CN118644322B - A product intelligent recommendation method and system for e-commerce user big data - Google Patents
A product intelligent recommendation method and system for e-commerce user big data Download PDFInfo
- Publication number
- CN118644322B CN118644322B CN202411110020.4A CN202411110020A CN118644322B CN 118644322 B CN118644322 B CN 118644322B CN 202411110020 A CN202411110020 A CN 202411110020A CN 118644322 B CN118644322 B CN 118644322B
- Authority
- CN
- China
- Prior art keywords
- user
- consumption
- product
- products
- similar
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Recommending goods or services
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Electronic shopping [e-shopping] by investigating goods or services
- G06Q30/0625—Electronic shopping [e-shopping] by investigating goods or services by formulating product or service queries, e.g. using keywords or predefined options
- G06Q30/0629—Electronic shopping [e-shopping] by investigating goods or services by formulating product or service queries, e.g. using keywords or predefined options by pre-processing results, e.g. ranking or ordering results
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a product intelligent recommendation method and a system for big data of e-commerce users, which relate to the technical field of e-commerce and data processing and comprise the following steps: acquiring historical consumption data of a user; obtaining at least one similar user set; obtaining at least one consumption time period and a non-consumption time period; obtaining a consumption trend product of a user in a consumption time period; obtaining user tendency characteristics, and recommending products according to the user tendency characteristics; obtaining possible consumption products, and recommending the possible consumption products in a consumption time period; and acquiring the recommended consumption proportion of the user, and judging whether the recommended consumption proportion is larger than a preset value. By arranging the user classification module, the recommendation test module, the product recommendation module and the judgment and identification module, the recommended products are adjusted in real time according to the monitoring result, so that the flexibility of an algorithm can be improved, the test range can be effectively reduced, and useless tests can be reduced.
Description
Technical Field
The invention relates to the technical field of electronic commerce and data processing, in particular to an intelligent product recommendation method and system for electronic commerce user big data.
Background
The electronic commerce refers to a transaction main body belonging to different environments, an electronic commerce platform and an online transaction platform are used for achieving transaction and payment settlement through the electronic commerce platform, and delivering commodities through the commodity stream to complete the transaction. The electronic commerce is used as a new business state, the traditional trade is networked and electronic, the electronic technology and logistics are used as main means, the business is used as a core, and the traditional sales and shopping channels are moved to the internet, so that the intermediate links can be reduced, and the cost is saved.
However, when the existing electronic commerce recommends products, the set algorithm is relatively fixed, the matching with the users is insufficient, when the matching degree with the users is insufficient, the users cannot easily adjust in time, and the recommendation algorithm is not intelligent enough for the users to explore the potential purchased products, and a large number of repeated tests are needed to explore the potential purchased products of the users.
Disclosure of Invention
In order to solve the technical problems, the technical scheme provides an intelligent product recommendation method and system for large data of an e-commerce user, and solves the problems that when the existing e-commerce provided in the background technology is used for recommending products, a set algorithm is relatively fixed and is easy to be matched with the user, when the matching degree with the user is insufficient, the user cannot easily adjust in time, the recommendation algorithm is not intelligent enough for the development of the user for potential purchasing products, and a large number of repeated tests are needed to discover the potential purchasing products of the user.
In order to achieve the above purpose, the invention adopts the following technical scheme:
An intelligent product recommendation method for big data of e-commerce users comprises the following steps:
acquiring historical consumption data of a user, wherein the historical consumption data comprises annual spending amount, shopping category and shopping time of the user;
dividing users to obtain at least one similar user set;
based on shopping time of a user, uniformly dividing daily time to obtain at least one consumption time period and a non-consumption time period;
Based on the historical consumption data, obtaining consumption tendency products of the user in a consumption time period;
When a user browses the similar consumption trend products, at least one similar display product which is not selected is obtained, at least one target product selected by the user is obtained, the target product and the at least one similar display product are analyzed to obtain user trend characteristics, the user trend characteristics are corresponding to a consumption time period, and product recommendation is carried out according to the user trend characteristics in the consumption time period;
Acquiring a non-consumption time period of a user, performing interest test in the non-consumption time period to obtain a possible consumption product, and recommending the possible consumption product in the consumption time period;
Acquiring a recommended consumption proportion of a user, wherein the recommended consumption proportion of the user refers to the proportion of transactions contributed by the user through recommendation in the consumption amount, judging whether the recommended consumption proportion is larger than a preset value, if so, continuing to recommend products according to the user trend characteristics in a consumption time period;
if not, using the similar user optimization recommendation mechanism to recommend the product to the user, wherein the similar user is the user in the same similar user set.
