CN112465598A - User portrait generation method, device and equipment and computer storage medium - Google Patents
User portrait generation method, device and equipment and computer storage medium Download PDFInfo
- Publication number
- CN112465598A CN112465598A CN202011403203.7A CN202011403203A CN112465598A CN 112465598 A CN112465598 A CN 112465598A CN 202011403203 A CN202011403203 A CN 202011403203A CN 112465598 A CN112465598 A CN 112465598A
- Authority
- CN
- China
- Prior art keywords
- user
- commodity
- information
- shopping
- type
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 230000006399 behavior Effects 0.000 claims description 35
- 238000010586 diagram Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000010219 correlation analysis Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 4
- 230000009193 crawling Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000000556 factor analysis Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000012790 confirmation Methods 0.000 description 2
- 239000002537 cosmetic Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 241000282326 Felis catus Species 0.000 description 1
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 235000014347 soups Nutrition 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0282—Rating or review of business operators or products
Abstract
The application discloses a user portrait generation method, a device, equipment and a computer storage medium. The user portrait generation method comprises the following steps: acquiring recommended commodity information pushed to a user account by a shopping application program; extracting commodity source side information from the recommended commodity information; classifying the recommended commodity information according to the commodity source information to obtain a corresponding commodity type; determining the shopping type distribution of each user group according to the commodity type and at least one user group to which the user corresponding to the user account belongs; the user group is determined according to user identity information of the user and online operation behavior data of the user; and generating a user portrait of the user according to the shopping type distribution of each user group. According to the embodiment of the invention, complete and real user shopping behavior data can be acquired, and an accurate user portrait can be generated.
Description
Technical Field
The present application belongs to the field of data analysis, and in particular, to a user portrait generation method, apparatus, device, and computer storage medium.
Background
With the popularization of the internet and mobile terminals, everyone can use the mobile terminal to access the internet at any time and any place. More and more people utilize the mobile terminal to finish shopping activities on the E-commerce platform, meanwhile, the mobile terminal can leave traces of shopping behaviors of users, for example, shopping application program push information of shopping information confirmation, commodity evaluation feedback, commodity promotion reminding and the like can be sent to the users by merchants according to the shopping behaviors of the users. The push information can reflect the shopping behavior of the user to a certain extent. Through the analysis of the shopping behaviors of the user, the shopping tendency of the user can be known, and corresponding commodity information is pushed to the user according to the shopping tendency of the user.
At present, in the prior art, a user portrait is generated by mainly extracting user shopping behavior data from a user shopping browsing record and a user shopping invoice to analyze the user shopping behavior. However, the shopping browsing record of the user only represents the shopping interest of the user, and it is impossible to detect whether the user completes the shopping behavior, and although the shopping invoice of the user may indicate that the user completes the shopping behavior, not all users will find the merchant to request the shopping invoice, so that the data of the user shopping behavior extracted from the shopping browsing record of the user and the shopping invoice of the user is incomplete and unreal, resulting in inaccurate final generated user image.
Disclosure of Invention
The embodiment of the application provides a user portrait generation method, device and equipment and a computer storage medium, which can acquire complete and real user shopping behavior data and generate accurate user portrait.
In one aspect, an embodiment of the present application provides a user portrait generation method, where the method includes:
acquiring recommended commodity information pushed to a user account by a shopping application program; extracting commodity source side information from the recommended commodity information; classifying the recommended commodity information according to the commodity source information to obtain a corresponding commodity type; determining the shopping type distribution of each user group according to the commodity type and at least one user group to which the user corresponding to the user account belongs; the user group is determined according to user identity information of the user and online operation behavior data of the user; and generating a user portrait of the user according to the shopping type distribution of each user group.
In the above technical solution, extracting the information of the source of the commodity from the recommended commodity information specifically includes:
and traversing the recommended commodity information based on the regular expression, and extracting commodity source side information corresponding to the recommended commodity information.
In any of the above technical solutions, the classifying the recommended commodity information according to the commodity source information to obtain a corresponding commodity type specifically includes:
matching the information of the commodity source party with a pre-acquired merchant information table, and determining a store name corresponding to the information of the commodity source party and a commodity type corresponding to the store name, wherein the merchant information table is obtained according to crawling shopping application program data; determining the commodity type corresponding to the shop name as the commodity type corresponding to the commodity source party information; and determining the commodity type corresponding to the commodity source party information as the commodity type corresponding to the recommended product information.
In any of the above technical solutions, before matching the information of the commodity source with the pre-obtained merchant information table, the method further includes:
and extracting merchant information meeting a first preset condition from the merchant information table, wherein the first preset condition comprises that the score of the shop score is greater than a preset score and/or the commodity sales volume is greater than a preset sales volume.
In any of the above technical solutions, determining the store name corresponding to the commodity source information and the commodity type corresponding to the store name specifically includes:
the character matching degree of the commodity source party information and the merchant information table meets a second preset condition to obtain a shop name corresponding to the commodity source party information, wherein the second preset condition comprises that the character matching degree of the commodity source party information and the shop name in the merchant information table is larger than a preset percentage; and obtaining the commodity type corresponding to the store name according to the store name corresponding to the commodity source side information.
