Interactive product recommendation method, device, computer equipment, and storage medium

Abstract

The present application relates to the field of big data technology, and in particular to an interactive product recommendation method, device, computer equipment, storage medium and computer program product. The method comprises: obtaining historical interaction data corresponding to a target object. Determining the historical time series relationship between two interactive products in each product combination in the historical interaction data. Filtering out multiple sequences to be processed from the current interaction data corresponding to the target object. Based on the multiple sequences to be processed, determining each group of product combinations to be processed, and determining the support of each group of product combinations to be processed based on the multiple historical time series relationships and each time node. For each sequence to be processed, determining the average support of the corresponding sequence to be processed based on the product combinations to be processed contained in the corresponding sequence to be processed. Based on the average support of each sequence to be processed, determining a target sequence that meets the strong timeliness condition from the multiple sequences to be processed. In this way, the effectiveness of interactive product recommendation is greatly improved.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to an interactive product recommendation method, apparatus, computer equipment, storage medium and computer program product. Background Art

[0002] With the development of big data technology, it is often necessary to analyze and process interactive data and recommend interactive products to users based on the processed interactive data. The characteristics of interactive data are large data volume and rapid changes.

[0003] Since interactive data is highly dispersed and poorly manageable, and there is a large amount of redundant data in the interactive data, it is difficult to ensure the data quality of the interactive data. At the same time, interactive data is a typical dynamic data, that is, the data reflecting personal attributes and personal status are constantly changing. Therefore, when analyzing and processing interactive data whose data quality cannot be ensured and changes rapidly, it is difficult to make the most of the effective information in the interactive data, so that it is impossible to obtain more credible processed interactive data, and further, it is impossible to ensure that effective and reliable interactive products are recommended to users, and there is a problem of low effectiveness of interactive product recommendations. Summary of the invention

[0004] Based on this, it is necessary to provide an interactive product recommendation method, device, computer equipment, computer-readable storage medium and computer program product to address the above technical problems.

[0005] In a first aspect, the present application provides an interactive product recommendation method. The method comprises:

[0006] Acquire historical interaction data corresponding to the target object, wherein the historical interaction data includes multiple product combinations, each product combination consisting of two adjacent interactive products;

[0007] Determine the historical temporal relationship between two interactive products in each product combination in the historical interaction data;

[0008] Filtering out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed containing interaction products corresponding to a plurality of time nodes respectively;

[0009] Determine each group of product combinations to be processed based on the multiple sequences to be processed, and determine the support level of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed;

[0010] For each sequence to be processed, based on the product combinations to be processed contained in the corresponding sequence to be processed, determine the average support of the corresponding sequence to be processed;

[0011] Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object in a corresponding time sequence.

[0012] In a second aspect, the present application also provides an interactive product recommendation device. The device comprises:

[0013] An acquisition module, used to acquire historical interaction data corresponding to a target object, wherein the historical interaction data includes a plurality of product combinations, each of which consists of two adjacent interactive products;

[0014] A first determination module is used to determine the historical time sequence relationship between two interactive products in each product combination in the historical interaction data;

[0015] A screening module, used to screen out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed containing interaction products corresponding to a plurality of time nodes respectively;

[0016] A second determination module is used to determine each group of product combinations to be processed based on the multiple sequences to be processed, and determine the support degree of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed;

[0017] A third determination module is used to determine, for each sequence to be processed, an average support of the corresponding sequence to be processed based on the combination of products to be processed contained in the corresponding sequence to be processed;

[0018] The fourth determination module is used to determine a target sequence that meets the strong timeliness condition from multiple sequences to be processed based on the average support of each sequence to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object in a corresponding time sequence.

[0019] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0020] Acquire historical interaction data corresponding to the target object, wherein the historical interaction data includes multiple product combinations, each product combination consisting of two adjacent interactive products;

[0021] Determine the historical temporal relationship between two interactive products in each product combination in the historical interaction data;

[0022] Filtering out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed containing interaction products corresponding to a plurality of time nodes respectively;

[0023] Determine each group of product combinations to be processed based on the multiple sequences to be processed, and determine the support level of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed;

[0024] For each sequence to be processed, based on the product combinations to be processed contained in the corresponding sequence to be processed, determine the average support of the corresponding sequence to be processed;

[0025] Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object in a corresponding time sequence.

[0026] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0027] Acquire historical interaction data corresponding to the target object, wherein the historical interaction data includes multiple product combinations, each product combination consisting of two adjacent interactive products;

[0028] Determine the historical temporal relationship between two interactive products in each product combination in the historical interaction data;

[0029] Filtering out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed containing interaction products corresponding to a plurality of time nodes respectively;

[0030] Determine each group of product combinations to be processed based on the multiple sequences to be processed, and determine the support level of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed;

[0031] For each sequence to be processed, based on the product combinations to be processed contained in the corresponding sequence to be processed, determine the average support of the corresponding sequence to be processed;

[0032] Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object in a corresponding time sequence.

[0033] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0034] Acquire historical interaction data corresponding to the target object, wherein the historical interaction data includes multiple product combinations, each product combination consisting of two adjacent interactive products;

[0035] Determine the historical temporal relationship between two interactive products in each product combination in the historical interaction data;

[0036] Filtering out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed containing interaction products corresponding to a plurality of time nodes respectively;

[0037] Determine each group of product combinations to be processed based on the multiple sequences to be processed, and determine the support level of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed;

[0038] For each sequence to be processed, based on the product combinations to be processed contained in the corresponding sequence to be processed, determine the average support of the corresponding sequence to be processed;

[0039] Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object in a corresponding time sequence.