Preferably, the dividing the users to obtain at least one similar user set includes the following steps:
calculating the deviation degree of the annual spending amounts of the two users, wherein the deviation degree of the annual spending amounts is equal to the difference value of the annual spending amounts of the two users divided by the sum of the annual spending amounts of the two users;
Calculating the deviation degree of the shopping categories of the two users, wherein the deviation degree of the shopping categories is equal to the sum of the different numbers of the shopping categories of the two users divided by the number of the shopping categories of the two users;
calculating the deviation degree of the shopping time of the two users, wherein the deviation degree of the shopping time is equal to the difference value of the shopping time of the two users divided by the sum of the shopping time of the two users;
And dividing the users into at least one similar user set according to the calculation result, wherein the deviation degree of annual spending amounts of any two users in the similar user set, the deviation degree of shopping categories and the deviation degree of shopping time are smaller than preset values.
Preferably, the uniformly dividing the daily time to obtain at least one consumption time period and a non-consumption time period includes the following steps:
Each user is uniformly divided into daily time, and the part except the shopping time in the daily time is used as the non-shopping time;
uniformly dividing shopping time to obtain at least one consumption time period;
and uniformly dividing the non-shopping time to obtain at least one non-consumption time period.
Preferably, the method for obtaining the consumption trend product of the user in the consumption time period based on the historical consumption data comprises the following steps:
counting the total times of products purchased by a user in a consumption time period as a first time;
counting the total times of similar products purchased by the user in the consumption time period to be used as the second times;
Comparing the second time with the first time to obtain the appearance ratio of the similar products;
When the appearance proportion of the similar products is larger than the preset proportion, the similar products are used as consumption tendency products.
Preferably, the analyzing the target product and the at least one similar display product to obtain the user tendency characteristics includes the following steps:
extracting features of at least one target product to obtain at least one target feature, wherein during feature extraction, the price, the appearance and the color of the target product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the target product as at least one target feature;
Extracting features of at least one similar display product to obtain at least one non-target feature, wherein during feature extraction, the price, the appearance and the color of the similar display product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the similar display product, and the price range, the possible value of the appearance and the possible value of the color are taken as the at least one non-target feature;
Acquiring the superposition of at least one target feature and at least one non-target feature to obtain at least one superposition feature;
And acquiring a part of the at least one target feature except the at least one coincident feature as a user tendency feature, and pairing the user tendency feature with the product type corresponding to the target product.
Preferably, the product recommendation according to the user trend feature in the consumption time period comprises the following steps:
In the consumption time period, at least one product type of the same type of the product type corresponding to the user tendency characteristics is obtained and used as a characteristic product;
The number of user tendency features in the feature products is obtained and used as the feature number;
accumulating the feature numbers of all feature products to obtain feature total number;
dividing the number of the features by the total number of the features to obtain the feature ratio of the feature product;
And obtaining the total recommended times recommended to the user in the consumption time period, and multiplying the total recommended times by the characteristic ratio of the characteristic product to obtain the display times of the characteristic product, wherein the recommended times of the characteristic product in the consumption time period are equal to the display times.
Preferably, the interest test during the non-consumption period, the obtaining the possible consumption product comprises the following steps:
Summarizing consumption trend products of all users in the similar user set to obtain at least one product to be tested;
Obtaining products which are different from consumption trend products of the single user in at least one product to be tested, and taking the products as pre-test products;
And recommending the product in the non-consumption time period of the user by using the pre-test product, and acquiring the pre-test product of which the repeated click times of the user exceed the preset times as a possible consumption product.
Preferably, the recommending of the possible consumer products during the consumption period comprises the steps of:
Extracting features of the possible consumer products to obtain at least one suspected feature, wherein during feature extraction, the price, the appearance and the color of the possible consumer products are identified to obtain the price range, the value possibility of the appearance and the value possibility of the color of the possible consumer products as at least one suspected feature;
Acquiring at least one product to be recommended, which has more than a preset number of suspected features and is similar to a possible consumer product;
and randomly recommending the product to be recommended in the consumption time period.
Preferably, the recommending the product to the user by using the similar user optimizing recommending mechanism comprises the following steps:
Ordering and numbering the consumption trend products of the users according to the purchase times of the users from small to large, and obtaining the consumption trend products with the numbers smaller than the preset numbers as replaceable consumption trend products;
Acquiring a similar user set where a user is located, and taking the similar user set as a characteristic similar user set;
Counting the purchase times of the consumption trend products of the users in the similar user set, sorting and numbering the consumption trend products from large to small according to the purchase times of the consumption trend products, and obtaining the consumption trend products with the number smaller than a preset number as pre-replacement consumption trend products;
When recommending a replaceable consumer trend product at the time of product recommendation, the replacement is performed using the pre-replacement consumer trend product.
