Search method and device, terminal and storage medium

Abstract

This application discloses a search method, apparatus, terminal, and storage medium. The method includes: acquiring user historical behavior data and the time period corresponding to the user historical behavior data; if the time period is less than a first preset time period, determining a first search result based on the user historical behavior data, preset product attributes, and current search information; if the time period is greater than or equal to the first preset time period, determining a second search result based on the user historical behavior data, current search information, and a target search model. This invention uses different search methods to determine search results based on different time periods corresponding to historical behavior data to cater to different customers. Furthermore, both the first and second search results are optimized using product attributes to improve search efficiency and thus enhance search effectiveness.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and more specifically, to a search method, apparatus, terminal, and storage medium. Background Technology

[0002] With the rapid development of the internet, search has become widely used. Not only do large companies like Google and Baidu, which originated from search, attach great importance to search, but e-commerce platforms like Alibaba and JD.com also place great emphasis on it, aiming to provide users with a very user-friendly search experience. In search scenarios, different users may return the same results for the same search term, but in reality, each user's intent when searching for the same term is different. For example, when a user searches for "coat," if we know from the user's historical behavior that the user prefers "black," we can rank black "coats" higher in the results, thus increasing the probability that the user will choose the product they prefer.

[0003] Currently, to effectively capture users' interests and preferences and return results that users are more likely to want based on their historical preferences, there are two main approaches: The first approach is machine learning models: Markov models, which generate a state transition matrix by observing users' short-term behavioral data, and predict user behavior at the next time point based on this matrix, effectively utilizing users' short-term interest and preference information; The second approach is deep learning models: classic YouTube models, as well as Alibaba's DIN network, etc. These models all take into account users' historical click behavior, historical search behavior, and user profile features.

[0004] However, neither of the above methods is applicable to searches in the field of digital asset management platforms. Summary of the Invention

[0005] The main objective of this application is to provide a search method, apparatus, terminal, and storage medium to address the problem that search methods in related technologies are not applicable to the field of digital asset management platforms.

[0006] To achieve the above objectives, firstly, this application provides a search method, comprising:

[0007] Obtain user historical behavior data and the corresponding time periods for user historical behavior data;

[0008] If the time period is shorter than the first preset time period, the first search result is determined based on the user's historical behavior data, preset product attributes, and current search information;

[0009] If the time period is greater than or equal to the first preset time period, the second search result is determined based on the user's historical behavior data, current search information, and target search model.

[0010] In one possible implementation, if the time period is shorter than a first preset time period, the first search result is determined based on the user's historical behavior data, preset product attributes, and current search information, including:

[0011] If the time period is shorter than the first preset time period, the initial search results are determined based on the user's historical behavior data and preset product attributes;

[0012] The initial search results are rearranged using a first matrix to obtain intermediate search results, wherein the first matrix is ​​determined by the user's historical behavior data and preset product attributes;

[0013] Based on the intermediate search results and the current search information, determine the target search result;

[0014] Based on the target search results, determine the search evaluation metrics;

[0015] If the search evaluation index meets the preset threshold, the target search result is used as the first search result.

[0016] In one possible implementation, determining the target search result based on the intermediate search results and the current search information includes:

[0017] Based on the intermediate search results and the product attributes corresponding to the current search information, a second matrix is ​​determined;

[0018] The first matrix and the second matrix are operated on to obtain the third matrix;

[0019] The intermediate search results are rearranged using the third matrix to obtain the target search result.

[0020] In one possible implementation, after determining the search evaluation metrics based on the target search results, the method further includes:

[0021] If the search evaluation metrics do not meet the first preset threshold, the user's historical behavior data and preset product attributes are adjusted to obtain the target user's historical behavior data and target product attributes.

[0022] In one possible implementation, if the time period is greater than or equal to a first preset time period, the second search result is determined based on the user's historical behavior data, current search information, and the target search model, including:

[0023] If the time period is greater than or equal to the first preset time period, the target search model is determined;

[0024] Input the user's historical behavior data and current search information into the target search model to obtain the second search result.

[0025] In one possible implementation, the target search model is determined, including:

[0026] Obtain the initial search model;

[0027] Input the user's historical behavior data into the initial search model, and obtain the target search model when the number of iterations reaches the second preset threshold.