[0040] The above-mentioned interactive product recommendation method, device, computer equipment, storage medium and computer program product obtain historical interaction data corresponding to the target object, and the historical interaction data includes multiple groups of product combinations, each group of product combinations consists of two adjacent interactive products. Determine the historical time series relationship between the two interactive products in each group of product combinations in the historical interaction data. By screening out multiple sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed contains interactive products corresponding to multiple time nodes, respectively, it is possible to clarify the multiple interaction situations of the current interactive data at multiple time nodes. Based on the multiple sequences to be processed, each group of product combinations to be processed is determined, and based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed, the support of each group of product combinations to be processed is determined, thereby being able to clearly reflect the strength of the time series relationship between the two interactive products in each group of product combinations to be processed. For each sequence to be processed, the average support of the corresponding sequence to be processed is determined based on the combination of products to be processed contained in the corresponding sequence to be processed, so that the temporal relationship represented by each sequence to be processed can be accurately and specifically determined, and then the strength of the temporal relationship of each sequence to be processed can be determined. Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to recommend to the terminal corresponding to the target object according to the corresponding time sequence. Therefore, based on the temporal relationship of each interactive product in the target sequence with strong timeliness conditions, interactive products can be reasonably and accurately recommended to the target object, greatly improving the effectiveness of interactive product recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A diagram of an application environment of an interactive product recommendation method in an embodiment;

[0042] Figure 2 A schematic diagram of a flow chart of an interactive product recommendation method in one embodiment;

[0043] Figure 3 A schematic diagram of obtaining a sequence to be processed in one embodiment;

[0044] Figure 4 A schematic flow chart of an interactive product recommendation method in another embodiment;

[0045] Figure 5 is a structural block diagram of an interactive product recommendation device in one embodiment;

[0046] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] The interactive product recommendation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 and the server 104 can be used separately to execute the interactive product recommendation method provided in the embodiment of the present application, or the system can be used to execute the interactive product recommendation method provided in the embodiment of the present application. For example, the server 104 can obtain historical interaction data corresponding to the target object from the terminal 102, and the historical interaction data includes multiple groups of product combinations, and each group of product combinations consists of two adjacent interactive products. The server 104 determines the historical time series relationship between two interactive products in each group of product combinations in the historical interaction data. The server 104 screens out multiple sequences to be processed from the current interaction data corresponding to the target object, and each of the sequences to be processed contains interactive products corresponding to multiple time nodes. The server 104 determines each group of product combinations to be processed based on multiple sequences to be processed, and determines the support of each group of product combinations to be processed based on multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed. For each sequence to be processed, the server 104 determines the average support of the corresponding sequence to be processed based on the product combinations to be processed contained in the corresponding sequence to be processed. Based on the average support of each sequence to be processed, the server 104 determines a target sequence that meets the strong timeliness condition from multiple sequences to be processed, and each interactive product in the target sequence is used to recommend to the terminal 102 corresponding to the target object according to the corresponding time sequence. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and Internet of Things devices. The Internet of Things devices can be smart speakers, smart TVs, etc. The server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0049] In one embodiment, Figure 2 As shown, an interactive product recommendation method is provided, and the method is applied to a computer device as an example for explanation. The computer device may be Figure 1 Terminal or server.

[0050] The interactive product recommendation method comprises the following steps:

[0051] Step S202, obtaining historical interaction data corresponding to the target object, wherein the historical interaction data includes a plurality of product combinations, each of which consists of two adjacent interactive products.

[0052] The target object may be a user, the historical interaction data is the data generated during the historical transaction of the target object, and the interactive product is the transaction commodity of the target object. The user is a user in the e-commerce platform scenario, and the data involved is the user's purchase record on the e-commerce platform.

[0053] Specifically, the computer device obtains the initial historical interaction data corresponding to the target object, and determines the type of the interaction product by traversing the interaction products in the initial historical interaction data. The computer device determines the identity of each interaction product in turn according to the type of the interaction product, and obtains the historical interaction data. The historical interaction data contains multiple groups of product combinations, and each group of product combinations consists of two adjacent interaction products.

[0054] For example, a computer device obtains the initial historical interaction data of each entity in a historical time period, where the entity is the target object. For entity A, based on the initial historical interaction data of entity A, all interactive products in the initial interaction data are traversed to obtain 100 non-repeating interactive products, and then the 100 non-repeating interactive products are assigned IDs (Identity documents, identity identification numbers) in turn to obtain historical interaction data. For example, the historical interaction data of entity A is (1,2,3,4,...,100). Among them, every two adjacent interactive products are a product combination. For example, an interactive product with ID 1 is adjacent to an interactive product with ID 2, which can be regarded as a product combination. Similar example code is as follows:

[0055]

[0056]

[0057] Step S204, determining the historical time sequence relationship between two interactive products in each product combination in the historical interaction data.

[0058] The historical time sequence relationship is used to characterize the customized transaction sequence of the two interactive products. For example, it can be determined by the transaction time sequence of the actual transaction of the two interactive products, or it can be determined by the reverse order of the transaction time sequence. Specifically, the computer device obtains the historical interaction data and determines each group of product combinations based on the historical interaction data. The computer device determines the historical time sequence relationship between the two interactive products in each group of product combinations based on each group of product combinations.

[0059] For example, for the historical interaction data (1, 2, 3, 4, 5) corresponding to the target object A, the historical interaction data is constructed according to the time sequence of transactions of each interactive product. The computer device determines 4 groups of product combinations based on the historical interaction data. Among them, product combination A is composed of interactive products with ID 1 and interactive products with ID 2. For product combination A, the computer device determines the historical time sequence relationship between the two interactive products in the product combination based on the order in which the two interactive products are traded. The historical time sequence relationship can be that the time sequence of the interactive product with ID 1 is before the interactive product with ID 2, or that the time sequence of the interactive product with ID 1 is after the interactive product with ID 2. The computer device traverses each target object in turn, determines the historical time sequence relationship between the two interactive products in multiple groups of product combinations in each target object in turn, and counts and counts each historical time sequence relationship. Similar example code is as follows:

[0060]

[0061] Step S206: Filter out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed includes interaction products corresponding to a plurality of time nodes.

[0062] Among them, the time node is used to represent the time point when the interactive product interacts. The sequence to be processed is the combination of interactive products at each time node in the target time period. The interaction order is the transaction order. For example, the target time period is a week, where the time node is a day. For target object A, at least one interactive product is purchased every day, such as a mn , m represents the day, and n represents the identity of the purchase on that day. Then for the target object A, one of the sequences to be processed can be (a 11 , a 23 , ..., a 75 ).

[0063] Specifically, the computer device obtains current interaction data corresponding to the target object, and determines the interaction products corresponding to each time node based on the target time period corresponding to the current interaction data and the current interaction data. The computer device determines multiple sequences to be processed based on the interaction products corresponding to each time node.

[0064] Step S208, determining each group of product combinations to be processed based on the multiple sequences to be processed, and determining the support level of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed.

