number of product inquiry behaviors of buyer users; the real-time message volume refers to the number of messages sent by buyer users to seller users;

The product access data includes product keywords, industry names, regions, and channels; the product keywords record the product keywords used by buyers when users visit; the industry names record the industry names of products; the channels record buyers The channel provider used by the user for access; the region record is the location of the buyer user; the region includes at least one country or region;

Step 2: Based on the product keywords selected by seller users, generate channels and regions suitable for seller users to promote, specifically: compare product keywords selected by seller users with popular product keywords; use consistent product keywords , obtain the corresponding channels and regions from the promotion resource database, and record them in the list of regions suitable for promotion and the list of channels respectively; the popular product keywords are arranged in descending order according to the number of visits, customers and business opportunities in the promotion resource database Access data; product keywords obtained from the product access data sorted before the preset position;

Step 3: Screen the areas with the largest number of visits and the areas with the greatest potential, specifically: arrange the product visit data in descending order according to the number of visits, customers and business opportunities, obtain the area from the product visit data sorted before the preset position, and record it in Access the largest regional directory; count product access data and corresponding business opportunities month by month, calculate the increase in business opportunities in the adjacent two months, select the product access data whose increase is greater than 5%, and access data from the product Get the area in the list and record it in the list of areas with the greatest potential;

Step 4: Match popular channels for the area selected by the seller user: Obtain the promotion area and promotion channel selected by the seller user; the promotion area is the list of areas suitable for promotion, the list of the largest area, and the list of areas with the greatest potential The promotion channel is selected from the list of channels suitable for promotion by sellers and users;

According to the matching relationship between the region and popular channels screened out in step 1, assign the promotion channel to the promotion region; call the API of the promotion channel, and send the associated promotion region and advertisement file to the promotion channel.

In the step 1, the promotion resource database is constructed by using the search engine Lucene tool; the step 1 collects the product access data and behavior data of the buyer user access platform in real time, specifically including:

Step 1-1: Obtain the keywords in the buyer's user access URL address as the product keyword; obtain the core words according to the product keywords, and obtain the industry name according to the mapping relationship between the core words and the industry preset by the platform; according to the HTTP request header Get the corresponding channel from the address of the channel provider recorded in the referer field value; get the region according to the IP field value in the HTTP request header;

Step 1-2: Update the number of visits and customers in the search resource database; specifically: in the promotion search resource database, search whether there is a product keyword, industry name, region, and channel provider that are consistent with step 1-1 Product access data; if there is a match, obtain the visit volume corresponding to the product visit data, and execute the visit volume +1; compare the Pid field value in the HTTP request header with the existing Pid value on the platform, if they are consistent, do not update the customer if not consistent, store the value of the Pid field, and update the customer quantity corresponding to the product access data +1; if not consistent, use the product keyword, the industry name, the region, and the channel as the newly added For product access data, the initial value of visits corresponding to the product access data is recorded as +1, the number of customers is recorded as 1; the amount of quotation requests is recorded as 0; the amount of real-time messages is recorded as 0;

Step 1-3: Update the number of business opportunities in the search resource database, specifically including: obtain the click behavior of the buyer user's inquiry button on the platform, if obtained, add 1 to the amount of executed inquiry; obtain the buyer user's inquiry button on the platform If the click behavior of the price button is obtained, the quotation request amount will be increased by 1; the information sending behavior of the buyer user on the platform will be obtained, and the number of information sent by the buyer user will be added to the real-time message volume; The quantity of quotation requests and the quantity of real-time messages mentioned above are the quantity of business opportunities.

In said step 1-1, when collecting product access data, the data of the URL address and the HTTP request header are cleaned, specifically including clearing crawler data, flow data of overfrequency access and data of malicious script attacks; said clearing crawler data is based on For the UserAgent field value and IP address of the HTTP request header, delete the product access data whose UserAgent value is a crawler and whose IP address is an abnormal address; the clearing of the traffic data of the overclocking access is by counting the access volume of the same Pid within 1 second, within 1 second For the visit volume of the same IP, delete the product visit data whose value exceeds the threshold. The process of clearing the script malicious attack is to delete the product visit data whose URL address contains the sign greater than and the sign less than.