In any of the above technical solutions, obtaining the type of the product corresponding to the store name according to the store name corresponding to the product source information specifically includes:
extracting a shop score corresponding to the shop name from the shop information table when the shop name corresponds to a plurality of commodity types; and determining the commodity type with the top ranking ratio as the commodity type corresponding to the store name according to the ranking ratio of the store scores in all stores corresponding to each commodity type.
In any of the above technical solutions, before determining the distribution of the shopping types of each user group according to the commodity type and at least one user group to which the user corresponding to the user account belongs, the method further includes:
acquiring user identity information of a user and online operation behavior data of the user; extracting characteristic variables of the user from user identity information of the user and online operation behavior data of the user; performing correlation analysis on the characteristic variables to determine mutually independent continuous characteristic variables; and clustering the users by taking the continuous characteristic variables as indexes based on a maximum expectation algorithm of the GMM Gaussian mixture model, and selecting the optimal classification number of the users by applying a Bayesian information criterion to obtain at least one user group to which the users belong.
In any of the above technical solutions, determining the shopping type distribution of each user group according to the commodity type and at least one user group to which the user corresponding to the user account belongs specifically includes:
according to the purchase proportion, drawing a shopping probability distribution diagram of each user group in different commodity types within a preset time period; and determining the shopping type distribution of each user group according to the shopping probability distribution map.
In any of the above technical solutions, generating a user representation of a user according to a shopping type distribution of each user group specifically includes:
selecting the shopping type with the largest shopping type distribution percentage and/or extracting the shopping type meeting a preset threshold value from the shopping type distribution of each user group, and determining the shopping type as a shopping tendency label of the user group; and generating a user portrait of the user according to the shopping tendency label and the category characteristics of the user group to which the user belongs.
In another aspect, an embodiment of the present application provides a user representation generating apparatus, including:
a recommended article information acquisition unit: the system comprises a shopping application program, a user account and a server, wherein the shopping application program is used for pushing a user account to a user;
commodity source side information extraction unit: the system is used for extracting commodity source side information from the recommended commodity information;
a commodity type matching unit: the system comprises a commodity information acquisition module, a commodity information classification module and a commodity information classification module, wherein the commodity information acquisition module is used for acquiring commodity source information of a commodity;
shopping type distribution statistical unit: the system comprises a commodity distribution server, a user account distribution server and a user group management server, wherein the commodity distribution server is used for determining shopping type distribution of each user group according to commodity types and at least one user group to which a user corresponding to the user account belongs; the user group is determined according to user identity information of the user and online operation behavior data of the user;
a user profile generation unit: and the user portrait of the user is generated according to the shopping type distribution of each user group.
In another aspect, an embodiment of the present application provides a user representation generating device, where the device includes: a processor and a memory storing computer program instructions; a processor, when executing computer program instructions, implements a user representation generation method as described in any of the above embodiments.
In yet another aspect, an embodiment of the present application provides a computer storage medium, where computer program instructions are stored, and when executed by a processor, the computer program instructions implement the user representation generation method as described in any one of the above embodiments.
According to the user portrait generation method, device and equipment and the computer storage medium, the user portrait of the user is generated according to the shopping type distribution of the user group where the user is located, the shopping type distribution of the user group is obtained based on the commodity types obtained by classifying the recommended commodity information pushed to the user account, and the recommended commodity information pushed to the user account is more comprehensive compared with the shopping browsing records of the user and the shopping invoice information of the user, so that the user portrait generated based on the method is more accurate.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart diagram illustrating a user representation generation method according to an embodiment of the present application;
FIG. 2 is a schematic flow chart diagram illustrating another user representation generation method according to another embodiment of the present application;
FIG. 3 is a user feature variable selection diagram of a user representation generation method according to yet another embodiment of the present application;
FIG. 4 is a schematic diagram of a user representation generation apparatus according to still another embodiment of the present application;
FIG. 5 is a schematic diagram of a user representation generating device according to another embodiment of the present application.
Detailed Description
Features and exemplary embodiments of various aspects of the present application will be described in detail below, and in order to make objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended to be illustrative only and are not intended to be limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present application by illustrating examples thereof.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
In order to solve the prior art problem, embodiments of the present application provide a user portrait generation method, apparatus, device, and computer storage medium, which can obtain complete and real user shopping behavior data and generate an accurate user portrait. The above-described technology of the present application will be described in detail below with reference to the accompanying drawings.
FIG. 1 is a flow chart illustrating a user representation generation method according to an embodiment of the present application. As shown in fig. 1, the method comprises the following steps:
step S100: acquiring recommended commodity information pushed to a user account by a shopping application program;
when a user finishes shopping operation in a shopping application program, the shopping application program often pushes corresponding recommended commodity information to a user account. As an example, the recommended merchandise information may include at least one of merchandise source information, shopping information confirmation, merchandise evaluation feedback, merchandise promotion reminder, and the like. As an example, the commodity source information may include a store name, a shopping platform name, and the like.
