An item display method, device and equipment based on user behavior graph centrality

Abstract

Embodiments of the present application disclose a kind of based on user behavior graph centrality item display method, device and equipment.The method comprises: obtaining user behavior dataset, wherein, user behavior dataset is made of user behavior data;User behavior data includes: user identification, item identification, user behavior type, and behavior time;According to behavior time, each user behavior data in user behavior dataset is sorted, and item behavior sequence is obtained;According to item behavior sequence, the position relationship between the items in item behavior sequence and the relative weight value of the items corresponding to the position relationship, the user behavior graph of the item is constructed;And according to the user behavior graph of the item, the interaction weight matrix of the item is determined;According to the feature vector of interaction weight matrix, the item quality value of each item is determined;According to item quality value, item display is carried out.The method considers user behavior and item interaction when carrying out item display, so that item display result is more objective.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, and in particular, to an item display method and device based on user behavior graph centrality. BACKGROUND

[0002] With the development of computer technology, item display can be applied in all aspects of people's daily life. For example, online shopping recommendation, information video recommendation, and music recommendation can all be performed through item display, such as "hot spot recommendation", "column recommendation", and "selected recommendation".

[0003] In the prior art, item display is usually determined according to a single type of behavior, and the display is performed according to the item popularity. For example, the items can be sorted according to the number of clicks, the number of views, or the sales volume.

[0004] However, the display method of determining item popularity according to a single type of behavior is easy to be cracked, such as affecting the entire product ecology by brushing clicks or sales volume, and leading to over-heated items that long-term dominate the list and exacerbate the "Matthew effect". SUMMARY

[0005] The present application provides an item display method and device based on user behavior graph centrality, which considers user behavior and item interaction when displaying items, so as to make the item display result more objective and avoid the "Matthew effect".

[0006] According to an aspect of the present application, an item display method based on user behavior graph centrality is provided, which comprises:

[0007] obtaining a user behavior data set, wherein the user behavior data set is composed of at least one user behavior data; each user behavior data includes a user identifier, an item identifier, a user behavior type, and a behavior time;

[0008] sorting each user behavior data in the user behavior data set according to the behavior time to obtain an item behavior sequence;

[0009] constructing a user behavior graph of an item according to the item behavior sequence, the position relationship between items in the item behavior sequence, and the item relative weight value corresponding to the position relationship; and determining an interaction weight matrix of the item according to the user behavior graph of the item;

[0010] determining an item quality value of each item according to the feature vector of the interaction weight matrix; and performing item display according to the item quality value.

[0011] According to another aspect of the present application, an item display device based on user behavior graph centrality is provided, which comprises:

[0012] a data set obtaining module, configured to obtain a user behavior data set, wherein the user behavior data set is composed of at least one user behavior data, and each user behavior data comprises a user identifier, an item identifier, a user behavior type, and a behavior time;

[0013] an item behavior sequence determining module, configured to sort each user behavior data in the user behavior data set according to the behavior time, and obtain an item behavior sequence;

[0014] an interaction weight matrix determining module, configured to construct an item user behavior graph according to the item behavior sequence, a position relationship between items in the item behavior sequence, and an item relative weight value corresponding to the position relationship, and determine an interaction weight matrix of the item according to the item user behavior graph;

[0015] an item display module, configured to determine an item quality value of each item according to a feature vector of the interaction weight matrix, and display the item according to the item quality value.

[0016] According to another aspect of the present application, an electronic device is provided, and the electronic device comprises:

[0017] at least one processor; and

[0018] a memory connected to the at least one processor in communication; wherein

[0019] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the item display method based on user behavior graph centrality according to any one of the embodiments of the present application.

[0020] According to another aspect of the present application, a computer readable storage medium is provided, and the computer readable storage medium stores computer instructions for enabling a processor to implement the item display method based on user behavior graph centrality according to any one of the embodiments of the present application when executed by the processor.

