CN116228282B - Intelligent commodity distribution method for user data tendency - Google Patents
Intelligent commodity distribution method for user data tendency Download PDFInfo
- Publication number
- CN116228282B CN116228282B CN202310510872.1A CN202310510872A CN116228282B CN 116228282 B CN116228282 B CN 116228282B CN 202310510872 A CN202310510872 A CN 202310510872A CN 116228282 B CN116228282 B CN 116228282B
- Authority
- CN
- China
- Prior art keywords
- user
- paid
- matrix
- blog
- similarity
- 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 239000011159 matrix material Substances 0.000 claims abstract description 272
- 238000004364 calculation method Methods 0.000 claims abstract description 29
- 238000005457 optimization Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000012937 correction Methods 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 24
- 230000006399 behavior Effects 0.000 description 16
- 238000004891 communication Methods 0.000 description 7
- 238000012300 Sequence Analysis Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 241000669618 Nothes Species 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000011524 similarity measure Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of commodity distribution, and discloses an intelligent commodity distribution method with user data tendency, which comprises the following steps: constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix; calculating to obtain a user similarity matrix, and carrying out high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix, so as to construct a user enhancement similarity matrix; and determining a user preference objective function representing the user tendency, carrying out optimization solving to obtain the user preference of different paid blogs, and carrying out paid blogs distribution processing. According to the application, a user preference objective function for representing the preference trend of similar users on the paid blogs is obtained according to the user-paid blogs preference matrix and the user enhanced similarity matrix, the user characteristic matrix and the paid blogs preference characteristic matrix are obtained through solving, the preference values of the users on different paid blogs are obtained through calculation, and the pay blogs distribution processing is realized.
Description
Technical Field
The application relates to the technical field of commodity distribution, in particular to an intelligent commodity distribution method with user data tendency.
Background
With the continuous development of the information society, people can conveniently acquire various information through the internet, but with the rapid increase of the information scale in the internet, people often need to spend more time and effort to find the information required by themselves. The recommendation system is used as an important technical means for relieving the information overload problem, and personalized recommendation is formed by mining historical behavior data of users. In the pay blog recommendation process, each blog is provided with an independent label for distinguishing the characteristics of the blog, the existing blog recommendation is mostly carried out based on the labels, but along with the increase of the index of the blog resource presentation, a large number of blogs with the same label type appear, and how to realize the refined recommendation of the pay blog under the condition is an important way for improving the user experience. Aiming at the problem, the application provides an intelligent commodity distribution method with user data tendency, which improves the user recommendation satisfaction of paid blogs.
Disclosure of Invention
In view of the above, the present application provides an intelligent commodity distribution method with user data tendency, which aims at: 1) According to the buying condition of a user on a paid blog and the quotation relation between the purchased paid blog and Yu Bo thereof, a user-paid blog collection matrix and a paid Fei Boke-paid blog quotation matrix are respectively constructed, by considering the problem that the preference of the user on the paid blog changes along with time, time weight is introduced when the similarity among users is calculated, namely, the larger the occupation ratio of the difference value between the time of buying the same paid blog and the time of buying the paid blog for the first time is, the worse the real-time performance of the paid blog to the user is indicated, the lower the corresponding time weight is, and the similarity among users is calculated based on the time weight and the behavior information of buying the paid blog by the user, so that the similarity measurement of different users is realized; 2) Constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix as user-paid blog preference matrices, sequentially converting the user similarity matrix into an adjacent matrix and an undirected graph, wherein the user is the vertex in the undirected graph, converting the user similarity into a high-order similarity measure combined with the weight of the adjacent user point edges by analyzing the closed triangle structure relationship among the users, further constructing a user enhanced similarity matrix, obtaining a user preference objective function representing the tendency of similar users to pay blog preferences according to the user-paid blog preference matrix and the user enhanced similarity matrix, carrying out optimization solving on the constructed user preference objective function, solving to obtain a user characteristic matrix and a paid blog preference characteristic matrix, further calculating to obtain the preference values of the users to different paid blogs, and carrying out paid blog distribution processing according to the sorting sequence of the preference values.
The application provides an intelligent commodity distribution method for user data trend, which comprises the following steps:
s1: collecting behavior information of purchasing a paid blog by a user and reference relation information among the paid blogs, respectively constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, and constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix;
s2: calculating to obtain a user similarity matrix according to the user-paid blog collection matrix and behavior information of the user purchasing the paid blog, and performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix, wherein the mode sequence analysis is a main implementation mode of the high-order similarity calculation;
s3: constructing a user enhancement similarity matrix according to the calculated user high-order similarity matrix and the user similarity matrix;
s4: determining a user preference objective function representing user tendency according to the constructed user enhancement similarity matrix;
s5: and carrying out optimization solving on the constructed user preference objective function to obtain the preferences of the user on different paid blogs and carrying out paid blogs distribution processing.
