CN113362098A - Data processing method, device and computer readable storage medium - Google Patents
Data processing method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN113362098A CN113362098A CN202010152306.4A CN202010152306A CN113362098A CN 113362098 A CN113362098 A CN 113362098A CN 202010152306 A CN202010152306 A CN 202010152306A CN 113362098 A CN113362098 A CN 113362098A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- target commodity
- time
- weight
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 28
- 238000013528 artificial neural network Methods 0.000 claims abstract description 42
- 238000012545 processing Methods 0.000 claims abstract description 32
- 230000003993 interaction Effects 0.000 claims abstract description 18
- 239000013598 vector Substances 0.000 claims description 24
- 238000011176 pooling Methods 0.000 claims description 15
- 238000000034 method Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000004590 computer program Methods 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 4
- 238000007477 logistic regression Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0254—Targeted advertisements based on statistics
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure provides a data processing method, a data processing device and a computer readable storage medium, and relates to the technical field of computers. The data processing method comprises the following steps: acquiring equipment interaction data and operation result data when a user performs online shopping operation on a target commodity each time; generating an equipment interaction weight by using the equipment interaction data, and performing weighted summation on the operation result data by using the equipment interaction weight to obtain operation result statistical data of the user about the target commodity; generating operation characteristics of the user about the target commodities by using the operation result statistical data, and generating historical operation characteristics of the user by using the operation characteristics of the user about each target commodity; and inputting the historical operation characteristics into a pre-trained neural network, obtaining the probability of the user performing online shopping operation on each target commodity next time, and pushing relevant data of each target commodity to the user according to the probability. The data pushing accuracy and the user experience of the data pushing service can be improved.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a data processing method and apparatus, and a computer-readable storage medium.
Background
With the development of electronic commerce, the e-commerce platform gradually accumulates massive operation result data, which includes browsing data, ordering data, collecting data, shopping cart adding data, and the like when a user performs an online shopping operation on a commodity.
On the other hand, the e-commerce platform also expands data push services. In the data push service, the e-commerce platform can push data related to commodities, such as commodity links, commodity activity information and the like, to a user. If the data pushed to the user by the data pushing service is the related data of the commodity concerned by the user, the user experience of the data pushing service is greatly improved. Therefore, how to improve the data pushing accuracy and the user experience of the data pushing service by using the operation result data accumulated by the e-commerce platform is becoming a popular research topic.
Disclosure of Invention
The technical problem solved by the present disclosure is how to improve the data push accuracy and user experience of the data push service.
According to an aspect of the embodiments of the present disclosure, there is provided a data processing method, including: acquiring equipment interaction data and operation result data when a user performs online shopping operation on a target commodity each time; generating a device interaction weight when a user performs online shopping operation on a target commodity by using the device interaction data, and performing weighted summation on operation result data by using the device interaction weight to obtain operation result statistical data of the user about the target commodity; generating operation characteristics of the user about the target commodities by using the operation result statistical data, and generating historical operation characteristics of the user by using the operation characteristics of the user about each target commodity; and inputting the historical operation characteristics into a pre-trained neural network, obtaining the probability of the user performing online shopping operation on each target commodity next time, and pushing relevant data of each target commodity to the user according to the probability.
In some embodiments, the operation result data includes the number of times the user browses the target commodity, and the device interaction data has different types; the method for generating the device interaction weight when the user performs online shopping operation on the target commodity by using the device interaction data comprises the following steps: respectively generating the times weight of browsing the target commodity by the corresponding type of users by utilizing different types of equipment interaction data; and generating a comprehensive times weight of each time the user browses the target commodity by using the times weights of at least one type so as to weight the times of the user browses the target commodity.
In some embodiments, the generating, by using different types of device interaction data, weights of the number of times that the corresponding types of users browse the target product each time includes: and normalizing the strength data of the target commodity link clicked on the equipment by the user each time to obtain the strength weight of the target commodity link clicked on the equipment by the user each time.
In some embodiments, the generating, by using different types of device interaction data, weights of the number of times that the corresponding types of users browse the target product each time includes: and normalizing the time length data of the target commodity browsed on the equipment by the user every time to obtain the time length weight of the target commodity browsed on the equipment by the user every time.
In some embodiments, the generating, by using different types of device interaction data, weights of the number of times that the corresponding types of users browse the target product each time includes: determining location data for a most frequent click location of a user on a device; determining the distance between the position of clicking the target commodity link on the equipment by the user each time and the most frequent clicking position on the equipment by the user according to the position data of clicking the target commodity link on the equipment by the user each time and the position data of the most frequent clicking position on the equipment by the user; and carrying out normalization processing after taking the reciprocal of the distance to obtain the position weight of the target commodity browsed on the equipment by the user each time.
In some embodiments, the operation result data includes the number of times the user places an order for the target commodity, and the device interaction data includes force data of the user clicking an order placing icon of the target commodity on the device; the method for generating the device interaction weight when the user performs online shopping operation on the target commodity by using the device interaction data comprises the following steps: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the order placing times of the target commodity by the user.
In some embodiments, the operation result data comprises a quota for ordering the target commodity by the user, and the equipment interaction data comprises strength data for clicking an ordering icon of the target commodity on the equipment by the user; the method for generating the device interaction weight when the user performs online shopping operation on the target commodity by using the device interaction data comprises the following steps: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the amount of the order placing of the target commodity by the user.