[0021] The technical scheme of the embodiment of the application comprises the following steps: obtaining a user behavior data set, wherein the user behavior data set is composed of user behavior data; the user behavior data comprises a user identifier, an item identifier, a user behavior type, and a behavior time; sorting each user behavior data in the user behavior data set according to the behavior time to obtain an item behavior sequence; constructing an item user behavior graph according to the item behavior sequence, a position relationship between items in the item behavior sequence, and an item relative weight value corresponding to the position relationship; determining an interaction weight matrix of the item according to the item user behavior graph; determining an item quality value of each item according to a feature vector of the interaction weight matrix; and displaying the item according to the item quality value, thereby solving the problem of item display and making the item display result more objective by considering user behavior and item interaction, so as to achieve the principle of 'dividing things according to groups' and avoid the 'Matthew effect'.

[0022] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the application, nor is it used to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0024] Figure 1a is a flow chart of an item display method based on user behavior graph centrality provided by the first embodiment of the application;

[0025] Figure 1b is a structural schematic diagram of an item user behavior graph provided by the first embodiment of the application;

[0026] Figure 2 is a flow chart of an item display method based on user behavior graph centrality provided by the second embodiment of the application;

[0027] Figure 3 is a structural schematic diagram of an item display device based on user behavior graph centrality provided by the third embodiment of the application;

[0028] Figure 4 is a structural schematic diagram of an electronic device for implementing the item display method based on user behavior graph centrality of the embodiment of the application. DETAILED DESCRIPTION

[0029] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application by persons in the technical field, obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons in the technical field without creative labor should belong to the scope of protection of the present application.

[0030] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data used in this way can be exchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0031] Embodiment one

[0032] Figure 1a It is a flowchart of a user behavior graph centrality-based item display method according to embodiment one of the present application, the present embodiment can be applied to the case of sorting and displaying items, for example, when recommending items, the technical solutions of the embodiments of the present application can be used to determine item quality values for item display. The method can be performed by a user behavior graph centrality-based item display device, which can be realized in the form of hardware and / or software, and can be configured in an electronic device such as a computer or a smart phone. As shown in Figure 1a The method comprises:

[0033] Step 110, obtaining a user behavior data set.

[0034] The user behavior data set is composed of at least one piece of user behavior data. In the embodiment, user behavior data of a user in a preset time period on a platform can be collected to form the user behavior data set. For example, user behavior data in the last 7 days can be collected to form the user behavior data set. Each piece of user behavior data includes a user identifier, an item identifier, a user behavior type, and a behavior time. The user identifier can be user ID information (userID), the item identifier can be item ID information (iteamID), the user behavior type can be a type of operation performed by the user on the item, such as clicking, collecting, loving, adding to shopping cart, or purchasing, and the behavior time can be the time when the user performs the operation on the item.

[0035] In step 120, each piece of user behavior data in the user behavior data set is sorted according to the behavior time to obtain an item behavior sequence.

[0036] Each piece of user behavior data in the user behavior data set can be stored in a row. That is, each row of data in the user behavior data set includes a user identifier, an item identifier, a user behavior type, and a behavior time. Each row of data can be sorted according to the behavior time. For example, when the behavior time is closest to the current time, the row of data can be arranged in the first row of the user behavior data set; when the behavior time is farthest from the current time, the row of data can be arranged in the last row of the user behavior data set; and the same applies to sorting each piece of user behavior data in the user behavior data set.

[0037] The item behavior sequence can be understood as the arrangement result of the item after sorting each piece of user behavior data in the user behavior data set. Further, in order to improve the accuracy of determining the item quality value, after sorting each piece of user behavior data in the user behavior data set according to the behavior time, the sorting result can be classified to obtain the item behavior sequence. For example, after sorting, user behavior data corresponding to a certain user identifier can be extracted; or user behavior data corresponding to a certain user behavior type can be extracted, and then the item behavior sequence is determined. Of course, user behavior data of a certain classification can also be extracted first, and then sorted to determine the item behavior sequence.