[0028] In one possible implementation, user historical behavior data is input into the initial search model. Once the number of iterations reaches a second preset threshold, a target search model is obtained, including:

[0029] Acquire historical search data for a second preset time period and historical user behavior data for a third preset time period;

[0030] Historical search data from the second preset time period and user historical behavior data from the third preset time period are input into the initial search model. The target search model is obtained when the number of iterations reaches the second preset threshold. The user historical behavior data includes historical search data and historical clicked product data, where historical clicked product data is represented by product attributes.

[0031] Secondly, embodiments of the present invention provide a search device, comprising:

[0032] The data acquisition module is used to acquire user historical behavior data and the time periods corresponding to the user historical behavior data;

[0033] The first result determination module is used to determine the first search result based on the user's historical behavior data, preset product attributes, and current search information if the time period is less than the first preset time period.

[0034] The second result determination module is used to determine the second search result based on the user's historical behavior data, current search information, and target search model if the time period is greater than or equal to the first preset time period.

[0035] Thirdly, embodiments of the present invention provide a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the search methods described above.

[0036] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the search methods described above.

[0037] This invention provides a search method, apparatus, terminal, and storage medium, comprising: firstly acquiring user historical behavior data and the corresponding time period; if the time period is less than a first preset time period, determining a first search result based on the user historical behavior data, preset product attributes, and current search information; and if the time period is greater than or equal to the first preset time period, determining a second search result based on the user historical behavior data, current search information, and a target search model. This invention uses different search methods to determine search results based on different time periods corresponding to historical behavior data to cater to different customers. Furthermore, product attributes are used to evaluate user preferences in determining both the first and second search results, thereby improving search efficiency and ultimately enhancing search effectiveness. Attached Figure Description

[0038] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings:

[0039] Figure 1 This is a flowchart illustrating the implementation of a search method provided in an embodiment of the present invention;

[0040] Figure 2 This is a schematic diagram of the structure of a search device provided in an embodiment of the present invention;

[0041] Figure 3 This is a schematic diagram of the terminal provided in an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in sequences other than those illustrated or described herein.

[0044] It should be understood that in the various embodiments of the present invention, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0045] It should be understood that in this invention, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or device.

[0046] It should be understood that in this invention, "multiple" refers to two or more. "And / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, "and / or B" can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "Contains A, B, and C", "Contains A, B, and C" means that all three A, B, and C are contained; "Contains A, B, or C" means that one of A, B, and C is contained; "Contains A, B, and / or C" means that any one, two, or three of A, B, and C are contained.

[0047] It should be understood that in this invention, "B corresponding to A", "B corresponding to A", "A and B correspond", or "B and A correspond" means that B is associated with A, and B can be determined based on A. Determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information. Matching A and B is defined as a similarity between A and B that is greater than or equal to a preset threshold.

[0048] Depending on the context, "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection."

[0049] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0050] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0051] In one embodiment, such as Figure 1 As shown, a search method is provided, including the following steps:

[0052] Step S101: Obtain user historical behavior data and the time period corresponding to the user historical behavior data;

[0053] Step S102: If the time period is less than the first preset time period, determine the first search result based on the user's historical behavior data, preset product attributes and current search information;

[0054] Step S103: If the time period is greater than or equal to the first preset time period, determine the second search result based on the user's historical behavior data, current search information and target search model.

[0055] Specifically, the first preset time period refers to the length of time a user spends searching using the search engine, as set by the system. This invention provides different search methods for users with different time periods and determines search results based on users' historical behavior data to adapt to different user needs and improve user experience.

[0056] Given a first search result, this invention derives an initial search result based on preset product attributes and user historical behavior (search or purchase behavior), where the preset product attributes are the attributes of the products corresponding to the user's historical behavior. Then, the initial search results are rearranged using a first matrix (i.e., a user feature matrix) to obtain intermediate search results. The target search result is then determined using the current search information and the intermediate search results. The search evaluation index mAP is then determined using the target search result, and it is judged whether the search evaluation index mAP has improved. If the search evaluation index mAP has improved, the target search result is taken as the first search result.

[0057] When determining the second search result, given the large amount of user historical behavior data, a model prediction method can be directly employed. Specifically, an initial search model is first selected, then trained based on user historical behavior data to determine the target search model. Finally, the user historical behavior data and current search information are input into the target search model to determine the second search result.