In some embodiments, the operation result data includes a number of times that the user added the target item to the shopping cart, and the device interaction data includes force data of the user clicking a shopping cart icon of the target item on the device; the method for generating the device interaction weight when the user performs online shopping operation on the target commodity by using the device interaction data comprises the following steps: and normalizing the force data of the shopping cart icon of the target commodity clicked on the equipment by the user each time to obtain the force weight of the shopping cart icon of the target commodity clicked on the equipment by the user each time so as to weight the times of adding the target commodity into the shopping cart by the user.
In some embodiments, the operation result data includes times of collecting the target commodity by the user, and the device interaction data includes force data of clicking a collection icon of the target commodity on the device by the user; the method for generating the device interaction weight when the user performs online shopping operation on the target commodity by using the device interaction data comprises the following steps: and normalizing the strength data of clicking the collection icon of the target commodity on the equipment by the user each time to obtain the strength weight of clicking the collection icon of the target commodity on the equipment by the user each time so as to weight the times of the target commodity collection by the user.
In some embodiments, the data processing method further comprises: training the neural network by using the historical operation sample characteristics of the user and the sample commodity label of the next online shopping operation executed by the user, so that the neural network can obtain the probability of the next online shopping operation executed by the user on each target commodity according to the historical operation characteristics.
In some embodiments, the neural network includes a first fully-connected layer including a plurality of fully-connected layers and a first cascade layer for cascading feature vectors output by the plurality of fully-connected layers.
In some embodiments, the neural network further comprises a discard layer, a second fully connected layer, a global average pooling layer, and a second cascade layer; the disposal layer is used for: performing discarding operation on the feature vectors output by the first cascade layer, and respectively inputting the feature vectors output after the discarding operation into the second full-connection layer and the global average pooling layer; the second cascaded layer is to: and cascading the feature vectors output by the second full-connection layer, the global average pooling layer and the first cascading layer, and carrying out batch standardization processing on the feature vectors after cascading.
According to another aspect of the embodiments of the present disclosure, there is provided a data processing apparatus including: a data acquisition module configured to: acquiring equipment interaction data and operation result data when a user performs online shopping operation on a target commodity each time; a data weighting module configured to: generating a device interaction weight when a user performs online shopping operation on a target commodity by using the device interaction data, and performing weighted summation on operation result data by using the device interaction weight to obtain operation result statistical data of the user about the target commodity; a feature generation module configured to: generating operation characteristics of the user about the target commodities by using the operation result statistical data, and generating historical operation characteristics of the user by using the operation characteristics of the user about each target commodity; a data push module configured to: and inputting the historical operation characteristics into a pre-trained neural network, obtaining the probability of the user performing online shopping operation on each target commodity next time, and pushing relevant data of each target commodity to the user according to the probability.
In some embodiments, the operation result data includes the number of times the user browses the target commodity, and the device interaction data has different types; the data weighting module is configured to: respectively generating the times weight of browsing the target commodity by the corresponding type of users by utilizing different types of equipment interaction data; and generating a comprehensive times weight of each time the user browses the target commodity by using the times weights of at least one type so as to weight the times of the user browses the target commodity.
In some embodiments, the data weighting module is configured to: and normalizing the strength data of the target commodity link clicked on the equipment by the user each time to obtain the strength weight of the target commodity link clicked on the equipment by the user each time.
In some embodiments, the data weighting module is configured to: and normalizing the time length data of the target commodity browsed on the equipment by the user every time to obtain the time length weight of the target commodity browsed on the equipment by the user every time.
In some embodiments, the data weighting module is configured to: determining location data for a most frequent click location of a user on a device; determining the distance between the position of clicking the target commodity link on the equipment by the user each time and the most frequent clicking position on the equipment by the user according to the position data of clicking the target commodity link on the equipment by the user each time and the position data of the most frequent clicking position on the equipment by the user; and carrying out normalization processing after taking the reciprocal of the distance to obtain the position weight of the target commodity browsed on the equipment by the user each time.
In some embodiments, the operation result data includes the number of times the user places an order for the target commodity, and the device interaction data includes force data of the user clicking an order placing icon of the target commodity on the device; the data weighting module is configured to: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the order placing times of the target commodity by the user.
In some embodiments, the operation result data comprises a quota for ordering the target commodity by the user, and the equipment interaction data comprises strength data for clicking an ordering icon of the target commodity on the equipment by the user; the data weighting module is configured to: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the amount of the order placing of the target commodity by the user.
In some embodiments, the operation result data includes a number of times that the user added the target item to the shopping cart, and the device interaction data includes force data of the user clicking a shopping cart icon of the target item on the device; the data weighting module is configured to: and normalizing the force data of the shopping cart icon of the target commodity clicked on the equipment by the user each time to obtain the force weight of the shopping cart icon of the target commodity clicked on the equipment by the user each time so as to weight the times of adding the target commodity into the shopping cart by the user.
In some embodiments, the operation result data includes times of collecting the target commodity by the user, and the device interaction data includes force data of clicking a collection icon of the target commodity on the device by the user; the data weighting module is configured to: and normalizing the strength data of clicking the collection icon of the target commodity on the equipment by the user each time to obtain the strength weight of clicking the collection icon of the target commodity on the equipment by the user each time so as to weight the times of the target commodity collection by the user.