As a further improvement of the present application:
optionally, in the step S1, collecting behavior information of purchasing a paid blog by a user and reference relation information between paid blogs, and constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, including:
collecting behavior information of purchasing a paid blog by a user, and referring relation information between the paid blog and the rest paid blogs, wherein the collected information is expressed in the following form:
；
；
wherein:
，indicating that user i purchased a paid blog j, < >>Indicating that user i did not purchase pay blog j, if +.>Then->Indicating the time when user i purchased pay blog j, if +.>Then->Is empty;
n represents the total number of selected users, m represents the total number of selected paid blogs;
，indicating that pay blog j does not reference pay blog h, < ->Indicating that pay blog j references pay blog h, if +.>Then->；
Respectively constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, wherein the constructed user-paid blog collection matrix G is as follows:
；
the constructed paid blog-paid blog reference matrix C is:
；
wherein:
representing a reference relationship between the paid blog 1 and the paid blog m.
Optionally, the step S1 constructs a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix, including:
constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix:
；
；
wherein:
indicating the preference degree of user i for pay blogs j, < >>Representing the total number of references to pay blogs purchased by user i.
Optionally, the calculating in step S2 to obtain the user similarity matrix includes:
calculating to obtain a user similarity matrix according to the user-paid blog collection matrix and the behavior information of purchasing the paid blog by the user, wherein the calculation flow of the user similarity matrix is as follows:
calculating time weights of different users on paid blogs:
；
wherein:
representing a small positive number;
representing the time weights of user i and user u for pay blogs j;
e represents a natural constant;
indicating the time when user i first purchased the pay blog,/->Representing the time when user u first purchased the pay blog, R represents a set time threshold,/->，；
Representing a time difference between a last purchase of the paid blog and a first purchase of the paid blog by the user i;
calculating the similarity between different users based on the time weight of the users to the paid blogs and the user-paid blogs collection matrix:
；
wherein:
representing the similarity between user i and user j;
constructing a user similarity matrix S:
；
wherein:
the values on the diagonal lines in the user similarity matrix S each represent the user' S own similarity, which is set to 1.
Optionally, in the step S2, performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix, including:
performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix, wherein the mode sequence analysis is a main implementation mode of the high-order similarity calculation, and the high-order similarity calculation flow is as follows:
s21: setting the values on the diagonal lines in the user similarity matrix S to 0 to obtain a matrix；
S22: matrix-basedCalculating to obtain an adjacent matrix B of the user similarity matrix:
；
；
wherein:
representing a hadamard product operation, representing an element-by-element multiplication operation;
t represents the transpose of the matrix;
s23: initializing an n rows and n columnsZero matrix of (2)；
S24: converting the adjacent matrix B into an undirected graph form, wherein the vertex of the undirected graph is the user, and the ith row or the ith column in the adjacent matrix B corresponds to the user i;
s25: vertices for user i and user uTraversing vertices->And any other vertex to obtain a plurality of closed triangle structures, wherein the closed triangle structures are +.>And +.>Calculating to obtain the closed triangle structure about +.>Higher order similarity of (2):
；
wherein:
representing a closed triangle structure->In relation to->Higher order similarity of (2);
representing vertices in undirected graphThe weights of the edges formed;
vertex of the vertexAll closed triangular structures with any other vertexThe higher-order similarity of the user i and the user u is accumulated, and the accumulated result is the higher-order similarity of the user i and the user u>And will->Update to zero matrix->And zero matrix +.>Ith row and column of (b);
s26: repeating step S25 to obtain high-order similarity between any two users, and updating the initialized zero matrix to obtain a user high-order similarity matrix. In the embodiment of the present application, the similarity greater than 1 is set to 1.
Optionally, in the step S3, constructing a user enhanced similarity matrix based on the user similarity matrix and the user higher-order similarity matrix includes:
based on user similarity matrix S and user high-order similarity matrixConstructing a user enhanced similarity matrix +.>：
；
Wherein:
the scale threshold is represented and set to 0.8.
Optionally, determining a user preference objective function representing a user tendency in the step S4 includes:
determining a user preference objective function representing user trends：
；
Wherein:
representing correction parameters, if->If not 0, then->If->0, then->；
Representing correction parameters, if->If not 0, then->If->0, then->，Representing user enhanced similarity matrix +.>Element values of the ith row and the qth column;
representing parameters to be optimally solved, U representing a user feature matrix, V representing a pay blog preference feature matrix,represents the ith row, < > in the user feature matrix U>Representing the j-th row in the pay blog preference feature matrix V;
representing a regularization term parameter, which is set to 0.2;
representing the L1 norm.
Optionally, in the step S5, the optimizing and solving the constructed user preference objective function includes:
carrying out optimization solution on the constructed user preference objective function to obtain a user feature matrix and a pay blog preference feature matrix, wherein the optimization solution flow is as follows:
s51: order theRepresenting the solution parameters to be optimized, converting the user preference objective function into +.>And initialize +.>Setting the iteration number of the current algorithm as d, the initial value of d as 0, the maximum iteration number as Max, and initializing +.>Is a unit matrix;
s52: calculation of；
S53: calculating the iteration step length of the (d+1) th iteration：
；
Wherein:
representing the trace of the calculated matrix;
s54: updating the solution parameters to be optimized based on the iteration step length:
；
wherein:
representing the identity matrix;
and updating the iteration step length:
；
；
；
if it isWill->As the result of the optimization solution, and extract the user feature matrix +.>Payment blog preference feature matrix +.>；
No order of noThe process returns to step S52.