[0065] Among them, support is used to reflect the situation in which a state timeliness rule with the same attribute is satisfied in the same data set. Among them, the timeliness rule, that is, timeliness, is time-related. This timeliness rule can be used to describe the degree of data update, so as to study the impact of data availability, and is mainly evaluated by three indicators: data freshness, decay, and timely availability. Among them, the timeliness used to judge the data can be determined based on the timeliness of the timing rule. The definition of timeliness is: on a relation R, if you want to determine the state timeliness between different tuples belonging to the same entity, you first need to determine the tuple attributes and value types. On the relation R, for the tuples (t1, t2) of the same data entity, if some attributes appear in the time order of the tuples, they increase, decrease, or obey a specific state change sequence: Let the relational database model R = (EID, A1, A2, A3, …, Am) be a database model containing m attributes, EID is used as an entity identifier, and the same EID identifier corresponds to the same entity (entity), and entity recognition technology can generate EID; A i The range of values ​​can be used ran(A i ) indicates that A i is the i-th attribute.

[0066] Specifically, the computer device obtains multiple sequences to be processed, and determines each group of product combinations to be processed from the multiple sequences to be processed. The computer device obtains multiple historical time series relationships, and determines the support of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed. Each group of product combinations to be processed consists of two adjacent interactive products. Step S210, for each sequence to be processed, based on the product combinations to be processed included in the corresponding sequence to be processed, determine the average support of the corresponding sequence to be processed.

[0067] Specifically, the computer device obtains multiple sequences to be processed and determines the product combinations to be processed contained in each sequence to be processed. The computer device determines the support of each product combination to be processed based on each product combination to be processed. For each sequence to be processed, the computer device determines the average support of the corresponding sequence to be processed based on the corresponding sequence to be processed and the support of the product combinations to be processed contained in the corresponding sequence to be processed.

[0068] Step S212, based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object according to the corresponding time sequence.

[0069] Specifically, the computer device obtains the average support and strong timeliness condition of each sequence to be processed, and determines a target sequence that meets the strong timeliness condition from multiple sequences to be processed based on the average support and strong timeliness condition of each sequence to be processed. Each interactive product in the target sequence is used to push the interactive product value to the terminal corresponding to the target object according to the corresponding time sequence. For example, for the target sequence of the target object (b 11 , b 26 , ..., b mn ), if the target object trades the interactive product b 11 , then based on the order of the interactive products in the target sequence, first the interactive product b 26 Recommended to the target person’s mobile phone.

[0070] In the above interactive product recommendation method, by obtaining historical interactive data corresponding to the target object, the historical interactive data includes multiple groups of product combinations, each group of product combinations consists of two adjacent interactive products. The historical time series relationship between the two interactive products in each group of product combinations in the historical interactive data is determined. By screening multiple sequences to be processed from the current interactive data corresponding to the target object, each of the sequences to be processed includes interactive products corresponding to multiple time nodes, respectively, so that the multiple interactive situations of the current interactive data in multiple time nodes can be clarified. Based on multiple sequences to be processed, each group of product combinations to be processed is determined, and based on multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed, the support of each group of product combinations to be processed is determined, so that the strength of the time series relationship between the two interactive products in each group of product combinations to be processed can be clearly reflected. For each sequence to be processed, based on the product combinations to be processed included in the corresponding sequence to be processed, the average support of the corresponding sequence to be processed is determined, so that the time series relationship represented by each sequence to be processed can be accurately and specifically determined, and then the strength of the time series relationship of each sequence to be processed can be determined. Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to recommend to the terminal corresponding to the target object in the corresponding time sequence. Therefore, based on the time sequence relationship of each interactive product in the target sequence with strong timeliness condition, interactive products can be reasonably and accurately recommended to the target object, greatly improving the effectiveness of interactive product recommendation.

[0071] In one embodiment, the method of selecting multiple sequences to be processed from the current interaction data corresponding to the target object includes: obtaining the current interaction data corresponding to the target object, wherein the current interaction data includes multiple interaction products corresponding to time nodes. Constructing a decision tree according to the time nodes corresponding to the interaction products, wherein the nodes at the same layer in the decision tree are interaction products with the same time nodes. Selecting one interaction product from each layer in turn according to the order of each time node for combination, and obtaining multiple sequences to be processed corresponding to the target object.

[0072] Among them, the decision tree is used to represent a mapping relationship between object attributes and object values. Each node in the decision tree represents an object, and each fork path represents a possible attribute value, and each leaf node corresponds to the value of the object represented by the path from the root node to the leaf node.

[0073] Specifically, the computer device obtains the current interaction data of the target object in the target time period, and the current interaction data includes multiple interaction products corresponding to the time nodes. Based on the interaction products of each time node in the current interaction data, the computer device uses the interaction product of the same time node as a layer of the decision tree to obtain a decision tree. Among them, different layers in the decision tree correspond to different time nodes. Based on the decision tree, the computer device selects one interaction product from each layer in turn according to the order of each time node for combination, and obtains multiple sequences to be processed corresponding to the target object.

[0074] For example, Figure 3 As shown, the computer device obtains the current interaction data of the target object A in 5 working days, and the current interaction data includes multiple interactive products corresponding to the time nodes. Among them, the time node unit is day. Based on the interactive products of each time node in the current interaction data, the computer device obtains a decision tree by taking the interactive products of the same time node as a layer of the decision tree. The first layer in the decision tree is that the target object purchased the interactive product with ID 1 at the Monday time node, purchased the interactive products with ID 2 and ID 3 at the Tuesday time node, purchased the interactive products with ID 4, ID 5, and ID 6 at the Wednesday time node, purchased the interactive product with ID 7 at the Thursday time node, and purchased the interactive products with ID 8 and ID 9 at the Friday time node. The computer device selects one interactive product from each layer in order from Monday to Friday for combination, and obtains 12 sequences to be processed corresponding to the target object, such as [1,2,4,7,8], [1,2,5,7,8], [1,2,6,7,8], [1,2,4,7,9], etc., a total of 12 sequences to be processed.

[0075] In this embodiment, the current interaction data corresponding to the target object is obtained, and the current interaction data includes multiple interaction products corresponding to the time nodes. A decision tree is constructed according to the time nodes corresponding to each interaction product, wherein the nodes at the same layer in the decision tree are interaction products with the same time nodes. In this way, the overall interaction situation of the current interaction data is presented more clearly and intuitively. According to the order of each time node, an interaction product is selected from each layer in turn for combination to obtain multiple sequences to be processed corresponding to the target object. In this way, based on multiple sequences to be processed carrying time sequence information, the various interaction situations of the current interaction data at multiple time nodes are clearly displayed, highlighting the information related to the time sequence in the current interaction data.