In some embodiments, the data processing apparatus further comprises a network training module configured to: training the neural network by using the historical operation sample characteristics of the user and the sample commodity label of the next online shopping operation executed by the user, so that the neural network can obtain the probability of the next online shopping operation executed by the user on each target commodity according to the historical operation characteristics.
In some embodiments, the neural network includes a first fully-connected layer including a plurality of fully-connected layers and a first cascade layer for cascading feature vectors output by the plurality of fully-connected layers.
In some embodiments, the neural network further comprises a discard layer, a second fully connected layer, a global average pooling layer, and a second cascade layer; the disposal layer is used for: performing discarding operation on the feature vectors output by the first cascade layer, and respectively inputting the feature vectors output after the discarding operation into the second full-connection layer and the global average pooling layer; the second cascaded layer is to: and cascading the feature vectors output by the second full-connection layer, the global average pooling layer and the first cascading layer, and carrying out batch standardization processing on the feature vectors after cascading.
According to still another aspect of the embodiments of the present disclosure, there is provided a data processing apparatus including: a memory; and a processor coupled to the memory, the processor configured to perform the aforementioned data processing method based on instructions stored in the memory.
According to still another aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the instructions, when executed by a processor, implement the aforementioned data processing method.
According to the online shopping operation prediction method and device, the device interaction data of the user when the user performs online shopping operation on the commodities are utilized, the operation result data are weighted to obtain the operation result statistical data, so that the probability that the user performs online shopping operation on each commodity next time is predicted more accurately, and the data pushing accuracy and the user experience of the data pushing service are improved.
Other features of the present disclosure and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained according to the drawings without inventive exercise.
Fig. 1 shows a flow diagram of a data processing method of some embodiments of the present disclosure.
Fig. 2 illustrates a schematic structural diagram of a neural network of some embodiments of the present disclosure.
Fig. 3 shows a schematic structural diagram of a data processing apparatus according to some embodiments of the present disclosure.
Fig. 4 shows a schematic structural diagram of a data processing apparatus according to further embodiments of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
The inventor has found that, with the development of mobile internet, the situation that people use e-commerce applications installed on mobile terminals and the like to perform online shopping is increasing. Therefore, the interest points of the user on the commodities can be acquired by combining the device interaction data when the user carries out online shopping operation.
The process of a user performing an online shopping operation on a device installed with an e-commerce application can be broken down as follows. The user enters a shopping e-commerce application on the device, searches for a target commodity interested by the user, clicks a target commodity link to enter a commodity detail page, slides a screen to browse the target commodity, clicks a collection icon on the screen to collect the target commodity, clicks a shopping cart icon on the screen to add the target commodity into a shopping cart, and clicks a list placing icon on the screen to place a list of the target commodity. In the process, when the user slides the screen to browse the target commodity, the screen can be slid slowly for the commodity which is more interested, the screen can be slid quickly for the commodity which is not interested, and the speed of the screen can be reflected by the time consumed by the user to browse the target commodity. Meanwhile, the greater the strength of the user clicking the link or icon on the screen, the higher the interest degree of the user in the target commodity; the closer the user clicks the location of the item link on the screen to the user's most frequently clicked location, the higher the user's interest level in the target item.
In view of the above, the present disclosure provides a data processing method, which digitizes device interaction behavior when a user performs online shopping operations on a device, and introduces device interaction data to generate historical operation characteristics of the user, so as to predict a probability that the user performs online shopping operations on each target product next time, so that the user pushes related data of each target product to the user according to the probability when retrieving products.
Some embodiments of the disclosed data processing method are first described in conjunction with fig. 1.
Fig. 1 shows a flow diagram of a data processing method of some embodiments of the present disclosure. As shown in fig. 1, the present embodiment includes steps S101 to S105.
In step S101, device interaction data and operation result data each time the user performs an online shopping operation on a target commodity are acquired.
The device interaction data can be divided into different types such as click force type, browsing duration type, click position type and the like. The various types of device interaction data can be captured by calling the application programming interface of the device. The click strength equipment interaction data specifically comprises strength data of clicking a target commodity link on equipment by a user each time, strength data of clicking an order-placing icon of a target commodity on the equipment by the user, and strength data of clicking the order-placing icon of the target commodity on the equipment by the user; the browsing duration class device interaction data specifically may include duration data of each time the user browses the target commodity on the device; the click location-based device interaction data may specifically include location data of a target commodity link clicked on the device each time by the user.
The operation result data can comprise the times of browsing the target commodity by the user, the times of placing an order for the target commodity by the user, the amount of placing the order for the target commodity by the user, the times of adding the target commodity into a shopping cart by the user and the times of collecting the target commodity by the user. The operation result data can be obtained statistically on the database side of the e-commerce platform.
In step S102, a device interaction weight when the user performs an online shopping operation on the target product each time is generated using the device interaction data, and the operation result data is subjected to weighted summation using the device interaction weight, so as to obtain operation result statistical data of the user about the target product.
The steps can be specifically divided into the following situations:
(1) and generating a comprehensive frequency weight of the target commodity browsed by the user each time by using different types of equipment interactive data, and performing weighted summation on the frequency of the target commodity browsed by the user by using the comprehensive frequency weight to obtain frequency statistical data of the target commodity browsed by the user.