An intelligent product recommendation system for electronic commerce user big data is used for realizing the intelligent product recommendation method for electronic commerce user big data, comprising the following steps:
the data acquisition module acquires historical consumption data of a user;
the user classification module divides the users to obtain at least one similar user set;
the time division module is used for evenly dividing the daily time based on the shopping time of the user to obtain at least one consumption time period and a non-consumption time period;
The data comparison module is used for obtaining consumption trend products of users in a consumption time period based on historical consumption data;
The product recommendation module is used for acquiring at least one similar display product which is not selected by a user when the user browses the similar consumption trend products, acquiring at least one target product selected by the user, analyzing the target product and the at least one similar display product to acquire user trend characteristics, and corresponding the user trend characteristics to a consumption time period, wherein the product recommendation is performed according to the user trend characteristics in the consumption time period;
the recommendation test module is used for acquiring a non-consumption time period of a user, performing interest test on the non-consumption time period to obtain a possible consumption product, and recommending the possible consumption product in the consumption time period;
The judging and identifying module is used for acquiring the recommended consumption proportion of the user, judging whether the recommended consumption proportion is larger than a preset value, if so, continuing to recommend the product according to the tendency characteristics of the user in the consumption time period; if not, using the similar user optimization recommendation mechanism to recommend the product to the user, wherein the similar user is the user in the same similar user set.
Compared with the prior art, the invention has the beneficial effects that:
By setting the user classification module, the recommendation test module, the product recommendation module and the judgment and identification module, similar products of user preference and non-preference are compared, so that user trend characteristics are obtained, recommendation efficiency can be improved according to the user trend characteristics, real-time monitoring is carried out on product recommendation of the user, and the recommended products are adjusted in real time according to monitoring results, so that products of users depending on similar user sets are adjusted, the matching degree of adjustment is high, the flexibility of an algorithm can be improved, meanwhile, when potential consumer products of the user are mined, the products of the user similar to the user are used for testing, the testing range can be effectively reduced, useless testing can be reduced, and the testing is carried out in a non-consumption time period without affecting the sales total amount of electronic commerce.
Drawings
FIG. 1 is a flow chart of the intelligent product recommendation method for electronic commerce user big data according to the present invention;
FIG. 2 is a flow chart of the present invention for partitioning users to obtain at least one set of similar users;
FIG. 3 is a flow chart of the present invention for evenly dividing the time of day to obtain at least one consumption period and a non-consumption period;
FIG. 4 is a flow chart of the present invention for deriving consumer-friendly products for a consumer over a period of consumption based on historical consumption data;
FIG. 5 is a flow chart of the present invention for analyzing a target product and at least one similar display product to obtain user trend characteristics;
FIG. 6 is a flow chart of product recommendation according to user trend characteristics during a consumption period of the present invention;
FIG. 7 is a schematic flow chart of interest testing during non-consumption time period to obtain possible consumption products according to the present invention;
FIG. 8 is a schematic flow chart of the present invention for recommending possible consumer products during a consumption time period;
fig. 9 is a schematic flow chart of product recommendation for users by using the similar user optimization recommendation mechanism.
Detailed Description
The following description is presented to enable a person skilled in the art to practice the invention. The preferred embodiments in the following description are by way of example only and other obvious variations will occur to the skilled person.
Referring to fig. 1, a product intelligent recommendation method for electronic commerce user big data includes:
acquiring historical consumption data of a user, wherein the historical consumption data comprises annual spending amount, shopping category and shopping time of the user;
dividing users to obtain at least one similar user set;
based on shopping time of a user, uniformly dividing daily time to obtain at least one consumption time period and a non-consumption time period;
Based on the historical consumption data, obtaining consumption tendency products of the user in a consumption time period;
When a user browses the similar consumption trend products, at least one similar display product which is not selected is obtained, at least one target product selected by the user is obtained, the target product and the at least one similar display product are analyzed to obtain user trend characteristics, the user trend characteristics are corresponding to a consumption time period, and product recommendation is carried out according to the user trend characteristics in the consumption time period;
Acquiring a non-consumption time period of a user, performing interest test in the non-consumption time period to obtain a possible consumption product, and recommending the possible consumption product in the consumption time period;
Acquiring a recommended consumption proportion of a user, wherein the recommended consumption proportion of the user refers to the proportion of transactions contributed by the user through recommendation in the consumption amount, judging whether the recommended consumption proportion is larger than a preset value, if so, continuing to recommend products according to the user trend characteristics in a consumption time period;
if not, using the similar user optimization recommendation mechanism to recommend the product to the user, wherein the similar user is the user in the same similar user set.
The recommended consumption ratio of the user refers to transactions which the user promotes through recommendation, and when the user shops on the e-commerce platform, part of the transactions are promoted by self-searching.
Referring to fig. 2, the user is divided to obtain at least one similar user set, which includes the following steps:
calculating the deviation degree of the annual spending amounts of the two users, wherein the deviation degree of the annual spending amounts is equal to the difference value of the annual spending amounts of the two users divided by the sum of the annual spending amounts of the two users;
Calculating the deviation degree of the shopping categories of the two users, wherein the deviation degree of the shopping categories is equal to the sum of the different numbers of the shopping categories of the two users divided by the number of the shopping categories of the two users;
calculating the deviation degree of the shopping time of the two users, wherein the deviation degree of the shopping time is equal to the difference value of the shopping time of the two users divided by the sum of the shopping time of the two users;
And dividing the users into at least one similar user set according to the calculation result, wherein the deviation degree of annual spending amounts of any two users in the similar user set, the deviation degree of shopping categories and the deviation degree of shopping time are smaller than preset values.