[0076] In one embodiment, the support of each group of product combinations to be processed is determined based on multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed, including: for each group of product combinations to be processed, based on multiple historical time series relationships, determining the target time series relationship between two interactive products in the corresponding product combination to be processed. For each group of product combinations to be processed, based on the target time series relationship between the two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in each group of product combinations to be processed, determining from multiple sequences to be processed a first quantity that satisfies the target time series relationship between the two interactive products in the corresponding product combination to be processed and a second quantity that does not satisfy the target time series relationship. Based on the first quantity and the second quantity corresponding to each group of product combinations to be processed, the support of each group of product combinations to be processed is determined.

[0077] Specifically, for each group of product combinations to be processed, the computer device determines two interactive products in the corresponding product combination to be processed, and determines the target timing relationship between the two interactive products in the corresponding product combination to be processed based on multiple historical timing relationships. The computer device judges the two interactive products in each group of product combinations to be processed based on multiple target timing relationships and the time nodes of the interactive products included in each group of product combinations to be processed, and determines a first quantity that satisfies the target timing relationship between the two interactive products in the corresponding product combination to be processed, and a second quantity that does not satisfy the target timing relationship. For each group of product combinations to be processed, the computer device determines the support of each group of product combinations to be processed based on the first quantity and the second quantity corresponding to each of the corresponding product combinations to be processed.

[0078] For example, for the 12 to-be-processed sequences of the target object A, such as [1,2,4,7,8], [1,2,5,7,8], [1,2,6,7,8], [1,2,4,7,9], etc. For the interactive product with ID 1 and the interactive product with ID 2, the historical time sequence relationship is that the time sequence of the interactive product with ID 1 is in front, and the time sequence of the interactive product with ID 2 is in the back. For the to-be-processed product combination 01 of the interactive product with ID 1 and the interactive product with ID 2, the computer device determines that the target sequence of the to-be-processed product combination 01 is that the time sequence of the interactive product with ID 1 is in front, and the time sequence of the interactive product with ID 2 is in the back. If there are 10 to-be-processed product combinations including the interactive product with ID 1 and the interactive product with ID 2 in the multiple to-be-processed sequences, among which the first number that meets the target time sequence is 7, and the second number that does not meet the target time sequence is 3. The computer device determines the support degree of the product combination 01 to be processed based on the first quantity and the second quantity in the product combination 01 to be processed.

[0079] In this embodiment, for each group of product combinations to be processed, the target timing relationship between two interactive products in the corresponding product combination to be processed is determined based on multiple historical timing relationships. In this way, by using the timing relationship between two interactive products in the products to be processed as a reference timing relationship, the steps in the process of determining the support of the product combination to be processed are effectively simplified. For each group of product combinations to be processed, based on the target timing relationship between the two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in each group of product combinations to be processed, the first number that satisfies the target timing relationship between the two interactive products in the corresponding product combination to be processed and the second number that does not satisfy the target timing relationship are determined from multiple sequences to be processed. Based on the first number and the second number corresponding to each group of product combinations to be processed, the support of each group of product combinations to be processed is determined. In this way, based on the support of each group of product combinations to be processed, the strength of the timing relationship between the two interactive products in each group of product combinations to be processed can be efficiently and clearly reflected.

[0080] In one embodiment, the support of each group of product combinations to be processed is determined based on the first quantity and the second quantity corresponding to each group of product combinations to be processed, including: if the second quantity of the product combinations to be processed is zero, then the support of the corresponding product combination to be processed is determined to be one numerical unit. If the second quantity of the product combinations to be processed is not zero, then the first quantity of the corresponding product combination to be processed is divided by the sum of the first quantity and the second quantity to obtain the support of the corresponding product combination to be processed.

[0081] Specifically, the computer device obtains the first quantity and the second quantity corresponding to each group of product combinations to be processed, and determines the second quantity in each product combination to be processed. If the second quantity of the product combination to be processed is zero, the support degree of the corresponding product combination to be processed is determined to be one numerical unit. If the second quantity of the product combination to be processed is not zero, the computer device divides the first quantity of the corresponding product combination to be processed by the sum of the first quantity and the second quantity to obtain the support degree of the corresponding product combination to be processed.

[0082] Among them, support is the ratio of the number of satisfied time-limited rules to the sum of satisfied and violated time-limited rules for a state expiration rule with the same attributes on the same data set, which is defined as support. The formula is as follows:

[0083]

[0084] In the above formula, Sc represents the support of rule C, Cnt(A->B) (i.e., S(c)) represents the number of cases that satisfy the invalid rule, and Cnt(B->A) (i.e., V(c)) represents the number of cases that violate the invalid rule.

[0085] For example, for the 12 to-be-processed sequences of target object A, such as [1,2,4,7,8], [1,2,5,7,8], [1,2,6,7,8], [1,2,4,7,9], etc. If there are 10 to-be-processed product combinations including an interactive product with ID 1 and an interactive product with ID 2 in multiple to-be-processed sequences, of which the first number that meets the target time sequence is 7 and the second number that does not meet the target time sequence is 3. Then the support for the to-be-processed product combination including the interactive product with ID 1 and the interactive product with ID 2 is 70%. Similar sample code is as follows:

[0086]

[0087] In this embodiment, if the second number of product combinations to be processed is zero, the support of the corresponding product combination to be processed is determined to be one numerical unit. In this way, when processing the current interactive data with a huge amount of data, by directly setting the support of the product combination to be processed with the second number of zero as one numerical unit, the amount of calculation is greatly reduced and the efficiency of data processing is improved. If the second number of product combinations to be processed is not zero, the first number of the corresponding product combination to be processed is divided by the sum of the first number and the second number to obtain the support of the corresponding product combination to be processed. In this way, based on the judgment of whether the second number of each product combination to be processed is zero, the correctness of the support calculation can be ensured in the process of reducing the amount of calculation, which greatly improves the efficiency of the support calculation for each group of combinations to be processed.

[0088] In one embodiment, the computer device obtains the first quantity and the second quantity corresponding to each group of product combinations to be processed, and determines the second quantity in each product combination to be processed. If the second quantity of the product combination to be processed is zero, the support degree of the corresponding product combination to be processed is determined to be one numerical unit. If the second quantity of the product combination to be processed is not zero, the computer device divides the second quantity of the corresponding product combination to be processed by the sum of the first quantity and the second quantity to obtain the support degree of the corresponding product combination to be processed.