[0038] Specifically, in an optional implementation of the embodiment, sorting each piece of user behavior data in the user behavior data set according to the behavior time to obtain an item behavior sequence includes: classifying each piece of user behavior data in the user behavior data set according to the user behavior type to obtain a user behavior data subset corresponding to each user behavior type; and sorting each piece of user behavior data in the user behavior subset according to the behavior time to obtain an item behavior sequence corresponding to each user behavior type.

[0039] In this embodiment, each user behavior data in the user behavior data set is classified according to user behavior types such as clicking, collecting, favoriting, adding to shopping cart, and purchasing, to obtain a user behavior data subset corresponding to each user behavior type. In the user behavior data subset, the user behavior data can be sorted according to behavior time, and then an item behavior sequence corresponding to each user behavior type can be obtained.

[0040] For example, the user behavior data subset of user 1 clicking an item is sorted according to behavior time as [user 1, item 3, click, time 1; user 1, item 4, click, time 2; user 1, item 2, click, time 3; user 1, item 1, click, time 4], where “;” represents line break. Then the item behavior sequence can be [item 3 (user 1, click, time 1), item 4 (user 1, click, time 2), item 2 (user 1, click, time 3), item 1 (user 1, click, time 4)]. Each item in the item behavior sequence can have a mapping relationship with the corresponding user identifier, user behavior type, and behavior time, which can facilitate locating the corresponding user behavior data and adding the user behavior type factor when determining the item quality value, thereby improving the objectivity of the item quality value determination.

[0041] Step 130: constructing an item user behavior graph according to the item behavior sequence, the positional relationship between items in the item behavior sequence, and the item relative weight value corresponding to the positional relationship; and determining an interaction weight matrix of the item according to the item user behavior graph.

[0042] The positional relationship between items in the item behavior sequence can be a positional difference value between two items. For example, in the item behavior sequence [item 3 (user 1, click, time 1), item 4 (user 1, click, time 2), item 2 (user 1, click, time 3), item 1 (user 1, click, time 4)], the positional relationship between item 3 and item 1 can be a positional difference value of 3, and the positional relationship between item 3 and item 2 can be a positional difference value of 2.

[0043] The item relative weight value corresponding to the positional relationship can be a corresponding weight value preset according to the positional difference value. The larger the positional difference value, the smaller the item relative weight value. For example, when the positional difference value is 1, the item relative weight value is 0.9; when the positional difference value is 2, the item relative weight value is 0.8; and when the positional difference value is 3, the item relative weight value is 0.7.

[0044] Figure 1b is a structural schematic diagram of an item user behavior graph provided by the embodiment one of the present application. As shown in Figure 1bAs shown, the items can be nodes of a user behavior graph of the items; connection lines can be set between the items to represent the user behavior interaction relationship between the items. The relative weight values of the items corresponding to the position relationship between the items can be marked on the connection lines. By constructing the user behavior graph of the items to determine the item quality values, the calculation result can be more objective, and the display effect can be improved.

[0045] In an optional embodiment of the embodiment of the present application, the user behavior graph of the items is constructed according to the item behavior sequence, the position relationship between the items in the item behavior sequence, and the relative weight values of the items corresponding to the position relationship, and the user behavior graph of the items is constructed according to the item position difference threshold value, the item behavior subsequence is selected from the item behavior sequence corresponding to each user behavior type, and the user behavior graph of the items is constructed according to the position relationship between the items and the relative weight values of the items corresponding to the position relationship in each item behavior subsequence corresponding to each user behavior type.

[0046] The item position difference threshold value can be understood as an upper limit of the number of items when the item behavior sequence is split. For example, the item position difference threshold value can be 3 or 5, etc. For example, when the item position difference threshold value is 3, for the item sequence [item 3 (user 1, click, time 1), item 4 (user 1, click, time 2), item 2 (user 1, click, time 3), item 1 (user 1, click, time 4)], the item behavior subsequence can be obtained: [item 3 (user 1, click, time 1), item 4 (user 1, click, time 2), item 2 (user 1, click, time 3)]; [item 4 (user 1, click, time 2), item 2 (user 1, click, time 3), item 1 (user 1, click, time 4)]; [item 2 (user 1, click, time 3), item 1 (user 1, click, time 4)]. The user behavior graph of the items can be constructed according to each item behavior subsequence, the cutting of the item sequence can be realized, the coherence of the user behavior interest can be introduced into the user behavior graph of the items, and the feature sparsity problem of the behavior can be relieved.