The purpose of dividing the users is to form a similar user set, and the users in the similar user set are users with similar shopping habits, so that products of the users in the similar user set have reference values for other users in the same set, and the products are used for testing or recommending, so that the matching degree is high, and the efficiency of a recommendation algorithm can be improved.
Referring to fig. 3, the uniform division of the daily time to obtain at least one consumption period and a non-consumption period includes the steps of:
Each user is uniformly divided into daily time, and the part except the shopping time in the daily time is used as the non-shopping time;
uniformly dividing shopping time to obtain at least one consumption time period;
and uniformly dividing the non-shopping time to obtain at least one non-consumption time period.
The purpose of obtaining at least one consumption time period and a non-consumption time period is to distinguish time, so that the time of product testing and the time of recommending products can be determined, and because the product testing is performed in the consumption time period, the consumption opportunity is occupied, the sales amount is reduced, which is disadvantageous to an e-commerce platform, but the product testing is performed in the non-consumption time period, the same effect can be obtained, the consumption opportunity is not occupied, and the sales amount is not reduced.
Referring to fig. 4, deriving consumer-preferred products for a consumer over a period of consumption based on historical consumption data includes the steps of:
counting the total times of products purchased by a user in a consumption time period as a first time;
counting the total times of similar products purchased by the user in the consumption time period to be used as the second times;
Comparing the second time with the first time to obtain the appearance ratio of the similar products;
When the appearance proportion of the similar products is larger than the preset proportion, the similar products are used as consumption tendency products.
The consumption trend product is used for recommending the product for the user, and the consumption trend product is used for recommending the product, so that the matching degree of recommendation can be improved, the purchase possibility of the user can be further increased, and the sales amount can be improved.
Referring to fig. 5, the analysis of the target product and at least one similar display product to obtain the user tendency characteristics includes the following steps:
extracting features of at least one target product to obtain at least one target feature, wherein during feature extraction, the price, the appearance and the color of the target product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the target product as at least one target feature;
Extracting features of at least one similar display product to obtain at least one non-target feature, wherein during feature extraction, the price, the appearance and the color of the similar display product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the similar display product, and the price range, the possible value of the appearance and the possible value of the color are taken as the at least one non-target feature;
Acquiring the superposition of at least one target feature and at least one non-target feature to obtain at least one superposition feature;
And acquiring a part of the at least one target feature except the at least one coincident feature as a user tendency feature, and pairing the user tendency feature with the product type corresponding to the target product.
And when the consumer trend products are analyzed, analyzing the target products and at least one similar display product, wherein the target products and the at least one similar display product are similar products of the consumer trend products, but the target products are products purchased by users, and the similar display products are products not purchased by users, so that the target products contain characteristics favored by the users and characteristics disfavored by the users, the similar display products contain characteristics disfavored by the users, the characteristics contained by the similar display products are used for removing part of the characteristics disfavored by the users contained by the target products, and the obtained user trend characteristics have higher correlation degree with the favorite degree of the users, so that the possibility of promoting the transaction according to the recommendation made by the user trend characteristics is higher.
Referring to fig. 6, during a consumption period, making product recommendations based on user-friendly features includes the steps of:
In the consumption time period, at least one product type of the same type of the product type corresponding to the user tendency characteristics is obtained and used as a characteristic product;
The number of user tendency features in the feature products is obtained and used as the feature number;
accumulating the feature numbers of all feature products to obtain feature total number;
dividing the number of the features by the total number of the features to obtain the feature ratio of the feature product;
And obtaining the total recommended times recommended to the user in the consumption time period, and multiplying the total recommended times by the characteristic ratio of the characteristic product to obtain the display times of the characteristic product, wherein the recommended times of the characteristic product in the consumption time period are equal to the display times.
When the recommendation is carried out, the recommendation duty ratio is required to be set according to the like degree of the user, otherwise, the sales of the product with low like degree of the user is reduced, the like degree of the user is related to the number of the user tendency characteristics in the characteristic product, and therefore the characteristic duty ratio is calculated and is used for recommendation distribution.
Referring to FIG. 7, the interest test performed during non-consumption periods, resulting in a possible consumer product, includes the steps of:
Summarizing consumption trend products of all users in the similar user set to obtain at least one product to be tested;
Obtaining products which are different from consumption trend products of the single user in at least one product to be tested, and taking the products as pre-test products;
And recommending the product in the non-consumption time period of the user by using the pre-test product, and acquiring the pre-test product of which the repeated click times of the user exceed the preset times as a possible consumption product.
The interest test aims at exploring potential consumer products of users, further expanding the product range of sales, and further improving sales.