Optionally, in the step S5, paid blog distribution processing is performed according to the preference of the user to different paid blogs, including:
user characteristic matrix obtained according to optimization solutionPayment blog preference feature matrix +.>Calculating preference values of the user i on any paid blog j:
；
wherein:
representing user i to arbitraryPreference value of pay blog j;
and pushing the paid blogs which are not purchased by the user i and have preference values larger than the preset preference threshold to the user i.
In order to solve the above-described problems, the present application provides an electronic apparatus including:
a memory storing at least one instruction;
the communication interface is used for realizing the communication of the electronic equipment; and a processor executing the instructions stored in the memory to implement the intelligent commodity distribution method for user data trends described above.
In order to solve the above-mentioned problems, the present application also provides a computer-readable storage medium having stored therein at least one instruction that is executed by a processor in an electronic device to implement the above-mentioned intelligent commodity distribution method for user data tendency.
Compared with the prior art, the application provides an intelligent commodity distribution method with user data tendency, which has the following advantages:
firstly, the scheme provides a user similarity measurement mode, and a user similarity matrix is obtained by calculation according to a user-paid blog collection matrix and behavior information of purchasing paid blogs by a user, wherein the calculation flow of the user similarity matrix is as follows: calculating time weights of different users on paid blogs:
；
；
wherein:representing a small positive number;Representing the time weights of user i and user u for pay blogs j; e, eRepresenting natural constants;Indicating the time when user i first purchased the pay blog,/->Representing the time when user u first purchased the pay blog, R represents a set time threshold,/->，；Representing a time difference between a last purchase of the paid blog and a first purchase of the paid blog by the user i; calculating the similarity between different users based on the time weight of the users to the paid blogs and the user-paid blogs collection matrix:
；
wherein:representing the similarity between user i and user j; constructing a user similarity matrix S:
；
wherein: the values on the diagonal lines in the user similarity matrix S each represent the user' S own similarity, which is set to 1. According to the scheme, a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix are respectively constructed according to the purchasing condition of a user on a paid blog and the reference relation between the purchased paid blog and the paid blog Yu Bo, and by considering the problem that the preference of the user on the paid blog changes along with time, time weight is introduced when the similarity among users is calculated, namely, the larger the ratio of the difference value between the time of purchasing the same paid blog and the time of purchasing the paid blog for the first time in the whole time is, the worse the real-time performance of the paid blog to the user is indicated, the lower the corresponding time weight is, and the similarity among users is calculated based on the time weight and the behavior information of purchasing the paid blog by the user, so that the similarity measurement of different users is realized.
Meanwhile, the scheme provides a pay blog distribution processing mode, and the user high-order similarity matrix is obtained by carrying out high-order similarity calculation on the user similarity matrix, wherein the high-order similarity calculation flow is as follows: setting the values on the diagonal lines in the user similarity matrix S to 0 to obtain a matrixThe method comprises the steps of carrying out a first treatment on the surface of the Based on matrix->Calculating to obtain an adjacent matrix B of the user similarity matrix:
；
；
wherein:representing a hadamard product operation, representing an element-by-element multiplication operation; t represents the transpose of the matrix; initializing a zero matrix of n rows and n columns>The method comprises the steps of carrying out a first treatment on the surface of the Converting the adjacent matrix B into an undirected graph form, wherein the vertex of the undirected graph is the user, and the ith row or the ith column in the adjacent matrix B corresponds to the user i; vertex for user i and user u +.>Traversing verticesAnd any other vertex to obtain a plurality of closed triangle structures, wherein the closed triangle structures are +.>And +.>Calculating to obtain the closed triangle structureHigher order similarity of (2):
；
wherein:representing a closed triangle structure->In relation to->Higher order similarity of (2);Representing vertex +_in undirected graph>The weights of the edges formed; vertex->All closed triangular structures with any other vertex are about +.>The higher-order similarity of the user i and the user u is accumulated, and the accumulated result is the higher-order similarity of the user i and the user u>And will->Update to zero matrix->And zero matrix +.>Ith row and column of (c). Determining a user preference objective function representing a user's tendency>：
；
Wherein:representing correction parameters, if->If not 0, then->If->0, then->=0；Representing correction parameters, if->If not 0, then->If->0, then->，Representing user enhanced similarity matrix +.>Element values of the ith row and the qth column;Representing parameters to be solved optimally, U representing a user feature matrix, V representing a pay blog preference feature matrix,/for>Represents the ith row, < > in the user feature matrix U>Representing the j-th row in the pay blog preference feature matrix V;Representing a regularization term parameter, which is set to 0.2;Representing the L1 norm. The scheme comprises the steps of constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix as user-paid blog preference matrices, sequentially converting a user similarity matrix into an adjacent matrix and an undirected graph, wherein a user is the vertex in the undirected graph, converting the user similarity into a high-order similarity measure combined with the weight of the adjacent user point side by analyzing the closed triangle structure relationship among the users, further constructing a user enhanced similarity matrix, obtaining a user preference objective function representing the preference tendency of similar users to the paid blog according to the user-paid blog preference matrices and the user enhanced similarity matrix, carrying out optimization solving on the constructed user preference objective function, solving to obtain a user characteristic matrix and a paid blog preference characteristic matrix, further calculating to obtain the preference values of the users to different paid blogs, and carrying out payment according to the ordering sequence of the preference valuesAnd (5) blog distribution processing.