[0047] In the embodiment, the interaction weight matrix of the items can be determined according to the user behavior graph of the items, that is, all the items are sorted, wherein i represents the i th item after sorting and the i th row in the interaction weight matrix, and j represents the j th item after sorting and the j th column in the interaction weight matrix. When i=j, the value of the i th row and the j th column in the interaction weight matrix is 0; when i≠j, the value of the i th row and the j th column in the interaction weight matrix is the relative weight value of the item i and the item j.

[0048] In step 140, the feature vectors of the interaction weight matrix are determined to determine the item quality values of each item, and the items are displayed according to the item quality values.

[0049] The eigenvector of the interaction weight matrix can be x satisfying Ax = λx, A is the interaction weight matrix, x is the eigenvector, and λ is the eigenvalue. The item quality value of the item i can be the ith value in x. The solving manner can be to use the power iteration method to make (A - λI)x = 0 to obtain the eigenvalue λ of A and the corresponding eigenvector x. I is the unit matrix of A. The item display according to the item quality value can be to preferentially display items with larger item quality values.

[0050] The technical scheme of the embodiment is to obtain a user behavior data set, wherein the user behavior data set is composed of user behavior data; the user behavior data includes a user identifier, an item identifier, a user behavior type, and a behavior time; each user behavior data in the user behavior data set is sorted according to the behavior time to obtain an item behavior sequence; an item user behavior graph is constructed according to the item behavior sequence, a position relationship between items in the item behavior sequence, and an item relative weight value corresponding to the position relationship; an interaction weight matrix of the item is determined according to the item user behavior graph; an item quality value of each item is determined according to an eigenvector of the interaction weight matrix; and item display is performed according to the item quality value, thereby solving the problem of item display. When performing item display, user behavior and item interaction can be considered, the item display result is more objective, the principle of “birds of a feather flock together” can be achieved, and the “Matthew effect” can be avoided.

[0051] Embodiment Two

[0052] Figure 2 is a flowchart of an item display method based on user behavior graph centrality according to the embodiment two of the present application. The embodiment is a further refinement of the above technical scheme, and the technical scheme in the embodiment can be combined with each optional scheme in one or more of the above embodiments. As shown in the figure, the method includes: Figure 2

[0053] Step 210, obtaining a user behavior data set.

[0054] The user behavior data set is composed of at least one piece of user behavior data; each user behavior data includes a user identifier, an item identifier, a user behavior type, and a behavior time.

[0055] Step 220, classifying each user behavior data in the user behavior data set according to the user behavior type to obtain a user behavior data subset corresponding to each user behavior type.

[0056] In order to improve the reliability of determining the item relative weight value, the user behavior data can be classified. Specifically, the classification can be performed according to the user behavior type. In some embodiments, the classification can also be performed according to the user identifier or the behavior time. ​

[0057] Exemplarily, the user behavior data set can be classified according to the user behavior type, and classified into user behavior data subsets corresponding to user behavior types such as clicking, collecting, favoriting, adding to shopping cart, and purchasing.

[0058] Step 230: According to the behavior time, each user behavior data in the user behavior subset is sorted to obtain an item behavior sequence corresponding to each user behavior type.

[0059] In an optional embodiment of the embodiment of the application, before each user behavior data in the user behavior data set is sorted according to the behavior time to obtain the item behavior sequence, it further includes: determining user repeated behavior data in the user behavior data set according to the user identifier, the item identifier, and the user behavior type; retaining user latest behavior data in the user repeated behavior data and deleting other user repeated behavior data according to the behavior time to update the user behavior data set.