When the shopping application program pushes the recommended commodity information to the user account, the embodiment of the application can acquire the recommended commodity information pushed to the user account by the shopping application program.
As an example, in order to obtain the recommended commodity information, the communication operator is first required to provide a data interface through which all the recommended commodity information pushed to the user account by the shopping application can be collected.
Step S110: extracting commodity source side information from the recommended commodity information;
since the product type is generally related to the product source information, it is necessary to extract the product source information from the recommended product information in order to identify the product type in the recommended product information.
In the embodiment of the present application, information extraction methods commonly used in the field may be adopted to extract the information of the commodity source in the commodity recommendation information, for example, the information of the commodity source in the commodity recommendation information is extracted by a content search method of BS4(Beautiful Soup 4). And privacy information may be included in the goods recommendation information. For example, the name of the commodity purchased by the user, the amount of the purchased commodity, and the like, in order not to violate the privacy information of the user, as an example, this step may employ a regular expression manner to extract the information of the commodity source in the commodity recommendation information, and taking Python as an example, re.
For example, the following steps are carried out: in the case where the source of the recommended product information is marked at the beginning of the recommended product information, for example, the shopping application program transmits the following information to a certain user account: "[ OLAY Tianmao flagship ] your light sense white bottle has been shipped! Add and give you a 30 yuan of voucher → m.tb.cn/h.4br4xcp with the big red bottle short of crease-resistant & compact! And returning to T and retreating. Through the regular expression, the commodity information source party can be identified to be the 'OLAY Tianmao flagship', the commodity source party information does not relate to the identity information of the user, but according to specific information content, the fact that the user purchases 'pinky bottles' can be definitely known, and the behavior belongs to the privacy behavior of invading the user. Therefore, in this embodiment, only the commodity source side information of the recommended commodity information, namely the "ocean sky cat flagship", needs to be identified by the regular expression, so that the privacy of the user cannot be violated.
Step S120: classifying the recommended commodity information according to the commodity source information to obtain a corresponding commodity type;
in general, the types of goods in the recommended goods information pushed by the shopping application may reflect the shopping tendency of the user, and therefore, the recommended goods information needs to be classified to obtain the types of goods corresponding to the recommended goods information.
In the embodiment of the present application, the commodity type of the recommended commodity information may be confirmed by the commodity type corresponding to the commodity source information extracted in step S110. As an example, the information of the commodity source and a merchant information table obtained by crawling the application data may be matched to obtain a store name corresponding to the information of the commodity source and a commodity type corresponding to the store name, the merchant information table includes the store name, the commodity type, the number of stores, and the like, for example, table 1 is the merchant information extracted from a certain application. Then, the commodity type corresponding to the store name may be determined as the commodity type corresponding to the commodity source information, and the commodity type corresponding to the commodity source information may be determined as the commodity type corresponding to the recommended commodity information. For example, table 2 shows the product types corresponding to the recommended product information obtained by matching the recommended product information of a certain shopping application with table 1.
TABLE 1 Merchant information Table
Table 2 commodity type matching table corresponding to commodity information recommended by shopping application
Step S130: determining the shopping type distribution of each user group according to the commodity type and at least one user group to which the user corresponding to the user account belongs; the user group is determined according to the user identity information of the user and the online operation behavior data of the user;
consumption ethology considers that a group of consumers with certain common characteristics will exhibit the same or similar consumption behaviors. Therefore, the shopping tendency of the user can be analyzed through the shopping type distribution of the user group to which the user belongs.
In the embodiment of the present application, the purchase proportion of each commodity type in each user group in a certain time period may be counted, for example, the purchase proportion represents a ratio of the total purchase times of each commodity type in the user group to the purchase times of all commodities; according to the purchase proportion, drawing a shopping probability distribution diagram of each user group in different commodity types within a certain time period; and obtaining the shopping type distribution of each user group through the shopping probability distribution map.
In the embodiment of the present application, in order to make the generated user image more representative, it is necessary to classify users by considering the attributes of the users in multiple aspects to obtain a representative user group.
As an example, as shown in fig. 2, a user needs to register an account with own identity information when using an application, so that the identity information and online operation behavior data of the user can be obtained from the user account, and the feature variable of the user is extracted from the user identity information and the online operation behavior data of the user; performing correlation analysis on the characteristic variables to obtain mutually independent continuous characteristic variables; and clustering the users by taking the continuous characteristic variables as indexes based on a maximum expectation algorithm of a GMM Gaussian mixture model, and selecting the optimal classification number of the users by applying a Bayesian information criterion to obtain at least one user group to which the users belong. Then, the purchase ratio of each item type in each user group in a certain time period may be counted, for example, the purchase ratio may be represented by a ratio of the total purchase number of each item type in the user group to the purchase number of all items. And finally, according to the purchase proportion, a shopping probability distribution diagram of each user group in different commodity types in a certain time period can be drawn, and further the shopping type distribution of each user group can be obtained.
As an example, a representative user group can be obtained by the following method.