Drawings
FIG. 1 is a flow chart of a method for intelligent commodity distribution with user data trends according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of an electronic device for implementing an intelligent commodity distribution method with user data tendency according to an embodiment of the present application.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The embodiment of the application provides an intelligent commodity distribution method with user data tendency. The execution subject of the user data-prone intelligent commodity distribution method includes, but is not limited to, at least one of a server, a terminal and the like capable of being configured to execute the method provided by the embodiment of the application. In other words, the user data-prone intelligent commodity distribution method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.
Example 1
S1: collecting behavior information of purchasing a paid blog by a user and reference relation information among the paid blogs, respectively constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, and constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix.
In the step S1, the behavior information of purchasing the paid blogs by the user and the reference relation information among the paid blogs are collected, and a user-paid blogs collection matrix and a paid Fei Boke-paid blogs reference matrix are constructed, which comprises the following steps:
collecting behavior information of purchasing a paid blog by a user, and referring relation information between the paid blog and the rest paid blogs, wherein the collected information is expressed in the following form:
；
；
wherein:
，indicating that user i purchased a paid blog j, < >>Indicating that user i did not purchase pay blog j, if +.>Then->Indicating the time when user i purchased pay blog j, if +.>Then->Is empty;
n represents the total number of selected users, m represents the total number of selected paid blogs;
，indicating that pay blog j does not reference pay blog h, < ->Representing pay blogsj refers to pay blog h, if +.>Then->；
Respectively constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, wherein the constructed user-paid blog collection matrix G is as follows:
；
the constructed paid blog-paid blog reference matrix C is:
；
wherein:
representing a reference relationship between the paid blog 1 and the paid blog m.
In the step S1, a user-paid blog preference matrix is constructed according to a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, which includes:
constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix:
；
；
wherein:
indicating the preference degree of user i for pay blogs j, < >>Representing the total number of references to pay blogs purchased by user i.
S2: and calculating to obtain a user similarity matrix according to the user-paid blog collection matrix and the behavior information of the user purchasing the paid blog, and performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix.
And step S2, calculating to obtain a user similarity matrix, wherein the step S comprises the following steps:
calculating to obtain a user similarity matrix according to the user-paid blog collection matrix and the behavior information of purchasing the paid blog by the user, wherein the calculation flow of the user similarity matrix is as follows:
calculating time weights of different users on paid blogs:
；
；
wherein:
representing a small positive number;
representing the time weights of user i and user u for pay blogs j;
e represents a natural constant;
indicating the time when user i first purchased the pay blog,/->R table indicating time of first purchase of paid blog by user uShowing the set time threshold, +.>，；
Representing a time difference between a last purchase of the paid blog and a first purchase of the paid blog by the user i;
calculating the similarity between different users based on the time weight of the users to the paid blogs and the user-paid blogs collection matrix:
；
wherein:
representing the similarity between user i and user j;
constructing a user similarity matrix S:
；
wherein:
the values on the diagonal lines in the user similarity matrix S each represent the user' S own similarity, which is set to 1.
In the step S2, performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix, including:
performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix, wherein the mode sequence analysis is a main implementation mode of the high-order similarity calculation, and the high-order similarity calculation flow is as follows:
s21: setting the values on the diagonal lines in the user similarity matrix S to 0 to obtain a matrix；
S22: matrix-basedCalculating to obtain an adjacent matrix B of the user similarity matrix:
；
；
wherein:
representing a hadamard product operation, representing an element-by-element multiplication operation;
t represents the transpose of the matrix;
s23: initializing a zero matrix of n rows and n columns；
S24: converting the adjacent matrix B into an undirected graph form, wherein the vertex of the undirected graph is the user, and the ith row or the ith column in the adjacent matrix B corresponds to the user i;
s25: vertices for user i and user uTraversing vertices->And any other vertex to obtain a plurality of closed triangle structures, wherein the closed triangle structures are +.>And +.>Calculating to obtain theClosed triangular structure about->Higher order similarity of (2):
；/>
wherein:
representing a closed triangle structure->In relation to->Higher order similarity of (2);
representing vertex +_in undirected graph>The weights of the edges formed;
vertex of the vertexAll closed triangular structures with any other vertexThe higher-order similarity of the user i and the user u is accumulated, and the accumulated result is the higher-order similarity of the user i and the user u>And will->Update to zero matrix->And zero matrix +.>Ith row and column of (b);
s26: repeating step S25 to obtain high-order similarity between any two users, and updating the initialized zero matrix to obtain a user high-order similarity matrix。
S3: and constructing a user enhancement similarity matrix according to the calculated user high-order similarity matrix and the user similarity matrix.