Referring to fig. 8, recommending possible consumer products for a period of consumption includes the steps of:
Extracting features of the possible consumer products to obtain at least one suspected feature, wherein during feature extraction, the price, the appearance and the color of the possible consumer products are identified to obtain the price range, the value possibility of the appearance and the value possibility of the color of the possible consumer products as at least one suspected feature;
Acquiring at least one product to be recommended, which has more than a preset number of suspected features and is similar to a possible consumer product;
and randomly recommending the product to be recommended in the consumption time period.
To further facilitate the transaction when the possible consumer products are acquired, therefore, recommendations need to be made during the consumer's period of time.
Referring to fig. 9, using the same type of user-optimized recommendation mechanism, product recommendation to a user includes the steps of:
Ordering and numbering the consumption trend products of the users according to the purchase times of the users from small to large, and obtaining the consumption trend products with the numbers smaller than the preset numbers as replaceable consumption trend products;
Acquiring a similar user set where a user is located, and taking the similar user set as a characteristic similar user set;
Counting the purchase times of the consumption trend products of the users in the similar user set, sorting and numbering the consumption trend products from large to small according to the purchase times of the consumption trend products, and obtaining the consumption trend products with the number smaller than a preset number as pre-replacement consumption trend products;
When recommending a replaceable consumer trend product at the time of product recommendation, the replacement is performed using the pre-replacement consumer trend product.
The product with poor sales condition can be replaced by using the pre-replacement consumption trend product, and the product with poor sales condition can be replaced by using the pre-replacement consumption trend product, wherein the pre-replacement consumption trend product is a product which is easy to purchase by a user in a similar user set, so that the matching degree after the replacement is high, and the current recommendation mechanism can be optimized rapidly after few times of replacement, so that the recommended consumption ratio of the user exceeds a preset value.
An intelligent product recommendation system for electronic commerce user big data is used for realizing the intelligent product recommendation method for electronic commerce user big data, comprising the following steps:
the data acquisition module acquires historical consumption data of a user;
the user classification module divides the users to obtain at least one similar user set;
the time division module is used for evenly dividing the daily time based on the shopping time of the user to obtain at least one consumption time period and a non-consumption time period;
The data comparison module is used for obtaining consumption trend products of users in a consumption time period based on historical consumption data;
The product recommendation module is used for acquiring at least one similar display product which is not selected by a user when the user browses the similar consumption trend products, acquiring at least one target product selected by the user, analyzing the target product and the at least one similar display product to acquire user trend characteristics, and corresponding the user trend characteristics to a consumption time period, wherein the product recommendation is performed according to the user trend characteristics in the consumption time period;
the recommendation test module is used for acquiring a non-consumption time period of a user, performing interest test on the non-consumption time period to obtain a possible consumption product, and recommending the possible consumption product in the consumption time period;
The judging and identifying module is used for acquiring the recommended consumption proportion of the user, judging whether the recommended consumption proportion is larger than a preset value, if so, continuing to recommend the product according to the tendency characteristics of the user in the consumption time period; if not, using the similar user optimization recommendation mechanism to recommend the product to the user, wherein the similar user is the user in the same similar user set.
Still further, the present solution also proposes a storage medium having a computer readable program stored thereon, which when called performs the above-mentioned product intelligent recommendation method for e-commerce user big data.
It is understood that the storage medium may be a magnetic medium, e.g., floppy disk, hard disk, magnetic tape; optical media such as DVD; or a semiconductor medium such as a solid state disk SolidStateDisk, SSD, etc.
In summary, the invention has the advantages that: by setting the user classification module, the recommendation test module, the product recommendation module and the judgment and identification module, similar products of user preference and non-preference are compared, so that user trend characteristics are obtained, recommendation efficiency can be improved according to the user trend characteristics, real-time monitoring is carried out on product recommendation of the user, and the recommended products are adjusted in real time according to monitoring results, so that products of users depending on similar user sets are adjusted, the matching degree of adjustment is high, the flexibility of an algorithm can be improved, meanwhile, when potential consumer products of the user are mined, the products of the user similar to the user are used for testing, the testing range can be effectively reduced, useless testing can be reduced, and the testing is carried out in a non-consumption time period without affecting the sales total amount of electronic commerce.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, but rather, the principles of the invention have been described in the foregoing examples and description, and that various changes and modifications may be effected therein without departing from the spirit and scope of the invention as defined by the claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (6)
1. An intelligent product recommendation method for big data of an e-commerce user is characterized by comprising the following steps:
acquiring historical consumption data of a user, wherein the historical consumption data comprises annual spending amount, shopping category and shopping time of the user;
dividing users to obtain at least one similar user set;
based on shopping time of a user, uniformly dividing daily time to obtain at least one consumption time period and a non-consumption time period;
Based on the historical consumption data, obtaining consumption tendency products of the user in a consumption time period;
When a user browses the similar consumption trend products, at least one similar display product which is not selected is obtained, at least one target product selected by the user is obtained, the target product and the at least one similar display product are analyzed to obtain user trend characteristics, the user trend characteristics are corresponding to a consumption time period, and product recommendation is carried out according to the user trend characteristics in the consumption time period;