Consumption ethology considers that a group of consumers with certain common characteristics will exhibit the same or similar consumption behaviors. Some scholars consider these characteristics to include internal and external factors. The physiological aspects of the internal factors include age and sex, the psychological aspects include character, psychological tendency and consumption habit, and the external factors include productivity development level, cultural background, ethnic group, religious belief, geographical climate condition and the like.
Therefore, in order to determine the user group to which the user belongs, the feature vector of the user can be extracted from the identity information of the user and the online operation behavior data to perform analysis statistics. The four interrelated characteristics of education, occupation, income level and residence are summarized by using the per-capita GDP of the city of the place where the user is located; the behavior habit of the online operation of the user is used for approximating the three characteristics of character, psychological tendency and behavior habit. As an example, the feature variables on which the user population is divided may be designed to be expanded from two aspects of user identity information and online operation behavior data of the user as shown in fig. 3. The description of the online operation behavior data of the user is completed through several ratios, and the ratio of the duration of using the video type App (application program), the duration of using the game type App, the duration of using the shopping type App and the social type APP to the total APP duration is calculated respectively.
In summary, 7 user feature vectors can be determined, which are respectively gender, age, average GDP of local market, total length of using time of video App, total length of using time of game App, total length of using time of shopping App, total length of using time of social App, and total length of using time of social App, wherein gender is a discrete variable, and others are continuous variables.
The identity information and the online operation behavior data of the user are extracted through the user account, and the specific obtaining steps are shown in table 3.
TABLE 3 details of the preprocessing of the user characteristic variables
After the feature variables of the user are extracted, correlation analysis needs to be performed on the feature variables to determine mutually independent continuous feature variables. And carrying out correlation analysis on user characteristic variables such as age, average GDP of local cities, ratio of video App use duration to total duration, ratio of game App use duration to total duration, ratio of shopping App use duration to total duration and the like, if the variables are independent, carrying out the next step, and if the variables with strong correlation exist, aggregating mutually independent factor variables through factor analysis. Wherein the correlation of the user characteristic variables can be determined according to the pearson coefficient: calculating the correlation between every two variables, wherein the correlation coefficient is less than 0.1, namely the variables are independent; the correlation coefficient is less than 0.5, is weak correlation and can be accepted according to the condition, and the correlation coefficient is relatively independent as a variable; the correlation coefficient is 0.5 to 0.7, which is stronger correlation; the correlation coefficient is 0.7 or more, and is a strong correlation.
After the user characteristic variables are determined, the users are clustered next. The user sample is divided into two subsamples by gender (male and female samples are separately clustered). Then, clustering is carried out on the two subsamples by taking the initial 6 continuous characteristic variables or the independent variables after factor aggregation as indexes. And determining the optimal class number of the user by using a Bayesian information criterion when the user is clustered by using an EM algorithm of a GMM Gaussian mixture model. The specific algorithm steps are as follows:
step1, the total data amount is N; let the initial number of subdistributions, i.e. the number of initial classes K, equal to 15
Step2. clustering data into K classes by K-means
Step3. class k data with a capacity of NkCalculating the weight pik＝NkN, mean vector μkCovariance matrix sigmak
Step4. estimate data xiProbability generated by kth sub-distribution
Wherein
D is the dimension of data, if the initial 6 characteristic variables are mutually independent, D is 6, otherwise, D is equal to the number of the factor variables aggregated after factor analysis
Step5. update
πk＝Nk/N
Step6. solving parameter mu by maximum likelihood estimationk，∑k
Step7, repeating the steps 4, 5 and 6 until the parameters are converged (unchanged), and entering the next Step
Step8. calculating likelihood function values
Step9. record the classification when K15 and the bayesian information content BIC 2kln (n) -2ln (l)
Step 2-step 9 are repeated until K is 2. K is K-1
Step11, outputting a line graph with the abscissa as K and the ordinate as BIC; determining the classification condition with the minimum BIC as the final classification result, and outputting the classification condition (including classification number, the class of each sample)
If the users are clustered into K classes by the algorithm and then combined with the gender, the users can be divided into 2K classes of groups.
And drawing a distribution graph of each index of each group, analyzing class characteristics through the distribution condition of the indexes, and naming the groups of different classes. For example, if a certain group is in a female subsample, the age is mainly distributed in 20-30 years, the average GDP of the local market is low between each category, and the using duration of the apps of the video and social class is higher than the duration indexes of the other two apps, the group can be named as a "young female group oriented to video entertainment in the area with low economy" or other names corresponding to research purposes. It should be noted that if the originally designed variables are aggregated into factor variables through factor analysis, the proper connotation of the factor variables can also be used to describe the population.
As an example, after obtaining the user group to which the user belongs, the purchase ratio of each commodity type in each user group in a certain time period is counted, and the following method may be used.