In the step S3, a user enhanced similarity matrix is constructed based on the user similarity matrix and the user higher-order similarity matrix, including:
based on user similarity matrix S and user high-order similarity matrixConstructing a user enhanced similarity matrix +.>：
；
Wherein:
the scale threshold is represented and set to 0.8.
S4: and determining a user preference objective function representing the user tendency according to the constructed user enhancement similarity matrix.
The step S4 of determining a user preference objective function representing a user tendency includes:
determining a user preference objective function representing user trends：
；
Wherein:
representing correction parameters, if->If not 0, then->If->0, then->；
Representing correction parameters, if->If not 0, then->If->0, then->，Representing user enhanced similarity matrix +.>Element values of the ith row and the qth column;
representing parameters to be solved optimallyThe number, U, represents the user feature matrix, V represents the pay blog preference feature matrix,represents the ith row, < > in the user feature matrix U>Representing the j-th row in the pay blog preference feature matrix V;
representing a regularization term parameter, which is set to 0.2;
representing the L1 norm. />
S5: and carrying out optimization solving on the constructed user preference objective function to obtain the preferences of the user on different paid blogs and carrying out paid blogs distribution processing.
And in the step S5, the constructed user preference objective function is optimally solved, and the method comprises the following steps:
carrying out optimization solution on the constructed user preference objective function to obtain a user feature matrix and a pay blog preference feature matrix, wherein the optimization solution flow is as follows:
s51: order theRepresenting the solution parameters to be optimized, converting the user preference objective function into +.>And initialize +.>Setting the iteration number of the current algorithm as d, the initial value of d as 0, the maximum iteration number as Max, and initializing +.>Is a unit matrix;
s52: calculation of；
S53: calculating the iteration step length of the (d+1) th iteration：
；
Wherein:
representing the trace of the calculated matrix;
s54: updating the solution parameters to be optimized based on the iteration step length:
；
wherein:
representing the identity matrix;
and updating the iteration step length:
；
；
；
if it isWill->As an optimizationSolving the result and extracting the user characteristic matrix from the solution>Payment blog preference feature matrix +.>；
No order of noThe process returns to step S52.
In the step S5, payment blog distribution processing is performed according to the preference of the user to different payment blogs, including:
user characteristic matrix obtained according to optimization solutionPayment blog preference feature matrix +.>Calculating preference values of the user i on any paid blog j:
；
wherein:
representing the preference value of the user i for any paid blog j;
and pushing the paid blogs which are not purchased by the user i and have preference values larger than the preset preference threshold to the user i.
Example 2
Fig. 2 is a schematic structural diagram of an electronic device for implementing an intelligent commodity distribution method for user data tendency according to an embodiment of the present application.
The electronic device 1 may comprise a processor 10, a memory 11, a communication interface 13 and a bus, and may further comprise a computer program, such as program 12, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may in other embodiments also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used not only for storing application software installed in the electronic device 1 and various types of data, such as codes of the program 12, but also for temporarily storing data that has been output or is to be output.
The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects respective parts of the entire electronic device using various interfaces and lines, executes or executes programs or modules (a program 12 for realizing intelligent commodity distribution, etc.) stored in the memory 11, and invokes data stored in the memory 11 to perform various functions of the electronic device 1 and process data.
The communication interface 13 may comprise a wired interface and/or a wireless interface (e.g. WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device 1 and other electronic devices and to enable connection communication between internal components of the electronic device.
The bus may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
Fig. 2 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 2 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 1 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.
The electronic device 1 may optionally further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 1 and for displaying a visual user interface.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The program 12 stored in the memory 11 of the electronic device 1 is a combination of instructions that, when executed in the processor 10, may implement:
collecting behavior information of purchasing a paid blog by a user and reference relation information among the paid blogs, respectively constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, and constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix;
calculating to obtain a user similarity matrix according to the user-paid blog collection matrix and the behavior information of the user purchasing the paid blog, and performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix;
constructing a user enhancement similarity matrix according to the calculated user high-order similarity matrix and the user similarity matrix;
determining a user preference objective function representing user tendency according to the constructed user enhancement similarity matrix;
and carrying out optimization solving on the constructed user preference objective function to obtain the preferences of the user on different paid blogs and carrying out paid blogs distribution processing.
Specifically, the specific implementation method of the above instruction by the processor 10 may refer to descriptions of related steps in the corresponding embodiments of fig. 1 to 2, which are not repeated herein.
It should be noted that, the foregoing reference numerals of the embodiments of the present application are merely for describing the embodiments, and do not represent the advantages and disadvantages of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method 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, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present application.