Acquiring a non-consumption time period of a user, performing interest test in the non-consumption time period to obtain a possible consumption product, and recommending the possible consumption product in the consumption time period;
Acquiring a recommended consumption proportion of a user, wherein the recommended consumption proportion of the user refers to the proportion of transactions contributed by the user through recommendation in the consumption amount, judging whether the recommended consumption proportion is larger than a preset value, if so, continuing to recommend products according to the user trend characteristics in a consumption time period;
If not, recommending the products to the users by using a similar user optimization recommendation mechanism, wherein the similar users are users in the same similar user set;
The method for obtaining the consumption trend product of the user in the consumption time period based on the historical consumption data comprises the following steps:
counting the total times of products purchased by a user in a consumption time period as a first time;
counting the total times of similar products purchased by the user in the consumption time period to be used as the second times;
Comparing the second time with the first time to obtain the appearance ratio of the similar products;
when the appearance proportion of the similar products is larger than the preset proportion, the similar products are used as consumption tendency products;
The analysis of the target product and at least one similar display product to obtain the user tendency characteristics comprises the following steps:
extracting features of at least one target product to obtain at least one target feature, wherein during feature extraction, the price, the appearance and the color of the target product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the target product as at least one target feature;
Extracting features of at least one similar display product to obtain at least one non-target feature, wherein during feature extraction, the price, the appearance and the color of the similar display product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the similar display product, and the price range, the possible value of the appearance and the possible value of the color are taken as the at least one non-target feature;
Acquiring the superposition of at least one target feature and at least one non-target feature to obtain at least one superposition feature;
acquiring a part of at least one target feature except at least one coincident feature as a user tendency feature, and pairing the user tendency feature with a product type corresponding to the target product;
The interest test is carried out in a non-consumption time period, and the possible consumption products are obtained by the following steps:
Summarizing consumption trend products of all users in the similar user set to obtain at least one product to be tested;
Obtaining products which are different from consumption trend products of the single user in at least one product to be tested, and taking the products as pre-test products;
Recommending the product in a non-consumption time period of the user by using the pre-test product, and acquiring the pre-test product with the repeated clicking times exceeding the preset times of the user as a possible consumption product;
The method for recommending the product to the user by using the similar user optimization recommendation mechanism comprises the following steps:
Ordering and numbering the consumption trend products of the users according to the purchase times of the users from small to large, and obtaining the consumption trend products with the numbers smaller than the preset numbers as replaceable consumption trend products;
Acquiring a similar user set where a user is located, and taking the similar user set as a characteristic similar user set;
Counting the purchase times of the consumption trend products of the users in the similar user set, sorting and numbering the consumption trend products from large to small according to the purchase times of the consumption trend products, and obtaining the consumption trend products with the number smaller than a preset number as pre-replacement consumption trend products;
When recommending a replaceable consumer trend product at the time of product recommendation, the replacement is performed using the pre-replacement consumer trend product.
2. The intelligent product recommendation method for electronic commerce user big data according to claim 1, wherein the dividing the users to obtain at least one similar user set comprises the following steps:
calculating the deviation degree of the annual spending amounts of the two users, wherein the deviation degree of the annual spending amounts is equal to the difference value of the annual spending amounts of the two users divided by the sum of the annual spending amounts of the two users;
Calculating the deviation degree of the shopping categories of the two users, wherein the deviation degree of the shopping categories is equal to the sum of the different numbers of the shopping categories of the two users divided by the number of the shopping categories of the two users;
calculating the deviation degree of the shopping time of the two users, wherein the deviation degree of the shopping time is equal to the difference value of the shopping time of the two users divided by the sum of the shopping time of the two users;
And dividing the users into at least one similar user set according to the calculation result, wherein the deviation degree of annual spending amounts of any two users in the similar user set, the deviation degree of shopping categories and the deviation degree of shopping time are smaller than preset values.
3. The intelligent product recommendation method for electronic commerce user big data according to claim 2, wherein the uniformly dividing the daily time to obtain at least one consumption time period and a non-consumption time period comprises the following steps:
Each user is uniformly divided into daily time, and the part except the shopping time in the daily time is used as the non-shopping time;
uniformly dividing shopping time to obtain at least one consumption time period;
and uniformly dividing the non-shopping time to obtain at least one non-consumption time period.
4. The intelligent product recommendation method for electronic commerce user big data according to claim 3, wherein the product recommendation according to the user tendency characteristics during the consumption period comprises the steps of:
In the consumption time period, at least one product type of the same type of the product type corresponding to the user tendency characteristics is obtained and used as a characteristic product;
The number of user tendency features in the feature products is obtained and used as the feature number;
accumulating the feature numbers of all feature products to obtain feature total number;
dividing the number of the features by the total number of the features to obtain the feature ratio of the feature product;
And obtaining the total recommended times recommended to the user in the consumption time period, and multiplying the total recommended times by the characteristic ratio of the characteristic product to obtain the display times of the characteristic product, wherein the recommended times of the characteristic product in the consumption time period are equal to the display times.