[0089] In this embodiment, if the second number of product combinations to be processed is zero, the support of the corresponding product combination to be processed is determined to be one numerical unit. In this way, when processing the current interactive data with a huge amount of data, by directly setting the support of the product combination to be processed with a second number of zero as one numerical unit, the amount of calculation is greatly reduced and the efficiency of data processing is improved. If the second number of product combinations to be processed is not zero, the second number of the corresponding product combination to be processed is divided by the sum of the first number and the second number to obtain the support of the corresponding product combination to be processed. In this way, based on the judgment of whether the second number of each product combination to be processed is zero, the correctness of the support calculation can be ensured in the process of reducing the amount of calculation, and the efficiency of the support calculation for each group of combinations to be processed is greatly improved.

[0090] In one embodiment, for each sequence to be processed, based on the product combinations to be processed included in the corresponding sequence to be processed, the average support of the corresponding sequence to be processed is determined, including: for each sequence to be processed, the support of each group of product combinations to be processed included in the corresponding sequence to be processed is determined, and the sequence length of the corresponding sequence to be processed is determined. For each sequence to be processed, the sum of the supports of the corresponding sequence to be processed is divided by the sequence length of the corresponding sequence to be processed, and a unit number is subtracted to obtain the average support of the corresponding sequence to be processed.

[0091] Specifically, for each sequence to be processed, the computer device determines the multiple groups of product combinations to be processed contained in each sequence to be processed, and determines the support of the multiple groups of product combinations to be processed contained in each sequence to be processed. The computer device obtains the sequence length of each sequence to be processed. For each sequence to be processed, the computer device divides the sum of the support of the corresponding sequence to be processed by the sequence length of the corresponding sequence to be processed, and subtracts a unit number to obtain the average support of the corresponding sequence to be processed. Similar sample code is as follows:

[0092]

[0093]

[0094] In this embodiment, for each sequence to be processed, the support of each group of product combinations to be processed contained in the corresponding sequence to be processed is determined, and the sequence length of the corresponding sequence to be processed is determined. For each sequence to be processed, the sum of the supports of the corresponding sequence to be processed is divided by the sequence length of the corresponding sequence to be processed, and a unit number is subtracted to obtain the average support of the corresponding sequence to be processed, so that the temporal relationship represented by each sequence to be processed can be accurately and specifically determined, and then the strength of the temporal relationship of each sequence to be processed can be determined.

[0095] In one embodiment, based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, including: comparing the average support of each sequence to be processed with a timeliness threshold, and taking the sequence to be processed whose average support is greater than or equal to the timeliness threshold as the target sequence.

[0096] Specifically, the computer device obtains the average support and timeliness threshold of each sequence to be processed, and compares each sequence to be processed with the timeliness threshold. Based on the comparison results of each sequence to be processed, the computer device takes the sequence to be processed whose average support is greater than or equal to the timeliness threshold as the target sequence.

[0097] In this embodiment, the average support of each sequence to be processed is compared with the timeliness threshold, and the sequence to be processed whose average support is greater than or equal to the timeliness threshold is used as the target sequence. In this way, through the temporal relationship of each interactive product in the target sequence with strong timeliness conditions, interactive products can be reasonably and accurately recommended to the target object, greatly improving the effectiveness of interactive product recommendation.

[0098] In order to more clearly understand the technical solution of the present application, a more detailed embodiment is provided for description. Figure 4As shown, the computer device selects the initial historical interaction data (data file) corresponding to the target object, and determines the type of the interaction product by traversing the interaction products in the initial historical interaction data. The computer device determines the identity (ID) of each interaction product in turn according to the type of the interaction product, and obtains the historical interaction data. The historical interaction data contains multiple groups of product combinations, and each group of product combinations consists of two adjacent interaction products. The computer device obtains the historical interaction data, and determines each group of product combinations based on the historical interaction data. The computer device determines the historical time sequence relationship between the two interaction products in each group of product combinations based on each group of product combinations. The computer device obtains the current interaction data of the target object in the target time period, and the current interaction data includes multiple interaction products corresponding to the time nodes. Based on the interaction products of each time node in the current interaction data, the computer device uses the interaction products of the same time node as a layer of the decision tree to obtain a decision tree. Among them, the time nodes corresponding to different layers in the decision tree are different. Based on the decision tree, the computer device selects an interaction product from each layer in turn according to the order of each time node for combination, and obtains multiple sequences to be processed corresponding to the target object. The computer device obtains multiple sequences to be processed, and determines each group of product combinations to be processed from the multiple sequences to be processed. For each group of product combinations to be processed, the computer device determines two interactive products in the corresponding product combination to be processed, and determines the target time sequence relationship between the two interactive products in the corresponding product combination to be processed based on multiple historical time sequence relationships. The computer device judges the two interactive products in each group of product combinations to be processed based on multiple target time sequence relationships and the time nodes of the interactive products included in each group of product combinations to be processed, and determines the first quantity that satisfies the target time sequence relationship between the two interactive products in the corresponding product combination to be processed, and the second quantity that does not satisfy the target time sequence relationship. The computer device judges the second quantity in each product combination to be processed. If the second quantity of the product combination to be processed is zero, the support of the corresponding product combination to be processed is determined to be one numerical unit. If the second quantity of the product combination to be processed is not zero, the computer device divides the first quantity of the corresponding product combination to be processed by the sum of the first quantity and the second quantity to obtain the support of the corresponding product combination to be processed. For each sequence to be processed, the computer device determines the multiple groups of product combinations to be processed contained in each sequence to be processed, and determines the support of the multiple groups of product combinations to be processed contained in each sequence to be processed. The computer device obtains the sequence length of each sequence to be processed. For each sequence to be processed, the computer device divides the sum of the support of the corresponding sequence to be processed by the sequence length of the corresponding sequence to be processed, and subtracts a unit number to obtain the average support of the corresponding sequence to be processed. The computer device obtains the timeliness threshold S (S = 0.85) and determines whether each sequence to be processed is greater than or equal to 0.85.Among them, the timeliness rule of the sequence to be processed with a value less than 0.85 is not obvious, and the timeliness rule of the sequence to be processed with a value greater than or equal to 0.85 is obvious. The computer device uses the sequence to be processed with a value greater than or equal to 0.85 as the target sequence. Each interactive product in the target sequence is used to recommend to the terminal corresponding to the target object according to the corresponding time sequence.

[0099] It should be noted that the technical solution of the present application can be applied to other scenarios, such as the scenario of purchasing financial products or the scenario of purchasing insurance products.