In the step 2, the product access data is arranged in descending order according to the amount of business opportunities, the amount of customers and the amount of visits in the promotion resource database, specifically: first arrange the product access data in descending order according to the amount of visits; if there are products with the same amount of visits For the access data, sort the access data of products with the same number of visits in descending order of the number of customers; if there is access data of products with the same number of customers, sort the access data of products with the same number of customers in descending order of the number of business opportunities; access data from products sorted before the preset position Obtain product keywords from the database and record them as popular product keywords;

In said step 2, if the product keywords selected by the seller user cannot be consistent with popular product keywords; obtain the product access data corresponding to the popular product keywords; and regions, which are respectively recorded in the list of areas suitable for promotion and the list of channels.

In said step 4, according to the type of the channel, a corresponding type of advertisement file is generated, specifically: if the type of the channel is social media, an advertisement file of the Feed type is generated, and the product data of the seller user is obtained from the product database of the platform and written into Feed advertisement file; if the channel type is a search engine, generate a CSV type advertisement file, obtain the PC-end address and mobile-end address of the product display from the product database of the platform and write it into the CSV advertisement file.

Step 5 is included after step 4: dynamically aggregate and promote products in the same region within the platform;

Step 5-1: According to the promotion area selected by the seller user, aggregate the seller users in the same promotion area; obtain the product data of the seller user, and aggregate the product data of the same industry according to the industry name;

Step 5-2: From the product data aggregated in step 5-1, select product data with the same product keywords, count the number of seller user IDs from the product data as the number of seller users, and count the number of product names as product number;

Step 5-2: When the number of users and products of the seller reaches the preset threshold, the product information is obtained from the aggregated product data and displayed centrally and visually on the platform.

In the step 5, the preset threshold for the number of seller users is 2, and the threshold for the number of products is 10, specifically: if the number of seller users is greater than or equal to 2 and the number of products is greater than or equal to 10, the product data Extract product information for centralized visualization; if any of the situations where the number of products is less than 10 or the number of customers is less than 2, emails will be sent to remind managers to manually supplement product information.

The beneficial effect that the present invention reaches:

1) Under the original database system, the present invention only relies on the existing operating data of the platform, not only can analyze and judge the access preferences of buyers and users in different regions and the popularity channels of each region, but also promote the products in the platform with good effect Implement external directional promotion in regions and channels; at the same time, it can also aggregate product data in the same region and industry, and implement dynamic aggregation and promotion in the platform, reducing the cost of repeated system building and ensuring that the original e-commerce platform does not change. Under the premise, more efficient implementation achieves the promotion effect.

2) The present invention collects effective buyer user visit data and counts the visit volume, customer volume, inquiry volume, inquiry volume and real-time message volume of the product, and analyzes the keyword preference of the buyer user and the concentration of the buyer user based on this. The promotion area and the access channels frequently used by buyers and users in a specific area, this method completely eliminates the process of first promoting and then accumulating and analyzing promotion data, and uses buyer behavior as the basis for promotion analysis, which is closer to the actual product needs of buyers. Improve the accuracy of product promotion advertisements, achieve rapid promotion effects at low cost, and achieve more sales revenue for sellers and users through the e-commerce platform;

3) The present invention records buyer user access data in real time by constructing promotional resource data, and matches popular product keywords obtained based on promotional resource database analysis with seller user’s own product keywords to generate a promotion plan for individual needs of seller users , to ensure that the product promotion plan can not only meet the seller's product positioning and promotion needs, but also take into account the product access habits and preferences of buyers and users. Since there is no need to collect additional promotion materials, the promotion cost is reduced and the promotion is greatly improved. efficiency.

Description of drawings

1 is a schematic flow diagram of an intelligent two-way promotion method for serving B2B e-commerce platform users in an embodiment of the present invention;

Fig. 2 is a flow chart of generating channels and regions suitable for seller user promotion in an embodiment of the present invention;

Fig. 3 is the flowchart of the dynamic aggregate display of seller user's product in the e-commerce platform in the embodiment of the present invention;

Fig. 4 is a schematic diagram of a visualization of aggregation and promotion of product areas in the same industry and the same promotion area in an embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

In the following embodiments of the present invention:

The region can include at least one country or region, according to geographical definition or artificially set the countries included in the region; such as geographically defined North America region, Asia region; North America includes the United States, Canada, Mexico and other countries; such as artificially set regions, Including Singapore, Malaysia, Indonesia, Myanmar, Thailand, India, Pakistan, Saudi Arabia, Russia, Ukraine and other countries;

The channels are divided into access channels and promotion channels according to the direction of information flow; among them, the channel that receives buyer user information and transmits it to the e-commerce platform is the access channel, and the channel that receives the advertisement files of the e-commerce platform is the promotion channel; well-known channels include Limited to Google, FaceBook, Bing, etc.;

The product data is product information obtained from the product database of the e-commerce platform, including seller user identification, product name, industry name, product keywords, and the like.

Fig. 1 is a schematic flow diagram of an intelligent two-way promotion method for serving B2B e-commerce platform users in an embodiment of the present invention, and the specific flow includes:

Step 101: Based on the product access data and behavior data of buyers and users accessing the platform collected in real time by the promotion resource database, count the channels with high access popularity in each region; specifically: divide the data in the promotion resource database by region, and target the same region According to the data of the channel, the corresponding visits are counted according to the channels; the channels with the visits exceeding 1 million are selected as popular channels, and a mapping relationship is established between the regions and the popular channels.

The data in the promotion resource database is uniquely identified by the product access data, and the visit volume, customer volume, and business opportunity volume corresponding to the product access data are counted in real time; the business opportunity volume includes inquiry volume, quotation request volume, and real-time message volume The total number of visits refers to the number of product visits by buyers; the number of customers refers to the number of buyers who visit products; the number of inquiries refers to the number of inquiries by buyers; Quotation request volume refers to the number of product inquiry behaviors of buyer users; the real-time message volume refers to the number of messages sent by buyer users to seller users;

The number of visits is an important factor to measure user stickiness; the number of customers represents the share of business opportunities in a specific area, that is, how big the business opportunities are; the amount of business opportunities represents the actual promotion value generated. If there is no traffic, there must be no customers and business opportunities; if there is a lack of business opportunities, even if the traffic is high and the number of customers is high, there is no way to measure the actual promotion value; if there is a lack of customers, there is no way to measure the potential market How big is the share and how big are the opportunities for commercial promotion. Therefore, visits, customers and business opportunities are indispensable.

Secondly, the number of inquiries, quotation requests and real-time news is chosen as the volume of business opportunities because the quantity of inquiries, quotation requests and real-time news represents the promotion value of different dimensions. Inquiry refers to the expression that the buyer asks the other party about the transaction conditions in order to purchase a certain product. In the actual business of international trade, the buyer user actively sends an inquiry to the seller user to consider the buyer's willingness and willingness to trade. sincerity.

Quotation request volume is used to consider the buyer's initiative to purchase products. On the e-commerce platform, buyer users actively publish procurement requirements, and seller users choose to select suitable buyer quotations. In layman's terms, the buyer publishes a purchase demand on the platform, and the seller user adjusts the quotation according to the demand. More than one buyer sent a request for quote. The advantage of quotation requests can improve buyers' purchasing efficiency, help sellers to better complete order conversion, and win more high-quality buyers.

The volume of real-time messages refers to the volume of real-time messages sent by buyers through instant messaging tools on the platform. By counting the volume of real-time messages, buyers are considered to be interested in the products of sellers. During the operation of the B2B e-commerce platform, buyers and users only send messages through chat tools without any inquiry or quotation request, and successfully reach an intended order.

To sum up, when counting the number of business opportunities, if any one is missing, the commercial value of product promotion cannot be fully measured.

The product access data includes product keywords, industry names, regions, and channels; the product keywords record the product keywords used by buyers when users visit; the industry names record the industry names of products; the channels record buyers The channel provider used by the user for access; the region record is the location of the buyer user; the region includes at least one country or region;

In the embodiment of the present invention, the promotion resource database is constructed using the search engine Lucene tool. Compared with other tools, the search engine Lucene tool has strong stability and high indexing performance; the accurate and high-performance search algorithm possessed by the tool will Improve the stable, efficient and accurate search effect of the present invention; in the embodiment of the present invention, the statistical granularity is set in the promotion resource database, and the statistics of visits, customers and business opportunities are counted according to the month; as shown in the following table, record product keywords The number of visits, customers and business opportunities of Led light in different regions and channels.