[0060] When the user identifier, the item identifier, and the user behavior type of two or more user behavior data in the user behavior data set are consistent, it can be determined that the two or more user behavior data are user repeated behavior data. For the user repeated behavior data, one can be selectively retained. For example, the user latest behavior data closest to the current time can be retained according to the behavior time. By deleting the repeated data in the original user behavior data set, the data processing amount can be reduced, the calculation efficiency can be improved, and the reliability of determining the item quality value can also be improved.

[0061] Step 240: According to the item position difference threshold, an item behavior sub-sequence is selected in the item behavior sequence corresponding to each user behavior type.

[0062] Step 250: In each item behavior sub-sequence corresponding to each user behavior type, an item user behavior graph is constructed according to the position relationship between items and the item relative weight value corresponding to the position relationship.

[0063] In an optional embodiment of the embodiment of the application, after the item user behavior graph is constructed, it further includes: detecting whether there are at least two item relative weight values between two items in the item user behavior graph according to the item identifier; if yes, determining a target item relative weight value for the two items, and updating the item user behavior graph.

[0064] In the different sub-sequences of the behavior of the different items, the position relationship between the items can be different, so that there can be multiple item relative weight values between the items, and each item relative weight value can be different. In this embodiment, a target item relative weight value can be determined in the multiple item relative weights. For example, the maximum value or the average value of each item relative weight between two items can be taken as the target item relative weight value between the two items. Alternatively, each item relative weight between two items can be combined with a user behavior type or a user attribute to obtain the target item relative weight value between the two items. The user attribute can indicate the importance of the user. For example, the user behavior types of purchase, add to cart, collect, favorite, and click can be set with weights, and the target item relative weight value between two items can be determined by the weights of the user behavior types and the corresponding item relative weights.

[0065] In step 260, an interaction weight matrix of the items is determined according to the user behavior graph of the items, a maximum eigenvalue of the interaction weight matrix is determined, and a maximum eigenvector corresponding to the maximum eigenvalue is determined.

[0066] The maximum eigenvalue can be the maximum value λ max of the solved λ max , and the maximum eigenvector can be x max that satisfies (A-λ max I)x max =0.

[0067] In step 270, the values in the maximum eigenvector are taken as item quality values of the items, and the items are displayed according to the item quality values.

[0068] In this embodiment, the maximum eigenvector can best represent the characteristics of the interaction weight matrix, so the values in the maximum eigenvector x max can be taken as the item quality values of the items. Specifically, the i th value in x max can be taken as the item quality value of item i, so as to improve the reliability and objectivity of the item display.

[0069] In an optional implementation of the embodiment of the present application, the item display according to the item quality values comprises: determining the item quality values of each item corresponding to each user behavior type according to the user behavior graph of the item corresponding to each user behavior type and the item quality values corresponding to the user behavior graph of the item; performing fusion calculation to obtain an item comprehensive quality value of each item according to each user behavior type and the item quality values of each item corresponding to each user behavior type; and displaying the items according to the item comprehensive quality values.

[0070] The item quality value can be determined according to a user behavior data subset corresponding to different user behavior types. Each user behavior type can have a different weight, and the weight and the item quality value are fused to calculate an item comprehensive quality value. For example, the item comprehensive quality value can be a weighted sum of the weights of the user behavior types and the item quality values. When displaying the item, the item with a higher item comprehensive quality value can be displayed first. The above display method can be applied to various scenarios, such as an actual hot recommendation scenario.