[0058] This invention provides a search method, comprising: first, acquiring user historical behavior data and the corresponding time period; if the time period is less than a first preset time period, determining a first search result based on the user historical behavior data, preset product attributes, and current search information; if the time period is greater than or equal to the first preset time period, determining a second search result based on the user historical behavior data, current search information, and a target search model. This invention uses different search methods to determine search results based on different time periods corresponding to historical behavior data, adapting to different customers. Furthermore, product attributes are used to evaluate user preferences in determining both the first and second search results, thereby improving search efficiency and ultimately enhancing search effectiveness.

[0059] In one embodiment, step S102 includes:

[0060] Step S201: If the time period is shorter than the first preset time period, determine the initial search results based on the user's historical behavior data and preset product attributes.

[0061] Specifically, there are two main ways to select user historical behavior data: The first is to directly specify a quantity, which requires a high level of exploratory data analysis, such as selecting 100 items. However, this approach has some drawbacks: firstly, historical behavior from a distant period has less relevance to the present; secondly, different users have different historical behavior counts—some users might click 100 times in an hour, while others might only click 100 times in a month. The second approach is to select user historical behavior data within a preset time frame, such as selecting items (materials, etc.) clicked within the past week.

[0062] Step S202: Rearrange the initial search results using the first matrix to obtain intermediate search results.

[0063] The first matrix is ​​determined by the user's historical behavior data and preset product attributes, and the elements in the first matrix are composed of the first score of the product attribute corresponding to the user's historical behavior data and the second score of the product attribute corresponding to the product.

[0064] Furthermore, the first score for the product attribute corresponding to the user's historical behavior data and the second score for the product attribute corresponding to the product are calculated based on the product (materials, etc.) attributes, thus summarizing the user's historical behavior and product attributes. Specifically, for user historical behavior statistics: if a user has previously clicked (or purchased) a black men's jacket, then the user's score for the "black" and "men's" attributes increases by 1, and so on; for product attributes: if a product (materials, etc.) has both "black" and "men's" attributes, then the product's score for these two attributes is 1; if it does not have these attributes, then the score is 0.

[0065] Step S203: Determine the target search result based on the intermediate search results and the current search information.

[0066] Specifically, determining the target search result requires first determining a second matrix based on the intermediate search results and the product attributes corresponding to the current search information, then performing operations on the first matrix and the second matrix to obtain a third matrix, and finally using the third matrix to rearrange the intermediate search results to obtain the target search result.

[0067] Step S204: Based on the target search results, determine the search evaluation indicators;

[0068] Step S205: If the search evaluation index meets the first preset threshold, the target search result is taken as the first search result;

[0069] Step S206: If the search evaluation index does not meet the first preset threshold, adjust the user historical behavior data and the preset product attributes to obtain the target user historical behavior data and the target product attributes.

[0070] The first preset threshold refers to a pre-set threshold, and the specific value is not limited here. It can be adjusted according to different needs.

[0071] In one possible implementation, step S103 includes:

[0072] Step S301: If the time period is greater than or equal to the first preset time period, determine the target search model.

[0073] Determining the target search model requires first obtaining an initial search model, then inputting user historical behavior data into the initial search model, and finally obtaining the target search model when the number of iterations reaches a second preset threshold.

[0074] Furthermore, to determine the target search model, it is necessary to first obtain historical search data for a second preset time period and user historical behavior data for a third preset time period. Then, input the historical search data for the second preset time period and the user historical behavior data for the third preset time period into the initial search model. After the number of iterations reaches a second preset threshold, the target search model is obtained. The user historical behavior data includes historical search data and historical clicked product data, where historical clicked product data is represented by product attributes.

[0075] The process of determining the target search model is illustrated in the following specific embodiment: For example, we will use the user's historical behavior data from the previous 0-14 days for training. First, we will use the data from the previous 4-14 days as the user's historical behavior data, thus obtaining the user's historical clicked product data (watch vector) and historical search data (search vector). Then, the user's historical search data (i.e., search terms) from the previous 0-3 days will be simulated as the user's current search data (currentsearch vector). The initial search model outputs the model's predicted result (label) at the final end. The true label is the user's actual data click results from the previous 0-3 days. After loss optimization iteration, a target search model based on the previous 14 days is constructed. When the user searches on the current day, we can use this target search model for inference.

[0076] Step S302: Input the user's historical behavior data and current search information into the target search model to obtain the second search result.