When building the promotion resource database, the existing data in the platform operation process is selected, which saves the time and implementation process of re-planning, creating the material library, collecting material data, and promotion effect data, and maximizes the promotion effect based on the existing data; In the step 1, use the promotion resource database to collect in real time the product access data and behavior data of buyers and users accessing the platform; specifically include:

Step 101-1: Obtain the keywords in the buyer's user access URL address as product keywords; obtain core words based on product keywords; the core words are words that usually define product types on e-commerce platforms, such as computers, notebooks, and light bulbs , dresses; obtain the industry name according to the core words preset by the platform and the mapping relationship between the industry according to the core words preset by the platform; obtain the corresponding channel according to the address of the channel provider recorded in the referer field value in the HTTP request header; obtain the corresponding channel according to the HTTP request header In the IP field value, get the country, and get the region according to the country;

Step 101-2: Update the number of visits and customers in the search resource database; specifically: in the promotion search resource database, search whether there is a product keyword, industry name, region, and channel provider that are consistent with step 101-1 Product access data; if there is a match, obtain the visit volume corresponding to the product visit data, and execute the visit volume +1; compare the Pid field value in the HTTP request header with the existing Pid value on the platform, if they are consistent, do not update the customer if not consistent, store the value of the Pid field, and update the customer quantity corresponding to the product access data +1; if not consistent, use the product keyword, the industry name, the region, and the channel as the newly added For product access data, the initial value of visits corresponding to the product access data is recorded as +1, the number of customers is recorded as 1; the amount of quotation requests is recorded as 0; the amount of real-time messages is recorded as 0;

Step 101-3: Update the amount of business opportunities in the search resource database, specifically including: obtain the click behavior of the buyer user's inquiry button on the platform, if obtained, add 1 to the execution inquiry amount; obtain the buyer user's inquiry button on the platform If the click behavior of the price button is obtained, the quotation request amount will be increased by 1; the information sending behavior of the buyer user on the platform will be obtained, and the number of information sent by the buyer user will be added to the real-time message volume; The quantity of quotation requests and the quantity of real-time messages mentioned above are the quantity of business opportunities.

Since the e-commerce platform system is frequently attacked by crawlers and malicious attacks, there are many ways to avoid crawler access and malicious attacks. In order to simplify the identification process, the present invention comprehensively considers not to increase the burden of the system structure, and can match the security level of the system at the same time; In the embodiment of the present invention, when collecting product access data, the data of the URL address and the HTTP request header are cleaned by clearing the crawler data, the flow data of overfrequency access and the data of malicious script attacks by combining three methods, specifically including:

Described clear crawler data is according to the UserAgent field value and IP address of HTTP request head, deletes the product visit data that UserAgent value is crawler and IP address is abnormal address; Described conventional crawler includes Google crawler Googlebot and adsbot, Microsoft bingbot, Baidu Baiduspider , Yahoo Yahoo! Slurp, 360spider, Speedy Spider, YandexBot, EasouSpider, LinkpadBot, MJ12bot; the abnormal address is the blacklist IP address preset by the platform;

The traffic data of clearing the overclocking access is to delete the product access data whose value exceeds 50 by counting the access volume of the same Pid within 1 second and the access volume of the same IP within 1 second;

The process of clearing the script malicious attack is to delete the product access data whose URL address contains a greater than or less than sign, such as: https://www.xxx.com/products-search/hot-china-products/<"PVC_Sheet.html .

Step 102: Based on the product keyword selected by the seller user, generate channels and areas suitable for seller user promotion, as shown in Figure 2, the flow chart of generating channels and areas suitable for seller user promotion in the embodiment of the present invention is specifically:

Step 102-1: Obtain popular product keywords, specifically including: first arrange product visit data in descending order of visit volume; There are product access data with the same number of customers, and the product access data with the same number of customers are arranged in descending order according to the number of business opportunities; product keywords are obtained from the top 20 product access data, and recorded as popular product keywords;

Step 102-2: Compare the product keywords selected by the seller user with popular product keywords;

Step 102-3: Using the consistent product keywords, obtain the corresponding channels and regions from the promotion resource database, and record them in the list of regions suitable for promotion and the list of channels respectively;