A time period can be set for the 2K user groups determined in the above embodiment, and the purchase proportion of each type of commodity in each user group in the time period is statistically analyzed, where the purchase proportion is the ratio of the total purchase frequency of each commodity type in the user group to the purchase frequency of all commodities, and the method specifically includes the following steps:
step1, counting the number N of users for each type of consumption group K of 1,2, … and 2KkBrowsing the class number M of productsk
Step2. in the k-th consumption group, for each type of products, counting the times t that each user i purchases j types of products within a certain period of time (according to analysis requirements, during activity, in a certain month, in a certain quarter and in a certain year)k(i,j)，j＝1,2,…,Mk
Step3. calculate the total times of all users in the k-th group purchasing j products in the period
Step4, calculating the proportion of the purchase times of various products to the total browsing times
Step6, outputting the proportion P of each consumption group to each product purchase amountk(j)＝{p(k,j)|j＝1,2,…,Mk}，k＝1,2,…,2K
As an example, according to the purchase proportion of each commodity type in each user group, a shopping probability distribution graph of the purchase amount ratio of users in each user group to different types of commodities can be drawn, that is, the shopping type distribution of different user groups can be evaluated, and the specific steps are as follows:
step1, compiling a browsing amount distribution table for each user group, wherein the column names are product types, the table elements are browsing amount ratios, and the browsing amount ratio of the consumption group k to j products is P in the output resultk(j)
Step2, for each user group, arranging the elements in the browsing amount distribution table in descending order, and arranging the product category with the largest browsing amount in the group at the first position
Step3, for each user group, taking the product category as an abscissa, drawing a histogram according to element values, and marking percentage
And step4. accordingly, the probability is estimated by using the frequency according to the law of large numbers, namely, the proportion or the possibility of the user group in a certain period of time to like a certain product can be obtained.
In the embodiment of the application, the shopping probability distribution map of each user group is drawn by statistically analyzing the purchase proportion of each type of commodity in each user group in a certain time period, so that the shopping type distribution of each user group is obtained. In the embodiment of the application, the shopping type distribution of each user group is determined, the weight proportion of the corresponding user group is obtained, the weight proportion is calculated by the actual shopping amount of the user group, the rigor is high, and the persuasion is sufficient.
For example, the following steps are carried out: a certain group is a female young group with the characteristics of video entertainment orientation in an economically developed area in a female child sample, and the consumed commodity types in the user group comprise skin care color cosmetics, ladies 'clothes, bags and the like through analysis, so that the shopping type distribution of the user group can be obtained through counting the shopping proportion of the commodity types of the skin care color cosmetics, the ladies' clothes, the bags and the like.
Step S140: and generating the user portrait of the user according to the shopping type distribution of each user group.
Because the user representation is a tagged user model abstracted from the user's information such as attributes, preferences, habits, behaviors, etc., the user can be described by some highly generalized features. Therefore, it is necessary to generate a user representation by combining the shopping type distribution of the user group and the user group characteristics.
In the embodiment of the application, the shopping tendency of each user group can be determined by analyzing the shopping type distribution of each user group, and then the user portrait of each user in the user group is generated by combining the group characteristics of each user group. As an example, the shopping type with the largest shopping type distribution percentage may be selected from the shopping type distribution of each user group and/or the shopping type satisfying a preset threshold may be extracted as the shopping tendency tag of the user group, and then the user representation of the user may be generated according to the shopping tendency tag and the category characteristics of the user group to which the user belongs.
In the embodiment of the present application, the shopping tendency tag extraction rule of the user group may be changed according to different scene requirements, and is not described in detail herein.
The foregoing is a specific implementation manner of the user portrait generation method provided in the embodiment of the present application. The user portrait of the user is generated according to the shopping type distribution of the user group where the user is located, the shopping type distribution of the user group is obtained based on the commodity types obtained by classifying the recommended commodity information pushed to the user account, and the recommended commodity information pushed to the user account is more comprehensive than the shopping browsing records of the user and the shopping invoice information of the user, so that the user portrait generated based on the method is more accurate.
As another implementation manner of the present application, in order to reduce the calculation difficulty, before matching the information of the commodity source with a merchant information table acquired in advance, the method may further include the following steps:
and extracting merchant information meeting a first preset condition from the merchant information table, wherein the first preset condition comprises that the score of the shop score is greater than a preset score and/or the commodity sales volume is greater than a preset sales volume.
As is known, when a shopping application program includes a large number of merchant stores, and the information of the source of each item of recommended merchandise information is extracted through a regular expression, all the merchant information in the shopping application program needs to be traversed, which makes the calculation difficult.
In the embodiment of the present application, as an example, a feasible method is to extract merchant information meeting the limitation condition by limiting the score of the store and the sales volume of the commodity, and then match the information of the commodity source side with the merchant information meeting the limitation condition. The method can reduce the total number of merchant samples and the calculation amount, and stores which push commodity information for marketing through the shopping application program are often stores with a certain sales amount and higher scores, so that the identification precision cannot be greatly influenced by the limiting method.
As another implementation manner of the present application, in order to save the calculation cost and improve the calculation efficiency, the method for matching the information of the commodity source with the pre-obtained merchant information table and determining the store name corresponding to the information of the commodity source and the commodity type corresponding to the store name specifically includes the following steps:
step A: the character matching degree of the commodity source party information and the character matching degree of the merchant information table meet preset conditions, and shop names corresponding to the commodity source party information are obtained, wherein the preset conditions comprise that the character matching degree of the commodity source party information and the shop names in the merchant information table is larger than preset percentage;
different distributors may exist for the same brand of goods, and the corresponding store names may be different in the shopping application, for example, the store name corresponding to the brand "nike" may be: "NIKE official flagship store", "NIKE exclusive store". When matching the store names in the commodity source information and the merchant information table, a plurality of store names may be identified at the same time, because the commodity types of the stores are the same, the result is not affected. However, when the information of the source of the goods is extracted, all the merchant information is traversed, which increases the calculation cost.