[0077] Based on the above specific embodiments, the user's historical behavior data from the previous 14 days is combined with the user's new search terms for the day (i.e., the current search information) and the target search results to jointly infer what result we should return to the user on that day (i.e., the second search result).

[0078] Furthermore, the advantages of the search model used in this invention are as follows: through the attention in the Transformer Block, the model can automatically learn which historical behavioral features are of reference value for current search optimization, without the need for manual adjustment of the number and attributes of historical behaviors; adding a zero vector to the user's historical data can effectively eliminate potential noise because the user's historical behavior may be irrelevant to the current search; and adding historical search terms is also a good source of historical behavioral data.

[0079] This invention combines the current state of search optimization in the production field and implements two methods—statistical method and Tezan vector search optimization—within our company. In the early stages, without a large amount of user historical behavior, directly using a model is not effective. It is more suitable to analyze user historical behavior to roughly know which products (materials, etc.) users prefer. By incorporating product (material, etc.) attributes into the analysis, we can more accurately grasp the user profile and ultimately rank the user's preferred materials at the top. In the later stages, after accumulating a certain amount of user historical behavior, we use a deep learning model. The model can learn more information, and many parameters can be learned by the model without manual adjustment.

[0080] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0081] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0082] Figure 2 The diagram illustrates a search device according to an embodiment of the present invention. For ease of explanation, only the parts relevant to the embodiment are shown. The search device includes a data acquisition module 21, a first result determination module 22, and a second result determination module 23, as detailed below:

[0083] Data acquisition module 21 is used to acquire user historical behavior data and the time period corresponding to the user historical behavior data;

[0084] The first result determination module 22 is used to determine the first search result based on user historical behavior data, preset product attributes and current search information if the time period is less than the first preset time period.

[0085] The second result determination module 23 is used to determine the second search result based on the user's historical behavior data, current search information and target search model if the time period is greater than or equal to the first preset time period.

[0086] In one possible implementation, the first result determination module 22 includes:

[0087] The initial search result determination submodule is used to determine the initial search results based on user historical behavior data and preset product attributes if the time period is less than the first preset time period.

[0088] The intermediate search result determination submodule is used to rearrange the initial search results using a first matrix to obtain intermediate search results, wherein the first matrix is ​​determined by the user's historical behavior data and preset product attributes;

[0089] The target search result determination submodule is used to determine the target search result based on the intermediate search results and the current search information;

[0090] The evaluation index determination submodule is used to determine the search evaluation index based on the target search results;

[0091] The first result determination submodule is used to take the target search result as the first search result if the search evaluation index meets the first preset threshold.

[0092] In one possible implementation, the target search result determination submodule includes:

[0093] A matrix determination unit is used to determine a second matrix based on the intermediate search results and the product attributes corresponding to the current search information;

[0094] A matrix calculation unit is used to perform operations on the first matrix and the second matrix to obtain a third matrix;

[0095] The target search result determination unit is used to rearrange the intermediate search results using the third matrix to obtain the target search result.

[0096] In one possible implementation, after the evaluation metrics are determined for the sub-module, the following is also included:

[0097] The parameter adjustment submodule is used to adjust the user's historical behavior data and preset product attributes if the search evaluation index does not meet the first preset threshold, so as to obtain the target user's historical behavior data and target product attributes.

[0098] In one possible implementation, the second result determination module 23 includes:

[0099] The target model determination submodule is used to determine the target search model if the time period is greater than or equal to the first preset time period.

[0100] The model prediction submodule is used to input user historical behavior data and current search information into the target search model to obtain the second search result.

[0101] In one possible implementation, the target model determination submodule includes:

[0102] The model acquisition unit is used to acquire the initial search model;

[0103] The model training unit is used to input user historical behavior data into the initial search model, and obtain the target search model when the number of iterations reaches the second preset threshold.

[0104] In one possible implementation, the model training unit includes:

[0105] The data acquisition subunit is used to acquire historical search data for a second preset time period and historical user behavior data for a third preset time period.

[0106] The model training subunit is used to input historical search data from a second preset time period and user historical behavior data from a third preset time period into the initial search model. Once the number of iterations reaches a second preset threshold, the target search model is obtained. The user historical behavior data includes historical search data and historical clicked product data, where historical clicked product data is represented by product attributes.