[0100] In this embodiment, by acquiring historical interaction data corresponding to the target object, the historical interaction data includes multiple groups of product combinations, each group of product combinations consists of two adjacent interactive products. The historical time series relationship between the two interactive products in each group of product combinations in the historical interaction data is determined. By screening multiple sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed includes interactive products corresponding to multiple time nodes, and the multiple interaction situations of the current interactive data in multiple time nodes can be clarified. Based on multiple sequences to be processed, each group of product combinations to be processed is determined, and based on multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed, the support of each group of product combinations to be processed is determined, thereby clearly reflecting the strength of the time series relationship between the two interactive products in each group of product combinations to be processed. For each sequence to be processed, based on the product combinations to be processed included in the corresponding sequence to be processed, the average support of the corresponding sequence to be processed is determined, thereby accurately and specifically determining the time series relationship represented by each sequence to be processed, and then determining the strength of the time series relationship of each sequence to be processed. Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to recommend to the terminal corresponding to the target object in the corresponding time sequence. Therefore, based on the time sequence relationship of each interactive product in the target sequence with strong timeliness condition, interactive products can be reasonably and accurately recommended to the target object, greatly improving the effectiveness of interactive product recommendation.

[0101] In addition, based on the target sequence, the interactive data of the target object can be finely managed, such as personalized recommendation, that is, personalized recommendation of services or bank products according to the user's preferences, such as first dividing the users according to their age, asset size, financial management preferences, etc., accurately locating the user group, and then filtering out the relevant data, and analyzing their potential financial service needs after timeliness processing, and then targeted marketing promotion. Or, for example, user life cycle management, that is, user life cycle management includes new user acquisition, user churn prevention and user win-back, etc. For example, for interactive insurance products, according to the purchase tendency of users of the same type for insurance products, high-yield financial products are sold to the top 20% of customers with churn rate to retain them, thereby reducing the customer churn rate. Among them, the target object can be a target user.

[0102] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0103] Based on the same inventive concept, the embodiment of the present application also provides an interactive product recommendation device for implementing the interactive product recommendation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more interactive product recommendation device embodiments provided below can refer to the limitations of the interactive product recommendation method above, and will not be repeated here.

[0104] In one embodiment, Figure 5 As shown, an interactive product recommendation device is provided, the device 500 includes: an acquisition module 502, a first determination module 504, a screening module 506, a second determination module 508, a third determination module 510 and a fourth determination module 512, wherein:

[0105] The acquisition module 502 is used to acquire the historical interaction data corresponding to the target object, wherein the historical interaction data includes a plurality of product combinations, and each product combination consists of two adjacent interactive products.

[0106] The first determining module 504 is used to determine the historical time sequence relationship between two interactive products in each product combination in the historical interaction data.

[0107] The screening module 506 is used to screen out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of which contains interaction products corresponding to a plurality of time nodes.

[0108] The second determination module 508 is used to determine each group of product combinations to be processed based on multiple sequences to be processed, and determine the support level of each group of product combinations to be processed based on multiple historical time series relationships and time nodes of interactive products included in each group of product combinations to be processed.

[0109] The third determination module 510 is used to determine, for each sequence to be processed, an average support degree of the corresponding sequence to be processed based on the combination of products to be processed included in the corresponding sequence to be processed.

[0110] The fourth determination module 512 is used to determine a target sequence that meets the strong timeliness condition from multiple sequences to be processed based on the average support of each sequence to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object in a corresponding time sequence.

[0111] In one embodiment, the screening module 506 is used to obtain current interaction data corresponding to the target object, wherein the current interaction data includes multiple interaction products corresponding to time nodes. A decision tree is constructed according to the time nodes corresponding to each interaction product, wherein nodes at the same layer in the decision tree are interaction products with the same time nodes. According to the order of each time node, one interaction product is selected from each layer in turn for combination to obtain multiple sequences to be processed corresponding to the target object.

[0112] In one embodiment, the second determination module 508 is used to determine, for each group of product combinations to be processed, a target timing relationship between two interactive products in the corresponding product combination to be processed based on a plurality of the historical timing relationships. For each group of product combinations to be processed, based on the target timing relationship between the two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in each group of product combinations to be processed, determine from a plurality of sequences to be processed a first quantity that satisfies the target timing relationship between the two interactive products in the corresponding product combination to be processed and a second quantity that does not satisfy the target timing relationship. Based on the first quantity and the second quantity corresponding to each group of product combinations to be processed, the support of each group of product combinations to be processed is determined.

[0113] In one embodiment, the second determination module 508 is configured to determine the support of the corresponding product combination to be processed as one numerical unit if the second number of the product combination to be processed is zero. If the second number of the product combination to be processed is not zero, the first number of the corresponding product combination to be processed is divided by the sum of the first number and the second number to obtain the support of the corresponding product combination to be processed.

[0114] In one embodiment, the third determination module 510 is used to determine the support of each group of product combinations to be processed contained in the corresponding sequence to be processed, and determine the sequence length of the corresponding sequence to be processed. For each sequence to be processed, the sum of the supports of the corresponding sequence to be processed is divided by the sequence length of the corresponding sequence to be processed, and a unit number is subtracted to obtain the average support of the corresponding sequence to be processed.

[0115] In one embodiment, the fourth determination module 512 is used to compare the average support of each sequence to be processed with the timeliness threshold, and take the sequence to be processed whose average support is greater than or equal to the timeliness threshold as the target sequence.

[0116] Each module in the above interactive product recommendation device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0117] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store interactive product recommendation data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an interactive product recommendation method is implemented.

[0118] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0119] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: obtaining historical interaction data corresponding to a target object, wherein the historical interaction data includes multiple groups of product combinations, each group of product combinations consisting of two adjacent interactive products. Determining the historical time sequence relationship between two interactive products in each group of product combinations in the historical interaction data. Filtering out multiple sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed includes interactive products corresponding to multiple time nodes respectively. Based on the multiple sequences to be processed, determining each group of product combinations to be processed, and determining the support of each group of product combinations to be processed based on the multiple historical time sequence relationships and the time nodes of the interactive products included in each group of product combinations to be processed. For each sequence to be processed, determining the average support of the corresponding sequence to be processed based on the product combinations to be processed included in the corresponding sequence to be processed. Based on the average support of each sequence to be processed, determining a target sequence that meets a strong timeliness condition from the multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object according to the corresponding time sequence.