(2) And generating a strength weight of each time when the user clicks the order placing icon of the target commodity by using the strength data of the order placing icon of the target commodity clicked by the user on the equipment, and performing weighted summation on the order placing times of the target commodity by the user by using the strength weight to obtain the statistical data of the order placing times of the target commodity by the user.
(3) And generating a strength weight of each time the user clicks the order placing icon of the target commodity by using the strength data of the order placing icon of the target commodity clicked by the user on the equipment, and carrying out weighted summation on the amount of the order placing of the target commodity by the user by using the strength weight to obtain the amount statistical data of the order placing of the target commodity by the user.
(4) And generating a strength weight of each time the user clicks the shopping cart icon of the target commodity by using the strength data of the shopping cart icon of the target commodity clicked by the user on the equipment, and performing weighted summation on the times of adding the target commodity into the shopping cart by the user by using the strength weight to obtain the statistical data of the times of adding the target commodity into the shopping cart by the user.
(5) And generating a strength weight of each time the user clicks the order placing icon of the target commodity by using the strength data of the order placing icon of the target commodity clicked by the user on the equipment, and carrying out weighted summation on the amount of the order placing of the target commodity by the user by using the strength weight to obtain the statistical data of the times of the target commodity collected by the user.
In step S103, the operation feature of the user with respect to the target product is generated using the operation result statistical data.
The operation result statistical data may include at least one of a number of times that the user browses the target commodity, a number of times that the user places an order for the target commodity, an amount statistical data that the user places an order for the target commodity, a number of times that the user adds the target commodity to a shopping cart, and a number of times that the user collects the target commodity. For example, if the user browses the target product 10 times, places an order 2 times, places an order amount of 1000 yuan, adds a shopping cart 3 times, and collects the item 2 times in one month, the operation characteristics of the user with respect to the target product are (10, 2, 1000, 3, 2).
In step S104, the user' S historical operation characteristics are generated using the operation characteristics of the user with respect to each target product.
The operation features of the user about each target commodity are used as row vectors, historical operation features of the user can be generated according to a preset target commodity sequence, the data form of the historical operation features is a matrix, and each row of the matrix represents the operation features of the user about one target commodity.
In step S105, the historical operation features are input into a pre-trained neural network, the probability of the user performing the online shopping operation on each target commodity next time is obtained, and the relevant data of each target commodity is pushed to the user according to the probability.
The neural network needs to be pre-trained before prediction can be performed using the neural network. During training, the neural network is trained by utilizing the historical operation sample characteristics of the user and the sample commodity label of the next online shopping operation executed by the user, so that the neural network can obtain the probability of the next online shopping operation executed by the user on each target commodity according to the historical operation characteristics.
After obtaining the probability of the user performing the online shopping operation on each target commodity next time, K target commodities with the highest probability can be taken, and relevant data of the K target commodities is pushed to the user, where K represents a preset positive integer according to business needs.
As the user corresponds to the historical operation characteristics in the form of a matrix, the historical operation characteristics in the form of the matrix corresponding to a plurality of users can be input into the neural network in batch when the neural network is trained and used, so as to realize the batch processing operation of the neural network on data.
According to the embodiment, the operation result data is weighted by using the device interaction data of the user when the user performs online shopping operation on the commodities to obtain the operation result statistical data, so that the probability of the user performing online shopping operation on each commodity next time is more accurately predicted, and the data push accuracy and the user experience of the data push service are improved.
The process of generating device interaction weights and weighting the operation result data is described in detail in the following cases.
Weighting the times of browsing target goods by user
And respectively generating the times of the target commodity browsing of the corresponding type of users by utilizing different types of equipment interaction data. Specifically, the following three cases can be classified.
(1) And normalizing the strength data of the target commodity link clicked on the equipment by the user each time to obtain the strength weight of the target commodity link clicked on the equipment by the user each time.
(2) And normalizing the time length data of the target commodity browsed on the equipment by the user every time to obtain the time length weight of the target commodity browsed on the equipment by the user every time.
(3) Determining location data for a most frequent click location of a user on a device; determining the distance between the position of clicking the target commodity link on the equipment by the user each time and the most frequent clicking position on the equipment by the user according to the position data of clicking the target commodity link on the equipment by the user each time and the position data of the most frequent clicking position on the equipment by the user; and carrying out normalization processing after taking the reciprocal of the distance to obtain the position weight of the target commodity browsed on the equipment by the user each time.
The three cases (1), (2) and (3) are described in further detail by taking the most complicated case (3) as an example.
Assume that the location where the user clicks on the target merchandise link on the device at a certain time is D (x)1,y1). When the most frequently clicked position of the user on the equipment is determined, the mobile phone screen can be divided into 7 rows and 3 columns of square grid areas, and for each square grid R, if the historical position D of the target commodity link clicked on the equipment by the user falls on a certain position of the square grid R, D belongs to R. Then, all the squares are counted, and the square is considered to be a square which the user has habitually clicked by observing which square has the largest number of clicks, and the position of the center point of the square is G (x)2,y2). Calculating D (x)1,y1) And G (x)2,y2) European distance betweenThen taking reciprocal to obtain w1And for the reciprocal value w1Performing a linear transformationThe position weight W of the target commodity browsed on the equipment by the user at the time can be obtained, and W in the formulaminRepresents the minimum value after taking the reciprocal, wmaxThe maximum value after the reciprocal is shown.