The foregoing description is only of the preferred embodiments of the present application, and is not intended to limit the scope of the application, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (6)
1. A method for intelligent commodity distribution of user data trends, the method comprising:
s1: collecting behavior information of purchasing a paid blog by a user and reference relation information among the paid blogs, respectively constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, and constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix;
collecting behavior information of purchasing a paid blog by a user and reference relation information among the paid blogs, and constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, comprising:
collecting behavior information of purchasing a paid blog by a user, and referring relation information between the paid blog and the rest paid blogs, wherein the collected information is expressed in the following form:
{(g ij ，t ij )|i∈[1，n]，j∈[1，m]}
{c j←h |j，h∈[1，m]，j≠h}
wherein:
g ij ＝{0,1}，g ij =1 means that user i purchased paid blogs j, g ij =0 means that user i does not purchase pay blog j, if g ij =1, then t ij Indicating the time when user i purchased paid blog j, if g ij =0, then t ij Is empty;
n represents the total number of selected users, m represents the total number of selected paid blogs;
c j←h ＝{-1,0,1}，c j←h =0 denotes that paid blog j does not reference paid blog h, c j←h =1 means that paid blog j references paid blog h, if j=h, c j←h ＝-1；
Respectively constructing a user-paid blog collection matrix and a paid Fei Boke-paid blog reference matrix, wherein the constructed user-paid blog collection matrix G is as follows:
the constructed paid blog-paid blog reference matrix C is:
wherein:
c 1←m representing a reference relationship between the paid blog 1 and the paid blog m;
constructing a user-paid blog preference matrix from the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix, comprising:
constructing a user-paid blog preference matrix according to the user-paid blog collection matrix and the paid Fei Boke-paid blog reference matrix:
wherein:
y ij indicating the preference degree of the user i for the paid blog j,representing the total number of references to pay blogs purchased by user i for pay blogs j;
s2: calculating to obtain a user similarity matrix according to the user-paid blog collection matrix and the behavior information of the user purchasing the paid blog, and performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix;
calculating to obtain a user similarity matrix, including:
calculating to obtain a user similarity matrix according to the user-paid blog collection matrix and the behavior information of purchasing the paid blog by the user, wherein the calculation flow of the user similarity matrix is as follows:
calculating time weights of different users on paid blogs:
wherein:
e represents a small positive number;
representing the time weights of user i and user u for pay blogs j;
e represents a natural constant;
R i representing the time when user i first purchases a pay blog, R u Representing the time when user u first purchased the pay blog, R represents a set time threshold, R>R i ，i∈[1,n]；
F i Representing a time difference between a last purchase of the paid blog and a first purchase of the paid blog by the user i;
calculating the similarity between different users based on the time weight of the users to the paid blogs and the user-paid blogs collection matrix:
wherein:
sim (i, u) represents the similarity between user i and user j;
constructing a user similarity matrix S:
wherein:
the values on the diagonal lines in the user similarity matrix S all represent the user similarity, and the value is set to be 1;
s3: constructing a user enhancement similarity matrix according to the calculated user high-order similarity matrix and the user similarity matrix;
s4: determining a user preference objective function representing user tendency according to the constructed user enhancement similarity matrix;
s5: and carrying out optimization solving on the constructed user preference objective function to obtain the preferences of the user on different paid blogs and carrying out paid blogs distribution processing.
2. The intelligent commodity distribution method according to claim 1, wherein in the step S2, the high-order similarity calculation is performed on the user similarity matrix to obtain the user high-order similarity matrix, and the method comprises the steps of:
performing high-order similarity calculation on the user similarity matrix to obtain a user high-order similarity matrix, wherein the high-order similarity calculation flow is as follows:
s21: setting the values on the diagonal lines in the user similarity matrix S to 0 to obtain the matrix S 0 ；
S22: based on matrix S 0 Calculating to obtain an adjacent matrix B of the user similarity matrix:
B＝(A·A)⊙A
A＝S 0 ⊙S 0 T
wherein:
the ". Alt represents Hadamard product operation, represents element-wise multiplication operation;
t represents the transpose of the matrix;
s23: initializing a zero matrix D of n rows and n columns 0 ；
S24: converting the adjacent matrix B into an undirected graph form, wherein the vertex of the undirected graph is the user, and the ith row or the ith column in the adjacent matrix B corresponds to the user i;
s25: vertex codes for user i and user u i ,code u Traversing vertex codes i ,code u And any other vertex to obtain a plurality of closed triangle structures, and for any closed triangle structure code i ,code u Code z Calculating to obtain the code of the closed triangle structure i ,code u Higher order similarity of (2):
p(code i ,code u ；code 2 )＝w(code i ,code z )×w(code u ,code 2 )
wherein:
p(code i ,code u ；code z ) Code representing closed triangle structure i ,code u ,code z In relation to code i ,code u Higher order similarity of (2);
w(code i ,code z ) Representing vertex codes in undirected graph i ,code z Is composed ofWeighting of edges;
encode the vertex i ,code u Code for all closed triangle structures composed with any other vertex i ,code u The higher-order similarity of the user i and the user u is accumulated, and the accumulated result is the higher-order similarity S 'of the user i and the user u' iu And S 'is carried out' iu Updating to zero matrix D 0 Ith row and ith column of (c) and zero matrix D 0 Ith row and column of (b);
s26: and repeating the step S25 to obtain the high-order similarity between any two users, and updating the initialized zero matrix to obtain a user high-order similarity matrix S'.