Step 102-4: If none of the product keywords selected by the seller user are consistent with the popular product keywords; obtain the product access data corresponding to the popular product keywords; compare the channels and regions in the product access data , respectively recorded in the list of areas suitable for promotion and the list of channels

Step 103: Screen the area with the largest scale of visits and the area with the greatest potential, specifically:

Step 103-1: Sort product visit data in descending order of visit volume; if there are product visit data with the same visit volume, sort product visit data with the same visit volume in descending order of customer volume; if customer volume is the same, sort customer volume in descending order of business opportunity volume The same product access data; get the area from the top 4 product access data, and record it in the list of areas with the largest access scale;

Step 103-2: Statistics of product access data and corresponding business opportunities month by month, calculating the growth rate of business opportunities in two adjacent months, selecting the product access data whose increase rate is greater than 5%, and obtaining from the product access data Regions, recorded in the list of regions with the greatest potential;

Step 104: For the area selected by the seller user, match the promotion channel, and push the advertisement file to the channel: obtain the promotion area and promotion channel selected by the seller user; the promotion area is the largest area from the list of areas suitable for promotion by the seller user list and the list of areas with the greatest potential; the promotion channel is selected by sellers and users from the list of channels suitable for promotion; according to the matching relationship between the area screened out in step 1 and the hot channel, the promotion channel is assigned to the promotion area ; Call the API of the promotion channel, and send the associated promotion area and advertisement file to the promotion channel.

In said step 104, according to the type of the channel, a corresponding type of advertisement file is generated, specifically: if the type of the channel is social media, an advertisement file of the Feed type is generated, and the product data of the seller user is obtained from the product database of the platform and written into Feed advertisement file; if the channel type is a search engine, generate a CSV type advertisement file, obtain the PC end address and mobile end address of the product display from the product database of the platform and write it into the CSV advertisement file;

Step 105: Dynamic aggregation and promotion of products in the same area within the platform; FIG. 3 is a flowchart of the dynamic aggregation and display of products of seller users on the e-commerce platform in the embodiment of the present invention, specifically including:

Step 105-1: Screen product data in the same promotion area and the same industry; aggregate seller users in the same promotion area according to the promotion area selected by the seller user; obtain the product data of the seller user, and aggregate products in the same industry according to the industry name data;

Step 105-2: Count the number of seller users and the number of products in the product data; from the product data aggregated in step 105-1, select product data with the same product keywords, and count the number of seller user identifiers from the product data as The number of seller users, the number of statistical product names is the number of products;

Step 105-2: According to the threshold of the number of seller users and the number of products, aggregate and display product data in the e-commerce platform; if the number of seller users is greater than or equal to 2 and the number of products is greater than or equal to 10, extract products from the product data The information is centralized and visualized, as shown in Figure 4; if the number of products is less than 10 or the number of customers is less than 2, an email will be sent to remind managers to manually supplement product information.

The beneficial effect that the present invention reaches:

1) Under the original database system, the present invention only relies on the existing operating data of the platform, not only can analyze and judge the access preferences of buyers and users in different regions and the popularity channels of each region, but also promote the products in the platform with good effect Implement external directional promotion in regions and channels; at the same time, it can also aggregate product data in the same region and industry, and implement dynamic aggregation and promotion in the platform, reducing the cost of repeated system building and ensuring that the original e-commerce platform does not change. Under the premise, more efficient implementation achieves the promotion effect.

2) The present invention collects effective buyer user visit data and counts the visit volume, customer volume, inquiry volume, inquiry volume and real-time message volume of the product, and analyzes the keyword preference of the buyer user and the concentration of the buyer user based on this. The promotion area and the access channels frequently used by buyers and users in a specific area, this method completely eliminates the process of first promoting and then accumulating and analyzing promotion data, and uses buyer behavior as the basis for promotion analysis, which is closer to the actual product needs of buyers. Improve the accuracy of product promotion advertisements, achieve rapid promotion effects at low cost, and achieve more sales revenue for sellers and users through the e-commerce platform;

3) The present invention records buyer user access data in real time by constructing promotional resource data, and matches popular product keywords obtained based on promotional resource database analysis with seller user’s own product keywords to generate a promotion plan for individual needs of seller users , to ensure that the product promotion plan can not only meet the seller's product positioning and promotion needs, but also take into account the product access habits and preferences of buyers and users. Since there is no need to collect additional promotion materials, the promotion cost is reduced and the promotion is greatly improved. efficiency.