By using the times weight of at least one type of the three cases (1), (2) and (3), the comprehensive times weight of each time the user browses the target product can be generated so as to weight the times of the user browsing the target product.
The optimal case is to generate the integrated times weight using the weights of the three cases. It is assumed that after the time sequence [1,0,1,1 … ] of browsing the target product by the user is obtained, the time sequence is not simply summed, but a comprehensive time weight of browsing the target product by the user each time is introduced, and then the sequence after the comprehensive time weight is introduced is summed. The sequences after the integrated number of times weight was introduced were [1 × 0.3 × 0.1 × 0.5,0 × 0.1 × 0.2 × 0.11,1 × 0.2 × 0.23 × 0.21,1 × 0.12 × 0.34 × 0.12 … ], where the first view of the target item was a strength weight of 0.3, a duration weight of 0.1, a position weight of 0.5, and an integrated number of times weight of 0.3 × 0.1 × 0.5.
Through multilayer weight superposition, the influence of interactive operation of the user on different layers on statistical data is introduced, and feature extraction is performed on the basis, so that the probability of the user performing online shopping operation on each commodity next time can be more accurately predicted.
(II) weighting the times of ordering the target commodity by the user
And normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the order placing times of the target commodity by the user.
(III) weighting the times of ordering the target commodity by the user
And normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the amount of the order placing of the target commodity by the user.
(IV) weighting the times of adding the target commodity into the shopping cart by the user
And normalizing the force data of the shopping cart icon of the target commodity clicked on the equipment by the user each time to obtain the force weight of the shopping cart icon of the target commodity clicked on the equipment by the user each time so as to weight the times of adding the target commodity into the shopping cart by the user.
And (V) normalizing the force data of clicking the collection icon of the target commodity on the equipment by the user each time, and obtaining the force weight of clicking the collection icon of the target commodity on the equipment by the user each time so as to weight the times of the target commodity collection by the user.
The cases (two) to (five) are simpler than the case (one), and those skilled in the art can implement the operations in the cases (two) to (five) according to the case (one), and will not be described herein too much.
The structure of the neural network according to the present disclosure is described in detail below with reference to fig. 2.
Fig. 2 illustrates a schematic structural diagram of a neural network of some embodiments of the present disclosure. As shown in fig. 2, the neural network includes a first fully-connected layer and a first cascade layer, the first fully-connected layer includes a plurality of fully-connected layers, and the first cascade layer is configured to cascade feature vectors output by the plurality of fully-connected layers.
In some embodiments, the neural network further includes a discard layer Dropout, a second fully connected layer, a global average pooling layer, and a second cascaded layer. Wherein the disposal layer is for: performing discarding operation on the feature vectors output by the first cascade layer, and respectively inputting the feature vectors output after the discarding operation into the second full-connection layer and the global average pooling layer; the second cascaded layer is to: and cascading the eigenvectors output by the second full-connection layer, the global average pooling layer and the first cascading layer, and carrying out Batch standardization processing on the cascaded eigenvectors.
In some embodiments, the neural network further comprises a third fully-connected layer and a logistic regression layer.
The above-described embodiments use a multi-tiered fully-connected network to non-linearly map the historical operating characteristics of the user. The input is the historical operation characteristics of the user processed according to the weight, then the first full-connection layer carries out characteristic mapping, and Dropout operation and Batch Normalization operation are added in the middle. Since the Dropout operation and the Batch Normalization operation close part of the neurons and open part of the neurons, overfitting can be prevented to improve the robustness of the neural network. Meanwhile, by using the idea of residual error network for reference, cascade concatenate operation is carried out, so that the original information of the neural network can not be lost along with the deepening of the number of neural layers. And finally, predicting the probability of outputting each target commodity through logistic regression softmax operation in the logistic regression layer.
The loss function of the neural network may be set toWherein c is the commodity identification, M is the total number of commodities, pcPredicting the probability of the next online shopping operation performed on the commodity c by the user for the neural network, and if the online shopping operation is actually performed on the commodity c by the user for the next time, ycGet 1, y if the user does not actually perform an online shopping operation on the item c the next timecTake 0. In the process of training the neural network, an adam optimizer can be used, and the training step length is dynamically adjusted along with the training process, so that the loss function is optimized, and the loss function value is reduced. After training is finished, 20% of training data sets can be divided to test and verify the neural network, and prediction accuracy is used as an evaluation index.
Some embodiments of the disclosed data processing apparatus are described below in conjunction with fig. 3.
Fig. 3 shows a schematic structural diagram of a data processing apparatus according to some embodiments of the present disclosure. As shown in fig. 3, the data processing apparatus 30 in the present embodiment includes: a data acquisition module 301 configured to: acquiring equipment interaction data and operation result data when a user performs online shopping operation on a target commodity each time; a data weighting module 302 configured to: generating a device interaction weight when a user performs online shopping operation on a target commodity by using the device interaction data, and performing weighted summation on operation result data by using the device interaction weight to obtain operation result statistical data of the user about the target commodity; a feature generation module 303 configured to: generating operation characteristics of the user about the target commodities by using the operation result statistical data, and generating historical operation characteristics of the user by using the operation characteristics of the user about each target commodity; a data push module 304 configured to: and inputting the historical operation characteristics into a pre-trained neural network, obtaining the probability of the user performing online shopping operation on each target commodity next time, and pushing relevant data of each target commodity to the user according to the probability.