5. The intelligent product recommendation method for electronic commerce user big data according to claim 4, wherein the recommending possible consumer products in the consumption time period comprises the steps of:
Extracting features of the possible consumer products to obtain at least one suspected feature, wherein during feature extraction, the price, the appearance and the color of the possible consumer products are identified to obtain the price range, the value possibility of the appearance and the value possibility of the color of the possible consumer products as at least one suspected feature;
Acquiring at least one product to be recommended, which has more than a preset number of suspected features and is similar to a possible consumer product;
and randomly recommending the product to be recommended in the consumption time period.
6. An intelligent product recommendation system for electronic commerce user big data, for implementing the intelligent product recommendation method for electronic commerce user big data according to any one of claims 1 to 5, comprising:
the data acquisition module acquires historical consumption data of a user;
the user classification module divides the users to obtain at least one similar user set;
the time division module is used for evenly dividing the daily time based on the shopping time of the user to obtain at least one consumption time period and a non-consumption time period;
The data comparison module is used for obtaining consumption trend products of users in a consumption time period based on historical consumption data;
The product recommendation module is used for acquiring at least one similar display product which is not selected by a user when the user browses the similar consumption trend products, acquiring at least one target product selected by the user, analyzing the target product and the at least one similar display product to acquire user trend characteristics, and corresponding the user trend characteristics to a consumption time period, wherein the product recommendation is performed according to the user trend characteristics in the consumption time period;
the recommendation test module is used for acquiring a non-consumption time period of a user, performing interest test on the non-consumption time period to obtain a possible consumption product, and recommending the possible consumption product in the consumption time period;
the judging and identifying module is used for acquiring the recommended consumption proportion of the user, judging whether the recommended consumption proportion is larger than a preset value, if so, continuing to recommend the product according to the tendency characteristics of the user in the consumption time period; if not, recommending the products to the users by using a similar user optimization recommendation mechanism, wherein the similar users are users in the same similar user set;
The method for obtaining the consumption trend product of the user in the consumption time period based on the historical consumption data comprises the following steps:
counting the total times of products purchased by a user in a consumption time period as a first time;
counting the total times of similar products purchased by the user in the consumption time period to be used as the second times;
Comparing the second time with the first time to obtain the appearance ratio of the similar products;
when the appearance proportion of the similar products is larger than the preset proportion, the similar products are used as consumption tendency products;
The analysis of the target product and at least one similar display product to obtain the user tendency characteristics comprises the following steps:
extracting features of at least one target product to obtain at least one target feature, wherein during feature extraction, the price, the appearance and the color of the target product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the target product as at least one target feature;
Extracting features of at least one similar display product to obtain at least one non-target feature, wherein during feature extraction, the price, the appearance and the color of the similar display product are identified to obtain the price range, the possible value of the appearance and the possible value of the color of the similar display product, and the price range, the possible value of the appearance and the possible value of the color are taken as the at least one non-target feature;
Acquiring the superposition of at least one target feature and at least one non-target feature to obtain at least one superposition feature;
acquiring a part of at least one target feature except at least one coincident feature as a user tendency feature, and pairing the user tendency feature with a product type corresponding to the target product;
The interest test is carried out in a non-consumption time period, and the possible consumption products are obtained by the following steps:
Summarizing consumption trend products of all users in the similar user set to obtain at least one product to be tested;
Obtaining products which are different from consumption trend products of the single user in at least one product to be tested, and taking the products as pre-test products;
Recommending the product in a non-consumption time period of the user by using the pre-test product, and acquiring the pre-test product with the repeated clicking times exceeding the preset times of the user as a possible consumption product;
The method for recommending the product to the user by using the similar user optimization recommendation mechanism comprises the following steps:
Ordering and numbering the consumption trend products of the users according to the purchase times of the users from small to large, and obtaining the consumption trend products with the numbers smaller than the preset numbers as replaceable consumption trend products;
Acquiring a similar user set where a user is located, and taking the similar user set as a characteristic similar user set;