[0071] The technical scheme of the embodiment of the application comprises the following steps: obtaining a user behavior dataset; classifying each user behavior data in the user behavior dataset according to a user behavior type to obtain a user behavior data subset corresponding to each user behavior type; sorting each user behavior data in the user behavior subset according to a behavior time to obtain an item behavior sequence corresponding to each user behavior type; selecting an item behavior subsequence in the item behavior sequence corresponding to each user behavior type according to an item position difference threshold; constructing an item user behavior graph in each item behavior subsequence corresponding to each user behavior type according to a position relationship between items and an item relative weight value corresponding to the position relationship; determining an interaction weight matrix of the item according to the item user behavior graph; determining a maximum eigenvalue of the interaction weight matrix and a maximum eigenvector corresponding to the maximum eigenvalue; taking a value in the maximum eigenvector as an item quality value of each item; and displaying the item according to the item quality value. The problem of item display is solved. When displaying the item, the user behavior and the item interaction can be considered, the item display result is more objective, the items can be classified according to the group, and the Matthew effect can be avoided. At the same time, the item behavior sequence is cut by the item position difference threshold, the coherence of the user behavior interest can be ensured to be integrated into the item user behavior graph, and the problem of feature sparsity can be alleviated. The item comprehensive quality value is determined by fusing the user behavior type and the corresponding item quality value, which can take into account various behaviors of the user to the item in various actual recommendation scenarios. The weight of the user behavior type can be configured, and the flexibility of the embodiment of the application can be improved.

[0072] In the technical scheme of the embodiment of the application, the acquisition, storage and application of the user behavior data comply with relevant laws and regulations and do not violate public order and good customs.

[0073] Embodiment three

[0074] Figure 3 is a structural schematic diagram of an item display device based on user behavior graph centrality provided by the embodiment three of the application. As shown in the figure, the device comprises a dataset acquisition module 310, an item behavior sequence determination module 320, an interaction weight matrix determination module 330 and an item display module 340. Figure 3 The dataset acquisition module 310 is configured to acquire a user behavior dataset. The item behavior sequence determination module 320 is configured to classify each user behavior data in the user behavior dataset according to a user behavior type to obtain a user behavior data subset corresponding to each user behavior type. The interaction weight matrix determination module 330 is configured to sort each user behavior data in the user behavior subset according to a behavior time to obtain an item behavior sequence corresponding to each user behavior type, select an item behavior subsequence in the item behavior sequence corresponding to each user behavior type according to an item position difference threshold, construct an item user behavior graph in each item behavior subsequence corresponding to each user behavior type according to a position relationship between items and an item relative weight value corresponding to the position relationship, and determine an interaction weight matrix of the item according to the item user behavior graph. The item display module 340 is configured to determine a maximum eigenvalue of the interaction weight matrix and a maximum eigenvector corresponding to the maximum eigenvalue, take a value in the maximum eigenvector as an item quality value of each item, and display the item according to the item quality value.

[0075] The data set acquisition module 310 is configured to acquire a user behavior data set, wherein the user behavior data set is composed of at least one piece of user behavior data, and each piece of user behavior data includes a user identifier, an item identifier, a user behavior type, and a behavior time.

[0076] The item behavior sequence determination module 320 is configured to sort each piece of user behavior data in the user behavior data set according to the behavior time, to obtain an item behavior sequence.

[0077] The interaction weight matrix determination module 330 is configured to construct an item user behavior graph according to the item behavior sequence, a positional relationship between items in the item behavior sequence, and an item relative weight value corresponding to the positional relationship, and determine an interaction weight matrix of the item according to the item user behavior graph.

[0078] The item display module 340 is configured to determine an item quality value of each item according to a feature vector of the interaction weight matrix, and display the items according to the item quality values.

[0079] Optionally, the apparatus further includes:

[0080] The user repeated behavior data determination module is configured to determine user repeated behavior data in the user behavior data set according to the user identifier, the item identifier, and the user behavior type, before sorting each piece of user behavior data in the user behavior data set according to the behavior time to obtain the item behavior sequence.

[0081] The user behavior data set update module is configured to retain user latest behavior data in the user repeated behavior data and delete other user repeated behavior data according to the behavior time, to update the user behavior data set.

[0082] Optionally, the item behavior sequence determination module 320 includes:

[0083] The user behavior data subset determination unit is configured to classify each piece of user behavior data in the user behavior data set according to the user behavior type, to obtain a user behavior data subset corresponding to each user behavior type.

[0084] The item behavior sequence determination unit is configured to sort each piece of user behavior data in the user behavior subset according to the behavior time, to obtain an item behavior sequence corresponding to each user behavior type.