The above embodiments do not limit the present invention in any way, and all other improvements and applications made to the above embodiments in an equivalent transformation form belong to the protection scope of the present invention.

Claims (7)

1. An intelligent bidirectional promotion method for serving a B2B e-commerce platform user is characterized by comprising the following steps:

step 1: based on product access data and behavior data of a buyer user access platform collected in real time by a popularization resource database, channels with high access heat in each area are counted; the method specifically comprises the following steps: dividing data in the popularization resource database by regions, and counting corresponding access quantity according to channels aiming at the data with the same regions; selecting the channel with the access quantity exceeding the threshold as a heat channel, and establishing a mapping relation between the area and the heat channel;

the data in the popularization resource database takes the product access data as a unique identifier, and the access quantity, the client quantity and the business opportunity quantity corresponding to the product access data are counted in real time; the business opportunity quantity is the sum of the inquiry quantity, the quotation request quantity and the real-time message quantity; the access amount refers to the product access amount of the buyer user; the customer quantity refers to the number of buyer users accessing the product; the inquiry quantity refers to the number of inquiry behaviors of a buyer user; the quotation request quantity refers to the number of product price inquiring behaviors of a buyer user; the real-time message quantity refers to the quantity of messages sent by a buyer user to a seller user;

the product access data comprises product keywords, industry names, areas and channels; the product keywords record product keywords used when the buyer user accesses; the industry name records the industry name of the product; the channel records the channel quotient used by the buyer user access; the zone record is the location of the buyer user; the region comprises at least 1 country or region;

step 2: based on the product keywords selected by the seller user, generating channels and areas suitable for popularization of the seller user, specifically: comparing product keywords selected by a seller user with popular product keywords; obtaining corresponding channels and areas from a popularization resource database by utilizing the product keywords which are consistent in comparison, and respectively recording an area directory and a channel directory which are suitable for popularization; the popular product keywords are product access data which are arranged in descending order according to the access quantity, the client quantity and the business opportunity quantity in the popularization resource database; product keywords obtained from product access data ordered before a preset position;

step 3: screening the area with the largest access rule and the area with the largest potential, specifically: arranging product access data in descending order according to the access quantity, the client quantity and the business opportunity quantity, acquiring an area from the product access data ordered before a preset position, and recording an area directory with the maximum access rule; calculating business opportunity fluctuation numbers of adjacent 2 months by counting the product access data and the corresponding business opportunity every month, selecting the product access data with the fluctuation number more than 5%, acquiring an area from the product access data, and recording an area name record with the maximum potential;

step 4: matching the heat channel for the region selected by the seller user: acquiring a promotion area and a promotion channel selected by a seller user; the promotion area is selected by a seller user from an area directory suitable for promotion, an area directory with the largest rule and an area directory with the largest potential; the promotion channel is selected by a seller user from a channel directory suitable for promotion;

distributing the promotion channels to the promotion areas according to the matching relation between the areas screened in the step 1 and the heat channels; and calling the API of the promotion channel, and transmitting the associated promotion area and advertisement file to the promotion channel.

2. The intelligent two-way popularization method for the service B2B E-commerce platform user according to claim 1, wherein in the step 1, a popularization resource database is constructed by utilizing a search engine Lucene tool; the step 1 is to collect product access data and behavior data of a buyer user access platform in real time, and specifically comprises the following steps:

step 1-1: obtaining keywords in URL addresses visited by buyer users as product keywords; obtaining core words according to product keywords, and obtaining industry names according to mapping relations between the core words and industries preset by a platform; acquiring a corresponding channel according to the address of the channel provider recorded by the reference field value in the HTTP request header; acquiring an area according to an IP field value in the HTTP request header;