According to the embodiment, the operation result data is weighted by using the device interaction data of the user when the user performs online shopping operation on the commodities to obtain the operation result statistical data, so that the probability of the user performing online shopping operation on each commodity next time is more accurately predicted, and the data push accuracy and the user experience of the data push service are improved.
In some embodiments, the operation result data includes the number of times the user browses the target commodity, and the device interaction data has different types; the data weighting module 302 is configured to: respectively generating the times weight of browsing the target commodity by the corresponding type of users by utilizing different types of equipment interaction data; and generating a comprehensive times weight of each time the user browses the target commodity by using the times weights of at least one type so as to weight the times of the user browses the target commodity.
In some embodiments, the data weighting module 302 is configured to: and normalizing the strength data of the target commodity link clicked on the equipment by the user each time to obtain the strength weight of the target commodity link clicked on the equipment by the user each time.
In some embodiments, the data weighting module 302 is configured to: and normalizing the time length data of the target commodity browsed on the equipment by the user every time to obtain the time length weight of the target commodity browsed on the equipment by the user every time.
In some embodiments, the data weighting module 302 is configured to: determining location data for a most frequent click location of a user on a device; determining the distance between the position of clicking the target commodity link on the equipment by the user each time and the most frequent clicking position on the equipment by the user according to the position data of clicking the target commodity link on the equipment by the user each time and the position data of the most frequent clicking position on the equipment by the user; and carrying out normalization processing after taking the reciprocal of the distance to obtain the position weight of the target commodity browsed on the equipment by the user each time.
In some embodiments, the operation result data includes the number of times the user places an order for the target commodity, and the device interaction data includes force data of the user clicking an order placing icon of the target commodity on the device; the data weighting module 302 is configured to: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the order placing times of the target commodity by the user.
In some embodiments, the operation result data comprises a quota for ordering the target commodity by the user, and the equipment interaction data comprises strength data for clicking an ordering icon of the target commodity on the equipment by the user; the data weighting module 302 is configured to: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the amount of the order placing of the target commodity by the user.
In some embodiments, the operation result data includes a number of times that the user added the target item to the shopping cart, and the device interaction data includes force data of the user clicking a shopping cart icon of the target item on the device; the data weighting module 302 is configured to: and normalizing the force data of the shopping cart icon of the target commodity clicked on the equipment by the user each time to obtain the force weight of the shopping cart icon of the target commodity clicked on the equipment by the user each time so as to weight the times of adding the target commodity into the shopping cart by the user.
In some embodiments, the operation result data includes times of collecting the target commodity by the user, and the device interaction data includes force data of clicking a collection icon of the target commodity on the device by the user; the data weighting module 302 is configured to: and normalizing the strength data of clicking the collection icon of the target commodity on the equipment by the user each time to obtain the strength weight of clicking the collection icon of the target commodity on the equipment by the user each time so as to weight the times of the target commodity collection by the user.
In some embodiments, the data processing apparatus 30 further comprises a network training module 300 configured to: training the neural network by using the historical operation sample characteristics of the user and the sample commodity label of the next online shopping operation executed by the user, so that the neural network can obtain the probability of the next online shopping operation executed by the user on each target commodity according to the historical operation characteristics.
In some embodiments, the neural network includes a first fully-connected layer including a plurality of fully-connected layers and a first cascade layer for cascading feature vectors output by the plurality of fully-connected layers.
In some embodiments, the neural network further comprises a discard layer, a second fully connected layer, a global average pooling layer, and a second cascade layer; the disposal layer is used for: performing discarding operation on the feature vectors output by the first cascade layer, and respectively inputting the feature vectors output after the discarding operation into the second full-connection layer and the global average pooling layer; the second cascaded layer is to: and cascading the feature vectors output by the second full-connection layer, the global average pooling layer and the first cascading layer, and carrying out batch standardization processing on the feature vectors after cascading.
Further embodiments of the data processing apparatus of the present disclosure are described below in conjunction with fig. 4.
Fig. 4 shows a schematic structural diagram of a data processing apparatus according to further embodiments of the present disclosure. As shown in fig. 4, the data processing apparatus 40 of this embodiment includes: a memory 410 and a processor 420 coupled to the memory 410, the processor 420 being configured to perform the data processing method of any of the foregoing embodiments based on instructions stored in the memory 410.
The data processing apparatus 40 may further include an input-output interface 430, a network interface 440, a storage interface 450, and the like. These interfaces 430, 440, 450 and the connection between the memory 410 and the processor 420 may be, for example, via a bus 460. The input/output interface 430 provides a connection interface for input/output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 440 provides a connection interface for various networking devices. The storage interface 450 provides a connection interface for external storage devices such as an SD card and a usb disk.
The present disclosure also includes a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement a data processing method in any of the foregoing embodiments.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only exemplary of the present disclosure and is not intended to limit the present disclosure, so that any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
Claims (15)
1. A method of data processing, comprising:
acquiring equipment interaction data and operation result data when a user performs online shopping operation on a target commodity each time;
generating a device interaction weight when the user performs online shopping operation on the target commodity each time by using the device interaction data, and performing weighted summation on the operation result data by using the device interaction weight to obtain operation result statistical data of the user about the target commodity;
generating operation characteristics of the user about the target commodities by using the operation result statistical data, and generating historical operation characteristics of the user by using the operation characteristics of the user about each target commodity;
and inputting the historical operation characteristics into a pre-trained neural network, obtaining the probability of the user performing online shopping operation on each target commodity next time, and pushing relevant data of each target commodity to the user according to the probability.