[0085] Optionally, the interaction weight matrix determination module 330 includes:

[0086] The item behavior sub-sequence selection unit is configured to select an item behavior sub-sequence from the item behavior sequence corresponding to each user behavior type according to an item position difference threshold.

[0087] The user behavior graph construction unit constructs a user behavior graph of the items according to the position relationship between the items and the item relative weight value corresponding to the position relationship in each item behavior subsequence corresponding to each user behavior type.

[0088] Optionally, the apparatus further comprises:

[0089] The item relative weight value detection module detects whether there are at least two item relative weight values between two items in the user behavior graph of the items according to the item identifiers after the user behavior graph of the items is constructed.

[0090] The user behavior graph updating module determines a target item relative weight value for the two items and updates the user behavior graph of the items if there are at least two item relative weight values between the two items.

[0091] Optionally, the item display module 340 comprises:

[0092] The maximum eigenvector determination unit determines a maximum eigenvalue of the interaction weight matrix and a maximum eigenvector corresponding to the maximum eigenvalue.

[0093] The item quality value determination unit takes the value in the maximum eigenvector as an item quality value of each item.

[0094] Optionally, the item display module 340 comprises:

[0095] The other item quality value determination unit determines an item quality value corresponding to each user behavior type for each item according to the user behavior graph of the item corresponding to each user behavior type and the item quality value corresponding to the user behavior graph of the item.

[0096] The item comprehensive quality value determination unit performs fusion calculation to obtain an item comprehensive quality value of each item according to each user behavior type and the item quality value corresponding to each user behavior type for each item.

[0097] The item display unit displays the items according to the item comprehensive quality value.

[0098] The item display apparatus based on the user behavior graph centrality provided in the embodiments of the application can perform the item display method based on the user behavior graph centrality provided in any of the embodiments of the application, has the function modules and beneficial effects corresponding to the execution method.

[0099] Embodiment Four

[0100] Figure 4A structural diagram of an electronic device 10 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present application described and / or claimed in this document.

[0101] As shown in Figure 4 The electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., connected in communication with the at least one processor 11, where the memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer programs stored in the read-only memory (ROM) 12 or loaded into the random access memory (RAM) 13 from the storage unit 18. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0102] Various components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc., an output unit 17, such as various types of displays, speakers, etc., a storage unit 18, such as a magnetic disk, an optical disk, etc., and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0103] The processor 11 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the item presentation method based on user behavior graph centrality.

[0104] In some embodiments, the item presentation method based on user behavior graph centrality can be implemented as a computer program tangibly embodied in a computer readable storage medium, e.g., storage unit 18. In some embodiments, parts or all of the computer program can be loaded and / or installed onto electronic device 10 via, e.g., ROM 12 and / or communication unit 19. When the computer program is loaded onto RAM 13 and executed by processor 11, one or more steps of the item presentation method based on user behavior graph centrality as described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the item presentation method based on user behavior graph centrality by way of other means (e.g., by way of firmware).

[0105] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0106] Computer programs used to implement the methods of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed by the processor of the machine, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0107] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0108] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0109] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0110] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0111] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, each step described in the present application can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which is not limited herein.

[0112] The above detailed description does not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An item presentation method based on user behavior graph centrality, characterized by, The method comprises the following steps: obtaining a user behavior data set, wherein the user behavior data set is composed of at least one user behavior data, each user behavior data comprises a user identifier, an item identifier, a user behavior type, and a behavior time; sorting each user behavior data in the user behavior data set according to the behavior time to obtain an item behavior sequence; constructing an item user behavior graph according to the item behavior sequence, a position relationship between items in the item behavior sequence, and an item relative weight value corresponding to the position relationship, and determining an interaction weight matrix of the item according to the item user behavior graph; determining an item quality value of each item according to a feature vector of the interaction weight matrix, and performing item display according to the item quality value; constructing an item user behavior graph according to the item behavior sequence, a position relationship between items in the item behavior sequence, and an item relative weight value corresponding to the position relationship, comprises: selecting an item behavior sub-sequence in the item behavior sequence corresponding to each user behavior type according to an item position difference threshold; constructing an item user behavior graph according to a position relationship between items and an item relative weight value corresponding to the position relationship in each item behavior sub-sequence corresponding to each user behavior type; determining an item quality value of each item according to a feature vector of the interaction weight matrix, comprises: determining a maximum eigenvalue of the interaction weight matrix and a maximum eigenvector corresponding to the maximum eigenvalue; corresponding values in the maximum eigenvector are taken as the item quality value of each item; wherein the position relationship between items in the item behavior sequence is a position difference value between two items, the item relative weight value corresponding to the position relationship is a corresponding weight value preset according to the position difference value, and the position difference value is inversely proportional to the item relative weight value.