[0120] In one embodiment, when the processor executes the computer program, the following steps are further implemented: obtaining current interaction data corresponding to the target object, wherein the current interaction data includes multiple interaction products corresponding to time nodes. Constructing a decision tree according to the time nodes corresponding to each interaction product, wherein nodes at the same layer in the decision tree are interaction products with the same time nodes. According to the order of each time node, selecting one interaction product from each layer in turn for combination, and obtaining multiple sequences to be processed corresponding to the target object.

[0121] In one embodiment, when the processor executes the computer program, the following steps are also implemented: for each group of product combinations to be processed, based on multiple historical time series relationships, determine the target time series relationship between two interactive products in the corresponding product combination to be processed. For each group of product combinations to be processed, based on the target time series relationship between two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in each group of product combinations to be processed, determine from multiple sequences to be processed a first quantity that satisfies the target time series relationship between two interactive products in the corresponding product combination to be processed and a second quantity that does not satisfy the target time series relationship. Based on the first quantity and the second quantity corresponding to each group of product combinations to be processed, determine the support of each group of product combinations to be processed.

[0122] In one embodiment, when the processor executes the computer program, the following steps are further implemented: if the second number of the product combination to be processed is zero, the support degree of the corresponding product combination to be processed is determined to be one numerical unit. If the second number of the product combination to be processed is not zero, the first number of the corresponding product combination to be processed is divided by the sum of the first number and the second number to obtain the support degree of the corresponding product combination to be processed.

[0123] In one embodiment, when the processor executes the computer program, the following steps are further implemented: for each sequence to be processed, the support of each group of product combinations to be processed contained in the corresponding sequence to be processed is determined, and the sequence length of the corresponding sequence to be processed is determined. For each sequence to be processed, the sum of the supports of the corresponding sequence to be processed is divided by the sequence length of the corresponding sequence to be processed, and a unit number is subtracted to obtain the average support of the corresponding sequence to be processed.

[0124] In one embodiment, when the processor executes the computer program, the following steps are also implemented: comparing the average support of each sequence to be processed with a timeliness threshold, and taking the sequence to be processed whose average support is greater than or equal to the timeliness threshold as the target sequence.

[0125] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: historical interaction data corresponding to a target object is obtained, wherein the historical interaction data includes multiple groups of product combinations, each group of product combinations consisting of two adjacent interactive products. The historical time sequence relationship between two interactive products in each group of product combinations in the historical interaction data is determined. Multiple sequences to be processed are screened out from the current interaction data corresponding to the target object, each of which includes interactive products corresponding to multiple time nodes. Based on the multiple sequences to be processed, each group of product combinations to be processed is determined, and based on the multiple historical time sequence relationships and the time nodes of the interactive products included in each group of product combinations to be processed, the support of each group of product combinations to be processed is determined. For each sequence to be processed, the average support of the corresponding sequence to be processed is determined based on the product combinations to be processed included in the corresponding sequence to be processed. Based on the average support of each sequence to be processed, a target sequence that meets a strong timeliness condition is determined from the multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object according to the corresponding time sequence.

[0126] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining current interaction data corresponding to the target object, wherein the current interaction data includes multiple interaction products corresponding to time nodes. Constructing a decision tree according to the time nodes corresponding to each interaction product, wherein nodes at the same layer in the decision tree are interaction products with the same time nodes. According to the order of each time node, selecting one interaction product from each layer in turn for combination, and obtaining multiple sequences to be processed corresponding to the target object.

[0127] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: for each group of product combinations to be processed, based on multiple historical time series relationships, determine the target time series relationship between two interactive products in the corresponding product combination to be processed. For each group of product combinations to be processed, based on the target time series relationship between two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in each group of product combinations to be processed, determine from multiple sequences to be processed a first quantity that satisfies the target time series relationship between two interactive products in the corresponding product combination to be processed and a second quantity that does not satisfy the target time series relationship. Based on the first quantity and the second quantity corresponding to each group of product combinations to be processed, determine the support of each group of product combinations to be processed.

[0128] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: if the second number of the product combination to be processed is zero, the support degree of the corresponding product combination to be processed is determined to be one numerical unit. If the second number of the product combination to be processed is not zero, the first number of the corresponding product combination to be processed is divided by the sum of the first number and the second number to obtain the support degree of the corresponding product combination to be processed.

[0129] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: for each sequence to be processed, the support of each group of product combinations to be processed contained in the corresponding sequence to be processed is determined, and the sequence length of the corresponding sequence to be processed is determined. For each sequence to be processed, the sum of the supports of the corresponding sequence to be processed is divided by the sequence length of the corresponding sequence to be processed, and a unit number is subtracted to obtain the average support of the corresponding sequence to be processed.

[0130] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: the average support of each sequence to be processed is compared with the timeliness threshold, and the sequence to be processed whose average support is greater than or equal to the timeliness threshold is used as the target sequence.

[0131] In one embodiment, a computer program product is provided, including a computer program, which implements the following steps when executed by a processor: obtaining historical interaction data corresponding to a target object, wherein the historical interaction data includes multiple groups of product combinations, each group of product combinations consisting of two adjacent interactive products. Determining the historical time sequence relationship between two interactive products in each group of product combinations in the historical interaction data. Filtering out multiple sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed includes interactive products corresponding to multiple time nodes. Based on the multiple sequences to be processed, determining each group of product combinations to be processed, and based on the multiple historical time sequence relationships and the time nodes of the interactive products included in each group of product combinations to be processed, determining the support of each group of product combinations to be processed. For each sequence to be processed, determining the average support of the corresponding sequence to be processed based on the product combinations to be processed included in the corresponding sequence to be processed. Based on the average support of each sequence to be processed, determining a target sequence that meets a strong timeliness condition from the multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object according to the corresponding time sequence.

[0132] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining current interaction data corresponding to the target object, wherein the current interaction data includes multiple interaction products corresponding to time nodes. Constructing a decision tree according to the time nodes corresponding to each interaction product, wherein nodes at the same layer in the decision tree are interaction products with the same time nodes. According to the order of each time node, selecting one interaction product from each layer in turn for combination, and obtaining multiple sequences to be processed corresponding to the target object.

[0133] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: for each group of product combinations to be processed, based on multiple historical time series relationships, determine the target time series relationship between two interactive products in the corresponding product combination to be processed. For each group of product combinations to be processed, based on the target time series relationship between two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in each group of product combinations to be processed, determine from multiple sequences to be processed a first quantity that satisfies the target time series relationship between two interactive products in the corresponding product combination to be processed and a second quantity that does not satisfy the target time series relationship. Based on the first quantity and the second quantity corresponding to each group of product combinations to be processed, determine the support of each group of product combinations to be processed.