[0107] Figure 3 This is a schematic diagram of a terminal provided in an embodiment of the present invention. Figure 3 As shown, the terminal 3 in this embodiment includes a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30. When the processor 30 executes the computer program 32, it implements the steps described in the various search method embodiments above, for example... Figure 1 Steps 101 to 103 are shown. Alternatively, when processor 30 executes computer program 32, it implements the functions of each module / unit in the above-described search device embodiments, for example... Figure 2 The functions of modules 21 to 23 are shown.

[0108] The present invention also provides a readable storage medium storing a computer program, which, when executed by a processor, is used to implement the methods provided in the various embodiments described above.

[0109] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0110] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0111] In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0112] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A search method, characterized in that, include: Obtain user historical behavior data and the time period corresponding to the user historical behavior data; If the time period is less than the first preset time period, the first search result is determined based on the user's historical behavior data, preset product attributes, and current search information; If the time period is greater than or equal to the first preset time period, the second search result is determined based on the user's historical behavior data, current search information, and target search model; If the time period is less than a first preset time period, the first search result is determined based on the user's historical behavior data, preset product attributes, and current search information, including: If the time period is less than the first preset time period, the initial search results are determined based on the user's historical behavior data and preset product attributes; The initial search results are rearranged using a first matrix to obtain intermediate search results, wherein the first matrix is ​​determined by the user's historical behavior data and preset product attributes; Based on the intermediate search results and the current search information, determine the target search result; Based on the target search results, determine the search evaluation metrics; If the search evaluation index meets the first preset threshold, the target search result will be used as the first search result; If the search evaluation index does not meet the first preset threshold, the user's historical behavior data and the preset product attributes are adjusted to obtain the target user's historical behavior data and the target product attributes. Determining the target search result based on the intermediate search results and the current search information includes: Based on the intermediate search results and the product attributes corresponding to the current search information, a second matrix is ​​determined; The first matrix and the second matrix are operated on to obtain the third matrix; The intermediate search results are rearranged using the third matrix to obtain the target search result.

2. The search method as described in claim 1, characterized in that, If the time period is greater than or equal to the first preset time period, the second search result is determined based on the user's historical behavior data, current search information, and target search model, including: If the time period is greater than or equal to the first preset time period, the target search model is determined; The user's historical behavior data and the current search information are input into the target search model to obtain the second search result.

3. The search method as described in claim 2, characterized in that, Determining the target search model includes: Obtain the initial search model; The user's historical behavior data is input into the initial search model, and the target search model is obtained when the number of iterations reaches a second preset threshold.

4. The search method as described in claim 3, characterized in that, The step of inputting the user's historical behavior data into the initial search model, and obtaining the target search model after the number of iterations reaches a second preset threshold, includes: Obtain historical search data for a second preset time period and historical user behavior data for a third preset time period; The historical search data of the second preset time period and the user historical behavior data of the third preset time period are input into the initial search model. When the number of iterations reaches the second preset threshold, the target search model is obtained. The user historical behavior data includes historical search data and historical clicked product data, wherein the historical clicked product data is characterized by product attributes.

5. A search device, characterized in that, include: The data acquisition module is used to acquire user historical behavior data and the time period corresponding to the user historical behavior data; The first result determination module is used to determine the first search result based on the user's historical behavior data, preset product attributes, and current search information if the time period is less than a first preset time period. The second result determination module is used to determine the second search result based on the user's historical behavior data, current search information, and target search model if the time period is greater than or equal to the first preset time period. If the time period is less than a first preset time period, the first search result is determined based on the user's historical behavior data, preset product attributes, and current search information, including: If the time period is less than the first preset time period, the initial search results are determined based on the user's historical behavior data and preset product attributes; The initial search results are rearranged using a first matrix to obtain intermediate search results, wherein the first matrix is ​​determined by the user's historical behavior data and preset product attributes; Based on the intermediate search results and the current search information, determine the target search result; Based on the target search results, determine the search evaluation metrics; If the search evaluation index meets the first preset threshold, the target search result will be used as the first search result; If the search evaluation index does not meet the first preset threshold, the user's historical behavior data and the preset product attributes are adjusted to obtain the target user's historical behavior data and the target product attributes. Determining the target search result based on the intermediate search results and the current search information includes: Based on the intermediate search results and the product attributes corresponding to the current search information, a second matrix is ​​determined; The first matrix and the second matrix are operated on to obtain the third matrix; The intermediate search results are rearranged using the third matrix to obtain the target search result.

6. A terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the searched method as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the search method as described in any one of claims 1 to 4.

Images