In the embodiment of the present application, as an example, a condition may be set, and when a store name satisfying a preset condition is matched when the store name satisfying the preset condition is matched with the commodity source information, the program is stopped, and the store name satisfying the condition is determined as the store name corresponding to the commodity source information.
For example, the preset condition is that the character matching degree of the store name in the commodity source side information and the merchant information table exceeds 75%, and the program is stopped. For example, when the commodity source party is [ NIKE brand exclusive shop ], 77.77% characters of the commodity source party can be matched with the shop name of the "NIKI brand flagship shop", so that the shop name corresponding to the "NIKI brand flagship shop" is directly determined, and the program does not need to be continuously run.
And B: extracting a store score corresponding to the store name from the merchant information table when the store name corresponds to a plurality of commodity types; and determining the commodity type with the top ranking ratio as the commodity type corresponding to the store name according to the ranking ratio of the store scores in all stores corresponding to each commodity type.
Under the condition that the same shop can operate a plurality of commodity types, in order to make calculation fast and convenient, the commodity type corresponding to the shop name can be determined through the shop score.
In the embodiment of the present application, as an example, a method may be possible in which a store score corresponding to a store name is extracted from a store information table, ranking ratios of the store scores in all stores corresponding to each item type are compared, and an item type with the top ranking ratio is determined as an item type corresponding to the store name.
For example, the store name is "NIKI brand flagship store", the commodity types of the store include men's shoes, women's shoes and bags, and the ranking proportion of the men's shoes as the most front commodity type can be obtained by comparing the ranking proportions of the stores of the "NIKI brand flagship store" in all stores corresponding to the several commodity types of the men's shoes, the women's shoes and the bags, so the men's shoes are determined as the commodity type of the "NIKI brand flagship store".
In the embodiment of the application, the character matching degree of the information of the commodity source party and the shop name in the merchant information table is larger than the preset percentage by presetting conditions to determine the shop name corresponding to the information of the commodity source party, so that the aim of saving the calculation cost can be fulfilled. In addition, the commodity type corresponding to the store name is determined by comparing the ranking proportion of the store scores in the stores of all the commodity types, so that the calculation process is quicker and more convenient.
FIG. 4 is a schematic diagram illustrating a user representation generating apparatus according to still another embodiment of the present application. As shown in fig. 4, the user representation generation apparatus includes:
the recommended article information acquisition unit 10: the system comprises a shopping application program, a user account and a server, wherein the shopping application program is used for pushing a user account to a user;
commodity source side information extraction unit 20: the system is used for extracting commodity source side information from the recommended commodity information;
the article type matching unit 30: the system comprises a commodity information acquisition module, a commodity information classification module and a commodity information classification module, wherein the commodity information acquisition module is used for acquiring commodity source information of a commodity;
shopping type distribution statistical unit 40: the system comprises a commodity distribution server, a user account distribution server and a user group management server, wherein the commodity distribution server is used for determining shopping type distribution of each user group according to commodity types and at least one user group to which a user corresponding to the user account belongs; the user group is determined according to user identity information of the user and online operation behavior data of the user;
user representation generation unit 50: the user portrait of the user is generated according to the shopping type distribution of each user group.
According to the user portrait generation device, complete and real user shopping behavior data can be acquired by comprehensively monitoring and collecting commodity information pushed by a shopping application program received by a user account, and accurate user portrait is generated. By extracting the commodity source side information from the commodity information pushed by the shopping application program, the detailed content of the pushed information is not involved, and the invasion of the privacy of the user is also avoided.
FIG. 5 shows a hardware structure diagram of a user representation generation device according to an embodiment of the present application.
The user representation generating device may include a processor 501 and a memory 502 storing computer program instructions.
Specifically, the processor 501 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of the embodiments of the present Application.
The memory may include Read Only Memory (ROM), Random Access Memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors), it is operable to perform operations described with reference to the methods according to an aspect of the present disclosure.
The processor 501 reads and executes computer program instructions stored in the memory 502 to implement any of the user representation generation methods in the above embodiments.
In one example, the user representation generating device may also include a communication interface 503 and bus 500. As shown in fig. 5, the processor 501, the memory 502, and the communication interface 503 are connected via the bus 500 to complete communication therebetween.
The communication interface 503 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiments of the present application.
In addition, in combination with the user portrait generation method in the foregoing embodiment, the embodiment of the present application may be implemented by providing a computer storage medium. The computer storage medium having computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement any of the user representation generation methods of the above embodiments.
It is to be understood that the present application is not limited to the particular arrangements and instrumentality described above and shown in the attached drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present application are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions or change the order between the steps after comprehending the spirit of the present application.