Counting the purchase times of the consumption trend products of the users in the similar user set, sorting and numbering the consumption trend products from large to small according to the purchase times of the consumption trend products, and obtaining the consumption trend products with the number smaller than a preset number as pre-replacement consumption trend products;
When recommending a replaceable consumer trend product at the time of product recommendation, the replacement is performed using the pre-replacement consumer trend product.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411110020.4A CN118644322B (en) | 2024-08-14 | 2024-08-14 | A product intelligent recommendation method and system for e-commerce user big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411110020.4A CN118644322B (en) | 2024-08-14 | 2024-08-14 | A product intelligent recommendation method and system for e-commerce user big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118644322A CN118644322A (en) | 2024-09-13 |
| CN118644322B true CN118644322B (en) | 2024-11-12 |
Family
ID=92661254
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411110020.4A Active CN118644322B (en) | 2024-08-14 | 2024-08-14 | A product intelligent recommendation method and system for e-commerce user big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118644322B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113239268A (en) * | 2021-04-29 | 2021-08-10 | 佛山科学技术学院 | Commodity recommendation method, device and system |
| CN113362144A (en) * | 2021-07-19 | 2021-09-07 | 海南炳祥投资咨询有限公司 | E-commerce shopping recommendation method and system based on big data |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5801257B2 (en) * | 2012-06-13 | 2015-10-28 | 日本電信電話株式会社 | Product diversification recommendation device, method and program |
| CN103325052A (en) * | 2013-07-03 | 2013-09-25 | 姚明东 | Commodity recommendation method based on multidimensional user consumption propensity modeling |
| CN112258260B (en) * | 2020-08-14 | 2024-12-31 | 北京沃东天骏信息技术有限公司 | Page display method, device, medium and electronic device based on user characteristics |
| CN116402581A (en) * | 2023-04-09 | 2023-07-07 | 湖北智慧商圈科技有限公司 | Electronic commerce data management method and system based on artificial intelligence |
| CN116664239A (en) * | 2023-06-02 | 2023-08-29 | 平安银行股份有限公司 | Product recommendation method, device, equipment and medium based on artificial intelligence |
| CN117495492A (en) * | 2023-11-03 | 2024-02-02 | 杭州观远数据有限公司 | Method for carrying out information reinforcement learning recommendation based on consumption upgrading recommendation model |
| CN118096292A (en) * | 2024-01-12 | 2024-05-28 | 广州市禾赢文化传播有限公司 | Intelligent recommendation method and system based on cloud shopping |
| CN117575662B (en) * | 2024-01-17 | 2024-06-07 | 深圳市微购科技有限公司 | Commercial intelligent business decision support system and method based on video analysis |
| CN118350896A (en) * | 2024-04-26 | 2024-07-16 | 合肥经斯新能源科技有限公司 | Clothing online sales buying tendency recommendation system based on weather prediction |
-
2024
- 2024-08-14 CN CN202411110020.4A patent/CN118644322B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113239268A (en) * | 2021-04-29 | 2021-08-10 | 佛山科学技术学院 | Commodity recommendation method, device and system |
| CN113362144A (en) * | 2021-07-19 | 2021-09-07 | 海南炳祥投资咨询有限公司 | E-commerce shopping recommendation method and system based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118644322A (en) | 2024-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118096292A (en) | Intelligent recommendation method and system based on cloud shopping | |
| CN107220852A (en) | Method, device and server for determining target recommended user | |
| US9087335B2 (en) | Multidimensional personal behavioral tomography | |
| CN107301592A (en) | The method and device excavated for commodity substitute | |
| CN107886241B (en) | Resource analysis method, device, medium, and electronic apparatus | |
| CN115496566B (en) | Regional specialty recommendation method and system based on big data | |
| CN113689259A (en) | A method and system for personalized product recommendation based on user behavior | |
| CN115099878B (en) | Marketing method based on big data analysis | |
| US20120330807A1 (en) | Systems and methods for consumer price index determination using panel-based and point-of-sale market research data | |
| CN107516246A (en) | Determination method, determining device, medium and the electronic equipment of user type | |
| Ming et al. | Consumer equilibrium, demand effects, and efficiency in group buying | |
| CN116739652A (en) | Clothing e-commerce sales prediction modeling method | |
| CN117635266A (en) | A platform optimization management system for product recommendation | |
| CN118644322B (en) | A product intelligent recommendation method and system for e-commerce user big data | |
| CN118822698B (en) | E-commerce data intelligent classification recommendation method and system based on artificial intelligence | |
| CN120387868B (en) | Intelligent information recommendation system for E-commerce sales | |
| CN119784288A (en) | Sales performance attribution analysis system and method based on store retail terminals | |
| CN119067729A (en) | Potential customer recommendation method, device and electronic equipment | |
| CN116611796B (en) | Exception detection method and device for store transaction data | |
| CN118799029A (en) | Intelligent recommendation method based on image recognition | |
| CN117273845A (en) | Marketing service e-commerce promotion service system based on big data technology | |
| CN116452299A (en) | Intelligent recommendation system and method for electronic commerce | |
| CN111507780B (en) | Method and device for improving guest unit price | |
| CN110490682A (en) | Method and device for analyzing commodity attributes | |
| Hackl et al. | How to measure product differentiation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20240913 Assignee: Dakanggu (Hangzhou) Network Technology Co.,Ltd. Assignor: NANCHANG INSTITUTE OF TECHNOLOGY Contract record no.: X2025980005217 Denomination of invention: A Product Intelligent Recommendation Method and System for E-commerce User Big Data Granted publication date: 20241112 License type: Common License Record date: 20250311 |