2. The data processing method according to claim 1, wherein the operation result data includes a number of times that a user browses a target commodity, the device interaction data having different types;
the generating of the device interaction weight when the user performs the online shopping operation on the target commodity by using the device interaction data comprises: respectively generating the times weight of browsing the target commodity by the corresponding type of users by using different types of equipment interaction data; and generating a comprehensive times weight of each time the user browses the target commodity by using the times weight of at least one type so as to weight the times of the user browses the target commodity.
3. The data processing method of claim 2, wherein the generating, by using the different types of device interaction data, weights of the number of times that the corresponding types of users browse the target product each time respectively comprises:
and normalizing the strength data of the target commodity link clicked on the equipment by the user each time to obtain the strength weight of the target commodity link clicked on the equipment by the user each time.
4. The data processing method of claim 2, wherein the generating, by using the different types of device interaction data, weights of the number of times that the corresponding types of users browse the target product each time respectively comprises:
and normalizing the time length data of the target commodity browsed on the equipment by the user every time to obtain the time length weight of the target commodity browsed on the equipment by the user every time.
5. The data processing method of claim 2, wherein the generating, by using the different types of device interaction data, weights of the number of times that the corresponding types of users browse the target product each time respectively comprises:
determining location data for a most frequent click location of a user on a device;
determining the distance between the position of clicking the target commodity link on the equipment by the user each time and the most frequent clicking position on the equipment by the user according to the position data of clicking the target commodity link on the equipment by the user each time and the position data of the most frequent clicking position;
and carrying out normalization processing after taking the reciprocal of the distance to obtain the position weight of the target commodity browsed on the equipment by the user each time.
6. The data processing method according to claim 1 or 2, wherein the operation result data includes the number of times the user places an order for the target commodity, and the device interaction data includes force data of the user clicking an order placing icon of the target commodity on the device;
the generating of the device interaction weight when the user performs the online shopping operation on the target commodity by using the device interaction data comprises: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the order placing times of the target commodity by the user.
7. The data processing method according to claim 1 or 2, wherein the operation result data includes an amount of the order placement of the target commodity by the user, and the device interaction data includes strength data of the order placement icon of the target commodity clicked by the user on the device;
the generating of the device interaction weight when the user performs the online shopping operation on the target commodity by using the device interaction data comprises: and normalizing the strength data of the order placing icon of the target commodity clicked by the user on the equipment every time to obtain the strength weight of the order placing icon of the target commodity clicked by the user on the equipment every time so as to weight the amount of the order placing of the target commodity by the user.
8. The data processing method according to claim 1 or 2, wherein the operation result data includes the number of times that the user adds the target commodity to the shopping cart, and the device interaction data includes force data of the user clicking a shopping cart icon of the target commodity on the device;
the generating of the device interaction weight when the user performs the online shopping operation on the target commodity by using the device interaction data comprises: and normalizing the force data of the shopping cart icon of the target commodity clicked on the equipment by the user each time to obtain the force weight of the shopping cart icon of the target commodity clicked on the equipment by the user each time so as to weight the times of adding the target commodity into the shopping cart by the user.
9. The data processing method according to claim 1 or 2, wherein the operation result data includes the times of collecting the target commodity by the user, and the device interaction data includes force data of clicking a collection icon of the target commodity on the device by the user;
the generating of the device interaction weight when the user performs the online shopping operation on the target commodity by using the device interaction data comprises: and normalizing the strength data of clicking the collection icon of the target commodity on the equipment by the user each time to obtain the strength weight of clicking the collection icon of the target commodity on the equipment by the user each time so as to weight the times of the target commodity collection by the user.
10. The data processing method of claim 1, further comprising:
training the neural network by utilizing the historical operation sample characteristics of the user and the sample commodity label of the next online shopping operation executed by the user, so that the neural network can obtain the probability of the next online shopping operation executed by the user on each target commodity according to the historical operation characteristics.
11. The data processing method of claim 1, wherein the neural network comprises a first fully-connected layer comprising a plurality of fully-connected layers and a first cascaded layer for cascading feature vectors output by the plurality of fully-connected layers.
12. The data processing method of claim 11, wherein the neural network further comprises a discard layer, a second fully connected layer, a global average pooling layer, and a second cascade layer;
the disposal layer is for: performing discarding operation on the feature vectors output by the first cascade layer, and respectively inputting the feature vectors output after the discarding operation into a second full-connection layer and a global average pooling layer;
the second cascaded layer is to: and cascading the feature vectors output by the second full connection layer, the global average pooling layer and the first cascading layer, and performing batch standardization processing on the feature vectors after cascading.
13. A data processing apparatus comprising:
a data acquisition module configured to: acquiring equipment interaction data and operation result data when a user performs online shopping operation on a target commodity each time;
a data weighting module configured to: generating a device interaction weight when the user performs online shopping operation on the target commodity each time by using the device interaction data, and performing weighted summation on the operation result data by using the device interaction weight to obtain operation result statistical data of the user about the target commodity;
a feature generation module configured to: generating operation characteristics of the user about the target commodities by using the operation result statistical data, and generating historical operation characteristics of the user by using the operation characteristics of the user about each target commodity;
a data push module configured to: and inputting the historical operation characteristics into a pre-trained neural network, obtaining the probability of the user performing online shopping operation on each target commodity next time, and pushing relevant data of each target commodity to the user according to the probability.