The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the present application are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware for performing the specified functions or acts, or combinations of special purpose hardware and computer instructions.
As described above, only the specific embodiments of the present application are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present application, and these modifications or substitutions should be covered within the scope of the present application.
Claims (10)
1. A user representation generation method, comprising:
acquiring recommended commodity information pushed to a user account by a shopping application program;
extracting commodity source side information from the recommended commodity information;
classifying the recommended commodity information according to the commodity source information to obtain a corresponding commodity type;
determining the shopping type distribution of each user group according to the commodity type and at least one user group to which the user corresponding to the user account belongs; the user group is determined according to the user identity information of the user and the online operation behavior data of the user;
and generating the user portrait of the user according to the shopping type distribution of each user group.
2. The user representation generation method of claim 1, wherein the extracting of the information of the source of the product from the recommended product information comprises:
traversing the recommended commodity information based on a regular expression, and extracting commodity source side information corresponding to the recommended commodity information.
3. The user representation generation method of claim 1, wherein the classifying the recommended merchandise information according to the merchandise source information to obtain a corresponding merchandise type specifically comprises:
matching the commodity source party information with a pre-acquired merchant information table, and determining a store name corresponding to the commodity source party information and a commodity type corresponding to the store name, wherein the merchant information table is obtained according to the crawling of the shopping application program data;
determining the commodity type corresponding to the shop name as the commodity type corresponding to the commodity source party information;
and determining the commodity type corresponding to the commodity source information as the commodity type corresponding to the recommended commodity information.
4. The user representation generation method of claim 3, wherein prior to matching the merchandise source information with a pre-obtained merchant information table, further comprising:
and extracting merchant information meeting a first preset condition from the merchant information table, wherein the first preset condition comprises that the score of the shop score is greater than a preset score and/or the commodity sales volume is greater than a preset sales volume.
5. The user representation generation method of claim 1, wherein before determining a shopping type distribution of each of the user groups according to the commodity type and at least one user group to which the user corresponding to the user account belongs, the method further comprises:
acquiring user identity information of the user and online operation behavior data of the user;
extracting characteristic variables of the user from the user identity information of the user and the online operation behavior data of the user;
performing correlation analysis on the characteristic variables to determine mutually independent continuous characteristic variables;
and clustering the users by taking the continuous characteristic variables as indexes based on a maximum expectation algorithm of a GMM Gaussian mixture model, and selecting the optimal classification number of the users by applying a Bayesian information criterion to obtain at least one user group to which the users belong.
6. The user representation generation method of claim 1, wherein the determining a shopping type distribution of each user group according to the commodity type and at least one user group to which the user corresponding to the user account belongs specifically comprises:
counting the purchase proportion of each commodity type in each user group in a preset time period, wherein the purchase proportion represents the ratio of the total purchase times of each commodity type in the user group to the purchase times of all commodities;
according to the purchase proportion, drawing a shopping probability distribution diagram of each user group in different commodity types within a preset time period;
and determining the shopping type distribution of each user group according to the shopping probability distribution map.
7. The method of claim 1, wherein generating a user representation of the user based on the shopping type distribution of each of the user groups comprises:
selecting the shopping type with the largest shopping type distribution percentage and/or extracting the shopping type meeting a preset threshold value from the shopping type distribution of each user group, and determining the shopping type as the shopping tendency label of the user group;
and generating a user portrait of the user according to the shopping tendency label and the category characteristics of the user group to which the user belongs.
8. A user representation generation apparatus, the apparatus comprising:
a recommended article information acquisition unit: the system comprises a shopping application program, a user account and a server, wherein the shopping application program is used for pushing a user account to a user;
commodity source side information extraction unit: the system is used for extracting commodity source side information from the recommended commodity information;
a commodity type matching unit: the system is used for classifying the recommended commodity information according to the commodity source information to obtain a corresponding commodity type;
shopping type distribution statistical unit: the shopping type distribution of each user group is determined according to the commodity type and at least one user group to which the user corresponding to the user account belongs; the user group is determined according to the user identity information of the user and the online operation behavior data of the user;
a user profile generation unit: and the user portrait of the user is generated according to the shopping type distribution of each user group.
9. A user representation generation apparatus, characterized in that the apparatus comprises: a processor and a memory storing computer program instructions;
the processor, when executing the computer program instructions, implements a user representation generation method as claimed in any of claims 1-7.