3. The intelligent commodity distribution method according to claim 2, wherein in the step S3, a user enhanced similarity matrix is constructed based on the user similarity matrix and the user higher-order similarity matrix, comprising:
constructing a user enhanced similarity matrix S 'based on the user similarity matrix S and the user high-order similarity matrix S':
S″＝mS+(1-α)S′
wherein:
alpha represents a proportional threshold, which is set to 0.8.
4. The intelligent commodity distribution method according to claim 1, wherein said step S4 of determining a user preference objective function indicative of user preferences comprises:
determining a user preference objective function f (U, V) representing user preferences:
wherein:
β ij representing the correction parameters, if y ij Is not 0, beta ij =1, if y ij 0, beta ij ＝0；
γ iq Indicating the correction parameters, if S iq Not 0, then gamma ik =1, if S "" iq Is 0, then gamma iq ＝0，S″ iq Element values representing the ith row and the qth column in the user enhanced similarity matrix S ";
u, V represents the parameters to be optimally solved, U represents the user feature matrix, V represents the pay blog preference feature matrix, U i Representing the ith row, V, in the user feature matrix U j Representing the j-th row in the pay blog preference feature matrix V;
sigma represents a regularization term parameter, which is set to 0.2;
|·| denotes the L1 norm.
5. The intelligent commodity distribution method according to claim 4, wherein said step S5 of optimally solving the constructed user preference objective function comprises:
carrying out optimization solution on the constructed user preference objective function to obtain a user feature matrix and a pay blog preference feature matrix, wherein the optimization solution flow is as follows:
s51: let θ= (U, V) represent the solution parameters to be optimized, convert the user preference objective function to f (θ), and initialize θ 0 Setting the iteration number of the current algorithm as d, the initial value of d as 0, the maximum iteration number as Max, and initializing L 0 Is a unit matrix;
s52: calculation of
S53: calculating the iteration step lambda at the (d+1) th iteration d ：
λ d ＝||grad(θ d )||tr(L d )
Wherein:
tr (·) represents the trace of the calculation matrix;
s54: updating the solution parameters to be optimized based on the iteration step length:
θ d+1 ＝θ d -(λ d I+L d )
wherein:
i represents an identity matrix;
and updating the iteration step length:
a d ＝f(θ d+1 )-f(θ d )
b d ＝grad(θ d+1 )-grad(θ d )
if d+1 is greater than or equal to Max, then θ d+1 As the optimized solving result, extracting the user characteristic matrix U * Payment blog preference feature matrix V * ；
If not, d=d+1, and the process returns to step S52.
6. An intelligent commodity distribution method according to claim 5, wherein in step S5, payment blogging distribution processing is performed according to user preferences of different payment blogs, including:
user characteristic matrix U obtained according to optimization solution * Payment blog preference feature matrix V * Calculating preference values of the user i on any paid blog j:
wherein:
h ij representing the preference value of the user i for any paid blog j;
and pushing the paid blogs which are not purchased by the user i and have preference values larger than the preset preference threshold to the user i.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310510872.1A CN116228282B (en) | 2023-05-09 | 2023-05-09 | Intelligent commodity distribution method for user data tendency |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310510872.1A CN116228282B (en) | 2023-05-09 | 2023-05-09 | Intelligent commodity distribution method for user data tendency |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116228282A CN116228282A (en) | 2023-06-06 |
| CN116228282B true CN116228282B (en) | 2023-08-11 |
Family
ID=86569878
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310510872.1A Active CN116228282B (en) | 2023-05-09 | 2023-05-09 | Intelligent commodity distribution method for user data tendency |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116228282B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105843860A (en) * | 2016-03-17 | 2016-08-10 | 山东大学 | Microblog attention recommendation method based on parallel item-based collaborative filtering algorithm |
| CN106776881A (en) * | 2016-11-28 | 2017-05-31 | 中国科学院软件研究所 | A kind of realm information commending system and method based on microblog |
| CN109241203A (en) * | 2018-09-27 | 2019-01-18 | 天津理工大学 | A kind of user preference and distance weighted clustering method of time of fusion factor |
| CN110110203A (en) * | 2018-01-11 | 2019-08-09 | 腾讯科技（深圳）有限公司 | Resource information method for pushing and server, resource information methods of exhibiting and terminal |
| CN110263189A (en) * | 2019-06-24 | 2019-09-20 | 腾讯科技（深圳）有限公司 | Recommended method, device, storage medium and the computer equipment of media content |
| CN111931023A (en) * | 2020-07-01 | 2020-11-13 | 西北工业大学 | Community structure identification method and device based on network embedding |
| CN113158041A (en) * | 2021-04-19 | 2021-07-23 | 电子科技大学 | Article recommendation method based on multi-attribute features |
| CN114461899A (en) * | 2021-12-24 | 2022-05-10 | 新奥新智科技有限公司 | Collaborative filtering recommendation method and device for user, electronic equipment and storage medium |
| CN115357803A (en) * | 2021-05-17 | 2022-11-18 | 连普乐士株式会社 | Account recommendation method, computer device and computer program |