step 1-2: updating the access amount and the client amount in the search resource database; the method specifically comprises the following steps: searching whether product access data consistent with the product keywords, the industry names, the areas and the channel providers in the step 1-1 exist in a popularization and search resource database; if the product access data are consistent, obtaining the corresponding access quantity of the product access data, and executing the access quantity +1; comparing the Pid field value in the HTTP request header with the Pid value existing in the platform, and if the Pid field value is consistent with the Pid value, not updating the client quantity; if the field values are inconsistent, storing the Pid field values, and updating the client quantity +1 corresponding to the product access data; if the product keywords, the industry names, the areas and the channels are inconsistent, the product access data is newly added, the initial value of the access quantity corresponding to the product access data is recorded as +1, and the client quantity is recorded as 1; the bid request amount is recorded as 0; the real-time message quantity is recorded as 0;

step 1-3: updating the business volume in the search resource database, which specifically comprises the following steps: acquiring clicking behaviors of a buyer user in a query button in a platform, and if the clicking behaviors are acquired, executing the query quantity plus 1; acquiring clicking behaviors of a buyer user in a price inquiring button in a platform, and adding 1 to a price request quantity if the clicking behaviors are acquired; acquiring information sending behaviors of a buyer user in a platform, and accumulating the quantity of information sent by the buyer user into real-time information quantity; and accumulating the inquiry quantity, the quotation request quantity and the real-time message quantity to obtain the business opportunity quantity.

3. The intelligent bidirectional promotion method of service B2B e-commerce platform users according to claim 2, wherein in the step 1-1, when collecting product access data, cleaning URL address and HTTP request header data, specifically including data for cleaning crawler data, traffic data accessed by over-frequency and script malicious attack; the step of eliminating the crawler data is to delete the product access data with the user agent value as the crawler and the IP address as the abnormal address according to the user agent field value and the IP address of the HTTP request head; and the flow data for clearing the over-frequency access is obtained by counting the access quantity of the same Pid in 1 second and the access quantity of the same IP in 1 second, and the flow of the malicious attack of the clearing script is to delete the product access data of which the URL address contains more than the number and less than the number.

4. The intelligent two-way promotion method for service B2B e-commerce platform users according to claim 3, wherein in the step 2, product access data are arranged in descending order according to the business volume, the customer volume and the access volume in the promotion resource database, specifically: firstly, arranging product access data according to the descending order of the access quantity; if the product access data with the same access quantity exist, the product access data with the same access quantity are arranged in a descending order according to the client quantity; if the product access data with the same customer quantity exist, arranging the product access data with the same customer quantity according to a descending order of the business quantity; obtaining product keywords from product access data sequenced before a preset position, and marking the product keywords as popular product keywords;

in the step 2, if none of the product keywords selected by the seller user can be aligned with the popular product keywords; acquiring product access data corresponding to the hot product keywords; and respectively recording the channels and the areas in the product access data into an area directory and a channel directory which are suitable for popularization.

5. The intelligent two-way promotion method for serving B2B e-commerce platform users of claim 4, wherein in step 4, according to the channel type, generating advertisement files of corresponding types, specifically: if the channel type is social media, generating a Feed type advertisement file, acquiring product data of a seller user from a product database of a platform, and writing the product data into the Feed advertisement file; if the channel type is a search engine, generating a CSV type advertisement file, acquiring a PC end address and a mobile end address of product display from a product database of a platform, and writing the PC end address and the mobile end address into the CSV advertisement file.

6. The intelligent two-way popularization method for the service B2B e-commerce platform user according to claim 5, wherein the step 4 comprises the following steps: dynamically polymerizing and popularizing products in the same area in a platform;

step 5-1: aggregating seller users in the same promotion area according to the promotion area selected by the seller users; acquiring product data of the seller user, and aggregating the product data of the same industry according to the industry name;

step 5-2: selecting product data with the same product keywords from the product data aggregated in the step 5-1, counting the number of seller user identifiers as the number of seller users and counting the number of product names as the number of products from the product data;

step 5-2: and when the number of the seller users and the number of the products reach a preset threshold value, acquiring product information from the aggregated product data, and concentrating the product information in a platform for visual display.

7. The intelligent two-way promotion method for serving B2B e-commerce platform users of claim 6, wherein in step 5, the preset threshold of the number of seller users is 2, the threshold of the number of products is 10, specifically: if the number of the seller users is more than or equal to 2 and the number of the products is more than or equal to 10, extracting product information from the product data for centralized visualization; if any of the product number is less than 10 and the customer number is less than 2, the mail reminds the manager to manually supplement the product information.

Images