[0134] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: if the second number of the product combination to be processed is zero, the support degree of the corresponding product combination to be processed is determined to be one numerical unit. If the second number of the product combination to be processed is not zero, the first number of the corresponding product combination to be processed is divided by the sum of the first number and the second number to obtain the support degree of the corresponding product combination to be processed.

[0135] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: for each sequence to be processed, the support of each group of product combinations to be processed contained in the corresponding sequence to be processed is determined, and the sequence length of the corresponding sequence to be processed is determined. For each sequence to be processed, the sum of the supports of the corresponding sequence to be processed is divided by the sequence length of the corresponding sequence to be processed, and a unit number is subtracted to obtain the average support of the corresponding sequence to be processed.

[0136] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: the average support of each sequence to be processed is compared with the timeliness threshold, and the sequence to be processed whose average support is greater than or equal to the timeliness threshold is used as the target sequence.

[0137] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Among them, for the push of interactive products or advertisements, users can refuse or conveniently refuse the push of interactive products, or refuse the push of advertisements, etc.

[0138] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0139] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0140] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An interactive product recommendation method, characterized in that: The method comprises: Acquire historical interaction data corresponding to the target object, wherein the historical interaction data includes multiple product combinations, each product combination consisting of two adjacent interactive products; Determine a historical temporal relationship between two interactive products in each product combination in the historical interaction data; Filtering out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of the sequences to be processed being a combination of interaction products at each time node in the target time period; Determine each group of product combinations to be processed based on the multiple sequences to be processed, and determine the support level of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed; For each sequence to be processed, based on the support of the corresponding sequence to be processed and the combination of products to be processed contained in the corresponding sequence to be processed, determine the average support of the corresponding sequence to be processed; Based on the average support of each sequence to be processed, a target sequence that meets the strong timeliness condition is determined from multiple sequences to be processed, and each interactive product in the target sequence is used to be recommended to the terminal corresponding to the target object according to the corresponding time sequence; The determining the support level of each group of product combinations to be processed based on the plurality of historical time sequence relationships and the time nodes of the interactive products included in each group of product combinations to be processed includes: For each group of product combinations to be processed, based on the plurality of historical time series relationships, determining a target time series relationship between two interactive products in the corresponding product combination to be processed; For each group of product combinations to be processed, based on the target timing relationship between two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in the corresponding product combination to be processed, determine from multiple sequences to be processed a first quantity that satisfies the target timing relationship between the two interactive products in the corresponding product combination to be processed and a second quantity that does not satisfy the target timing relationship; The support degree of each group of product combinations to be processed is determined based on the first quantity and the second quantity corresponding to each group of product combinations to be processed.

2. The method according to claim 1, characterized in that The step of selecting a plurality of sequences to be processed from the current interaction data corresponding to the target object includes: Acquire current interaction data corresponding to the target object, wherein the current interaction data includes a plurality of interaction products corresponding to time nodes; Constructing a decision tree according to the time nodes corresponding to each interactive product, wherein the nodes at the same layer in the decision tree are interactive products with the same time node; According to the order of each time node, one interactive product is selected from each layer in turn for combination to obtain a plurality of sequences to be processed corresponding to the target object.

3. The method according to claim 1, characterized in that The step of determining the support level of each group of product combinations to be processed based on the first quantity and the second quantity respectively corresponding to each group of product combinations to be processed includes: If the second number of the product combination to be processed is zero, the support degree of the corresponding product combination to be processed is determined to be one numerical unit; If the second quantity of the product combination to be processed is not zero, the first quantity of the corresponding product combination to be processed is divided by the sum of the first quantity and the second quantity to obtain the support degree of the corresponding product combination to be processed.

4. The method according to claim 1, characterized in that: For each sequence to be processed, based on the combination of products to be processed included in the corresponding sequence to be processed, determining the average support of the corresponding sequence to be processed includes: For each sequence to be processed, determine the support of each group of product combinations to be processed contained in the corresponding sequence to be processed, and determine the sequence length of the corresponding sequence to be processed; For each sequence to be processed, the sum of the supports of the corresponding sequence to be processed is divided by the sequence length of the corresponding sequence to be processed, and a unit number is subtracted to obtain the average support of the corresponding sequence to be processed.

5. The method according to any one of claims 1 to 4, characterized in that The step of determining a target sequence satisfying a strong timeliness condition from a plurality of sequences to be processed based on the average support of each sequence to be processed includes: The average support of each sequence to be processed is compared with the timeliness threshold, and the sequence to be processed whose average support is greater than or equal to the timeliness threshold is taken as the target sequence.

6. An interactive product recommendation device, characterized in that: The device comprises: An acquisition module, used to acquire historical interaction data corresponding to a target object, wherein the historical interaction data includes a plurality of product combinations, each of which consists of two adjacent interactive products; A first determination module is used to determine the historical time sequence relationship between two interactive products in each product combination in the historical interaction data; A screening module, used to screen out a plurality of sequences to be processed from the current interaction data corresponding to the target object, each of which is a combination of interaction products at each time node in the target time period; A second determination module is used to determine each group of product combinations to be processed based on the multiple sequences to be processed, and determine the support degree of each group of product combinations to be processed based on the multiple historical time series relationships and the time nodes of the interactive products included in each group of product combinations to be processed; A third determination module is used to determine, for each sequence to be processed, an average support of the corresponding sequence to be processed based on the support of the corresponding sequence to be processed and the combination of products to be processed contained in the corresponding sequence to be processed; A fourth determination module is used to determine a target sequence that meets a strong timeliness condition from multiple sequences to be processed based on the average support of each sequence to be processed, wherein each interactive product in the target sequence is used to be recommended to a terminal corresponding to the target object in a corresponding time sequence; The second determination module is specifically used to determine, for each group of product combinations to be processed, a target timing relationship between two interactive products in the corresponding product combination to be processed based on multiple historical timing relationships; for each group of product combinations to be processed, based on the target timing relationship between the two interactive products in the corresponding product combination to be processed and the time nodes of the interactive products included in the corresponding product combination to be processed, determine from multiple sequences to be processed a first quantity that satisfies the target timing relationship between the two interactive products in the corresponding product combination to be processed and a second quantity that does not satisfy the target timing relationship; determine the support of each group of product combinations to be processed based on the first quantity and the second quantity corresponding to each group of product combinations to be processed.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images