| CN115545349A (en) * | 2022-11-24 | 2022-12-30 | 天津师范大学 | Time sequence social media popularity prediction method and device based on attribute sensitive interaction |
| CN116028727A (en) * | 2023-03-30 | 2023-04-28 | 南京邮电大学 | Video recommendation method based on image data processing |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009048937A1 (en) * | 2007-10-08 | 2009-04-16 | Imedia Streams, Llc | Method and system for integrating rankings of journaled internet content and consumer media preferences for use in marketing profiles |
| JP2011253462A (en) * | 2010-06-03 | 2011-12-15 | Sony Corp | Contents recommendation system, contents recommendation device and contents recommendation method |
| US10515127B2 (en) * | 2015-04-09 | 2019-12-24 | Oath Inc. | Inductive matrix completion and graph proximity for content item recommendation |
| US20180032882A1 (en) * | 2016-07-27 | 2018-02-01 | Fuji Xerox Co., Ltd. | Method and system for generating recommendations based on visual data and associated tags |
-
2023
- 2023-05-09 CN CN202310510872.1A patent/CN116228282B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105843860A (en) * | 2016-03-17 | 2016-08-10 | 山东大学 | Microblog attention recommendation method based on parallel item-based collaborative filtering algorithm |
| CN106776881A (en) * | 2016-11-28 | 2017-05-31 | 中国科学院软件研究所 | A kind of realm information commending system and method based on microblog |
| CN110110203A (en) * | 2018-01-11 | 2019-08-09 | 腾讯科技（深圳）有限公司 | Resource information method for pushing and server, resource information methods of exhibiting and terminal |
| CN109241203A (en) * | 2018-09-27 | 2019-01-18 | 天津理工大学 | A kind of user preference and distance weighted clustering method of time of fusion factor |
| CN110263189A (en) * | 2019-06-24 | 2019-09-20 | 腾讯科技（深圳）有限公司 | Recommended method, device, storage medium and the computer equipment of media content |
| CN111931023A (en) * | 2020-07-01 | 2020-11-13 | 西北工业大学 | Community structure identification method and device based on network embedding |
| CN113158041A (en) * | 2021-04-19 | 2021-07-23 | 电子科技大学 | Article recommendation method based on multi-attribute features |
| CN115357803A (en) * | 2021-05-17 | 2022-11-18 | 连普乐士株式会社 | Account recommendation method, computer device and computer program |
| CN114461899A (en) * | 2021-12-24 | 2022-05-10 | 新奥新智科技有限公司 | Collaborative filtering recommendation method and device for user, electronic equipment and storage medium |
| CN115545349A (en) * | 2022-11-24 | 2022-12-30 | 天津师范大学 | Time sequence social media popularity prediction method and device based on attribute sensitive interaction |
| CN116028727A (en) * | 2023-03-30 | 2023-04-28 | 南京邮电大学 | Video recommendation method based on image data processing |
Non-Patent Citations (1)
| Title |
|---|
| 融合用户画像与协同过滤的知识付费平台个性化推荐模型;魏玲;情报理论与实践;第44卷(第03期);第188-193页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116228282A (en) | 2023-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210150378A1 (en) | Recommendation method, recommendation apparatus, recommendation device, recommendation system and storage medium | |
| CN113449187B (en) | Product recommendation method, device, equipment and storage medium based on double images | |
| CN107220217A (en) | Characteristic coefficient training method and device that logic-based is returned | |
| CN111652278B (en) | User behavior detection method, device, electronic equipment and medium | |
| CN108074003B (en) | Prediction information pushing method and device | |
| CN109213936B (en) | Commodity searching method and device | |
| CN107274209A (en) | The method and apparatus for predicting advertising campaign sales data | |
| CN113837631B (en) | Employee evaluation method and device, electronic equipment and readable storage medium | |
| CN111966886A (en) | Object recommendation method, object recommendation device, electronic equipment and storage medium | |
| CN115391669A (en) | Intelligent recommendation method and device and electronic equipment | |
| CN114912948A (en) | Cloud service-based cross-border e-commerce big data intelligent processing method, device and equipment | |
| CN111652282B (en) | Big data-based user preference analysis method and device and electronic equipment | |
| CN107346344A (en) | The method and apparatus of text matches | |
| CN113723747A (en) | Analysis report generation method, electronic device and readable storage medium | |
| CN112328869A (en) | User loan willingness prediction method and device and computer system | |
| CN112487021A (en) | Correlation analysis method, device and equipment for business data | |
| CN110866625A (en) | Promotion index information generation method and device | |
| CN107729931A (en) | Picture methods of marking and device | |
| CN115048174A (en) | User-based personalized menu generation method, device, equipment and storage medium | |
| CN106157281A (en) | A kind of image subject recognition methods and device | |
| CN116228282B (en) | Intelligent commodity distribution method for user data tendency | |
| CN113778979A (en) | Method and device for determining live broadcast click rate | |
| CN114625975B (en) | Knowledge graph-based customer behavior analysis system | |
| CN113139129B (en) | Virtual reading trajectory graph generation method and device, electronic equipment and storage medium | |
| CN113705686B (en) | Image classification method, device, electronic equipment and readable storage medium |
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 |