10. A computer storage medium having computer program instructions stored thereon that, when executed by a processor, implement a user representation generation method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403203.7A CN112465598A (en) | 2020-12-04 | 2020-12-04 | User portrait generation method, device and equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403203.7A CN112465598A (en) | 2020-12-04 | 2020-12-04 | User portrait generation method, device and equipment and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112465598A true CN112465598A (en) | 2021-03-09 |
Family
ID=74805405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011403203.7A Pending CN112465598A (en) | 2020-12-04 | 2020-12-04 | User portrait generation method, device and equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112465598A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113313546A (en) * | 2021-04-19 | 2021-08-27 | 深圳市竹芒信息技术有限公司 | Information recommendation method and device, computer equipment and storage medium |
| CN113407826A (en) * | 2021-06-09 | 2021-09-17 | 广州三七极创网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
| CN115170212A (en) * | 2022-09-08 | 2022-10-11 | 咚咚来客(广州)信息技术有限公司 | Private domain operation data management method based on chain brands and related device |
| CN116109338A (en) * | 2022-12-12 | 2023-05-12 | 广东南粤分享汇控股有限公司 | Electric business analysis method and system based on artificial intelligence |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150061082A (en) * | 2013-11-25 | 2015-06-04 | 에스케이플래닛 주식회사 | System, apparatus and mehtod for performing product recommendation based on personal information |
| CN109558530A (en) * | 2018-10-23 | 2019-04-02 | 深圳壹账通智能科技有限公司 | User's portrait automatic generation method and system based on data processing |
| CN110610384A (en) * | 2019-09-20 | 2019-12-24 | 上海掌门科技有限公司 | User portrait generation method, information recommendation method, device and readable medium |
| CN112001754A (en) * | 2020-08-21 | 2020-11-27 | 上海风秩科技有限公司 | User portrait generation method, device, equipment and computer readable medium |
-
2020
- 2020-12-04 CN CN202011403203.7A patent/CN112465598A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150061082A (en) * | 2013-11-25 | 2015-06-04 | 에스케이플래닛 주식회사 | System, apparatus and mehtod for performing product recommendation based on personal information |
| CN109558530A (en) * | 2018-10-23 | 2019-04-02 | 深圳壹账通智能科技有限公司 | User's portrait automatic generation method and system based on data processing |
| CN110610384A (en) * | 2019-09-20 | 2019-12-24 | 上海掌门科技有限公司 | User portrait generation method, information recommendation method, device and readable medium |
| CN112001754A (en) * | 2020-08-21 | 2020-11-27 | 上海风秩科技有限公司 | User portrait generation method, device, equipment and computer readable medium |
Non-Patent Citations (1)
| Title |
|---|
| 王庆福;: "贝叶斯网络在用户兴趣模型构建中的研究", 无线互联科技, no. 12 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113313546A (en) * | 2021-04-19 | 2021-08-27 | 深圳市竹芒信息技术有限公司 | Information recommendation method and device, computer equipment and storage medium |
| CN113407826A (en) * | 2021-06-09 | 2021-09-17 | 广州三七极创网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
| CN115170212A (en) * | 2022-09-08 | 2022-10-11 | 咚咚来客(广州)信息技术有限公司 | Private domain operation data management method based on chain brands and related device |
| CN115170212B (en) * | 2022-09-08 | 2022-11-25 | 咚咚来客(广州)信息技术有限公司 | Private domain operation data management method based on chain brands and related device |
| CN116109338A (en) * | 2022-12-12 | 2023-05-12 | 广东南粤分享汇控股有限公司 | Electric business analysis method and system based on artificial intelligence |
| CN116109338B (en) * | 2022-12-12 | 2023-11-24 | 广东南粤分享汇控股有限公司 | Electric business analysis method and system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112465598A (en) | User portrait generation method, device and equipment and computer storage medium | |
| CN106776619B (en) | Method and device for determining attribute information of target object | |
| CN107833082B (en) | Commodity picture recommendation method and device | |
| CN109064285B (en) | Commodity recommendation sequence and commodity recommendation method | |
| CN110807669B (en) | Cross-platform user information management method and device | |
| CN109816134B (en) | Method and device for predicting delivery address and storage medium | |
| CN110933472B (en) | Method and device for realizing video recommendation | |
| CN108805598A (en) | Similarity information determines method, server and computer readable storage medium | |
| CN109711917B (en) | Information pushing method and device | |
| CN111198989A (en) | Method and device for determining travel recommendation data, storage medium and electronic equipment | |
| CN114581207A (en) | Commodity image big data accurate pushing method and system for E-commerce platform | |
| CN103577472A (en) | Method and system for obtaining and presuming personal information as well as method and system for classifying and retrieving commodities | |
| CN114519616A (en) | Commodity popularization method | |
| CN111340566B (en) | Commodity classification method and device, electronic equipment and storage medium | |
| CN111709809B (en) | Shopping mall shopping consumption recommendation system based on block chain | |
| CN106204163B (en) | Method and device for determining user attribute characteristics | |
| CN111225009B (en) | Method and device for generating information | |
| CN114841760B (en) | Advertisement recommendation management method and system based on audience behavior characteristic analysis | |
| CN115953225A (en) | Commodity recommendation method and device based on user emotion | |
| CN116127184A (en) | Product recommendation method and device, nonvolatile storage medium and electronic equipment | |
| CN115131108A (en) | E-commerce commodity screening system | |
| CN112070562B (en) | Commodity identification method based on big data and electronic commerce and e-commerce system | |
| CN111127128B (en) | Commodity recommendation method, commodity recommendation device and storage medium | |
| CN113327132A (en) | Multimedia recommendation method, device, equipment and storage medium | |
| CN114078037A (en) | Commodity recommendation method and device based on label vectorization |
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 |