14. A data processing apparatus comprising:
a memory; and
a processor coupled to the memory, the processor configured to perform the data processing method of any of claims 1 to 12 based on instructions stored in the memory.
15. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions which, when executed by a processor, implement a data processing method as claimed in any one of claims 1 to 12.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010152306.4A CN113362098A (en) | 2020-03-06 | 2020-03-06 | Data processing method, device and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010152306.4A CN113362098A (en) | 2020-03-06 | 2020-03-06 | Data processing method, device and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113362098A true CN113362098A (en) | 2021-09-07 |
Family
ID=77524133
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010152306.4A Pending CN113362098A (en) | 2020-03-06 | 2020-03-06 | Data processing method, device and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113362098A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113793180A (en) * | 2021-09-15 | 2021-12-14 | 北京沃东天骏信息技术有限公司 | User preference analysis method, device, equipment and computer storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10002322B1 (en) * | 2017-04-06 | 2018-06-19 | The Boston Consulting Group, Inc. | Systems and methods for predicting transactions |
CN108632311A (en) * | 2017-03-20 | 2018-10-09 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
CN108648049A (en) * | 2018-05-03 | 2018-10-12 | 中国科学技术大学 | A kind of sequence of recommendation method based on user behavior difference modeling |
KR20190012894A (en) * | 2017-07-29 | 2019-02-11 | 한남대학교 산학협력단 | Personalized prediction method for purchase timing using purchased pattern data and target promotion system |
CN109685540A (en) * | 2018-08-31 | 2019-04-26 | 深圳壹账通智能科技有限公司 | Electric business platform and its commercial promotions control method, equipment, readable storage medium storing program for executing |
-
2020
- 2020-03-06 CN CN202010152306.4A patent/CN113362098A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108632311A (en) * | 2017-03-20 | 2018-10-09 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
US10002322B1 (en) * | 2017-04-06 | 2018-06-19 | The Boston Consulting Group, Inc. | Systems and methods for predicting transactions |
KR20190012894A (en) * | 2017-07-29 | 2019-02-11 | 한남대학교 산학협력단 | Personalized prediction method for purchase timing using purchased pattern data and target promotion system |
CN108648049A (en) * | 2018-05-03 | 2018-10-12 | 中国科学技术大学 | A kind of sequence of recommendation method based on user behavior difference modeling |
CN109685540A (en) * | 2018-08-31 | 2019-04-26 | 深圳壹账通智能科技有限公司 | Electric business platform and its commercial promotions control method, equipment, readable storage medium storing program for executing |
Non-Patent Citations (1)
Title |
---|
李裕礞;练绪宝;徐博;王健;林鸿飞;: "基于用户隐性反馈行为的下一个购物篮推荐", 中文信息学报, no. 05, pages 215 - 222 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113793180A (en) * | 2021-09-15 | 2021-12-14 | 北京沃东天骏信息技术有限公司 | User preference analysis method, device, equipment and computer storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111523976B (en) | Commodity recommendation method and device, electronic equipment and storage medium | |
CN110503531B (en) | Dynamic social scene recommendation method based on time sequence perception | |
CN109697629B (en) | Product data pushing method and device, storage medium and computer equipment | |
CN110148023A (en) | The electric power integral Method of Commodity Recommendation and system that logic-based returns | |
CN111815415A (en) | Commodity recommendation method, system and equipment | |
CN107103049A (en) | A kind of recommendation method and the network equipment | |
CN112632405B (en) | Recommendation method, recommendation device, recommendation equipment and storage medium | |
US20100239084A1 (en) | Memorable Telephone Number Identification | |
CN112085565A (en) | Deep learning-based information recommendation method, device, equipment and storage medium | |
CN106294882A (en) | Data digging method and device | |
CN110910201A (en) | Information recommendation control method and device, computer equipment and storage medium | |
CN110858219A (en) | Logistics object information processing method and device and computer system | |
CN111310052A (en) | User portrait construction method and device and computer readable storage medium | |
CN115797005A (en) | Prediction model training method, prediction model training device, prediction model commodity recommendation method, prediction model training device, commodity recommendation device, electronic equipment and medium | |
CN113378067B (en) | Message recommendation method, device and medium based on user mining | |
CN113362098A (en) | Data processing method, device and computer readable storage medium | |
CN113435900A (en) | Transaction risk determination method and device and server | |
CN110851708B (en) | Negative sample extraction method, device, computer equipment and storage medium | |
Wang et al. | A Comparative Study on Contract Recommendation Model: Using Macao Mobile Phone Datasets | |
CN116127188A (en) | Target feedback value determining method and device, electronic equipment and storage medium | |
WO2023048807A1 (en) | Hierarchical representation learning of user interest | |
CN112084408B (en) | List data screening method, device, computer equipment and storage medium | |
CN110827044A (en) | Method and device for extracting user interest mode | |
Hanif | Applications of data mining techniques for churn prediction and cross-selling in the telecommunications industry | |
CN111475721A (en) | Information pushing method, device, equipment and 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 |