2. The method of claim 1, wherein, Before sorting each user behavior data in the user behavior data set according to the behavior time to obtain an item behavior sequence, the method further comprises: determining user repeated behavior data in the user behavior data set according to the user identifier, the item identifier, and the user behavior type; retaining user latest behavior data in the user repeated behavior data and deleting other user repeated behavior data according to the behavior time to update the user behavior data set.

3. The method of claim 2, wherein, Sorting each user behavior data in the user behavior data set according to the behavior time to obtain an item behavior sequence comprises: classifying each user behavior data in the user behavior data set according to the user behavior type to obtain a user behavior data subset corresponding to each user behavior type; sorting each user behavior data in the user behavior subset according to the behavior time to obtain an item behavior sequence corresponding to each user behavior type.

4. The method of claim 1, wherein, After constructing an item user behavior graph, the method further comprises: detecting whether there are at least two item relative weight values between two items in the item user behavior graph according to the item identifier. If yes, a target item relative weight value is determined for the two items, and a user behavior graph of the items is updated.

5. The method according to claim 1 or 4, characterized in that, Item display is performed according to the item quality values. An item quality value corresponding to each item and each user behavior type is determined according to the user behavior graph of the item corresponding to each user behavior type and the item quality value corresponding to the user behavior graph of the item. An item comprehensive quality value of each item is obtained through fusion calculation according to each user behavior type and the item quality value corresponding to each user behavior type of each item. Item display is performed according to the item comprehensive quality values.

6. An article display device based on user behavior graph centrality, comprising: It comprises: a data set acquisition module for acquiring a user behavior data set, wherein the user behavior data set is composed of at least one user behavior data; each user behavior data comprises a user identifier, an item identifier, a user behavior type, and a behavior time; an item behavior sequence determination module for sorting each user behavior data in the user behavior data set according to the behavior time to obtain an item behavior sequence; an interaction weight matrix determination module for constructing a user behavior graph of an item according to the item behavior sequence, a position relationship between items in the item behavior sequence, and an item relative weight value corresponding to the position relationship, and determining an interaction weight matrix of the item according to the user behavior graph of the item; an item display module for determining an item quality value of each item according to a feature vector of the interaction weight matrix, and performing item display according to the item quality value; the interaction weight matrix determination module comprises: an item behavior subsequence selection unit for selecting an item behavior subsequence in the item behavior sequence corresponding to each user behavior type according to an item position difference threshold value; an item user behavior graph construction unit for constructing a user behavior graph of an item in each item behavior subsequence corresponding to each user behavior type according to a position relationship between items and an item relative weight value corresponding to the position relationship; the item display module comprises: a maximum feature vector determination unit for determining a maximum eigenvalue of the interaction weight matrix and a maximum feature vector corresponding to the maximum eigenvalue; an item quality value determination unit for taking a value in the maximum feature vector as an item quality value of each item; wherein the position relationship between items in the item behavior sequence is a position difference value between two items; the item relative weight value corresponding to the position relationship is a corresponding weight value preset according to the position difference value; the position difference value is inversely proportional to the item relative weight value.

7. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the item display method based on user behavior graph centrality in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing the processor to implement the item display method based on the user behavior graph centrality according to any one of claims 1-5 when executed.

Images