Resource content push method and related equipment

Abstract

The present application discloses a resource content push method and related equipment, which includes: obtaining multiple resource contents searched by a target object within a first historical time period and the corresponding search times; and obtaining a first number of reference objects that generate search behavior and a second number of reference objects that search for each resource content within a second historical time period; scoring the target object's preference for the resource content based on the search time, the second number, and the first number corresponding to any resource content, to obtain a time decay score and an importance preference score for the resource content; determining the target object's target preference score for the resource content based on the time decay score and the importance preference score; determining the target object's target preference score for the resource content based on the target object's target preference score for each resource content; and pushing preferred resource content from multiple resource contents based on the target object's target preference score for each resource content; accurately characterizing the target object's behavioral preferences to achieve personalized push of resource content.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a resource content push method and related equipment. Background Art

[0002] In resource content push scenarios, it's often possible to profile the target's behavioral preferences based on their historical search data, allowing personalized push of resource content based on their preferences. For example, this can be achieved by analyzing the resource content that prompted the target to pay over a period of time. Therefore, accurately profiling the target's behavioral preferences to achieve personalized push of resource content is a current research hotspot. Summary of the Invention

[0003] The embodiments of the present application provide a resource content push method, apparatus, device, storage medium, and computer program product, which can accurately characterize the behavioral preferences of a target object to achieve personalized push of resource content.

[0004] In one aspect, an embodiment of the present application provides a method for pushing resource content, including:

[0005] Obtaining multiple resource contents searched by a target object within a first historical time period and the search time corresponding to each resource content, and obtaining a first number of reference objects that generated search behavior within a second historical time period, and a second number of reference objects that searched for each resource content within the second historical time period; the second historical time period includes the first historical time period;

[0006] For any resource content among the multiple resource contents, based on the search time corresponding to the any resource content, the first quantity and the second quantity corresponding to the any resource content, the target object's preference for the any resource content is scored to obtain a time decay score and an importance preference score for the target object for the any resource content; the time decay score is used to indicate the influence of the search time corresponding to the any resource content on the target object's preference for the any resource content, and the importance preference score is used to indicate the importance of the any resource content to the target object;

[0007] Determining a target preference score of the target object for the any resource content based on the time decay score and the importance preference score; the target preference score is used to indicate: the degree of preference of the target object for the any resource content within the first historical time period;

[0008] Determine preferred resource content from the multiple resource contents, and push the preferred resource content to the target object; the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents.

[0009] In one aspect, an embodiment of the present application provides a resource content push device, comprising:

[0010] an acquisition unit, configured to acquire multiple resource contents searched by a target object within a first historical time period and the search time corresponding to each resource content, and acquire a first number of reference objects that generated search behavior within a second historical time period, and a second number of reference objects that searched for each resource content within the second historical time period; the second historical time period includes the first historical time period;

[0011] a processing unit for scoring the target object's preference for any resource content among the multiple resource contents based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content, to obtain a time decay score and an importance preference score for the target object for the any resource content; the time decay score is used to indicate the influence of the search time corresponding to the any resource content on the target object's preference for the any resource content, and the importance preference score is used to indicate the importance of the any resource content to the target object;

[0012] The processing unit is further configured to determine a target preference score of the target object for the any resource content based on the time decay score and the importance preference score; the target preference score is configured to indicate the target object's preference for the any resource content within the first historical time period;

[0013] The processing unit is further configured to determine preferred resource content from the multiple resource contents and push the preferred resource content to the target object; the target object's target preference score for the preferred resource content is greater than the target object's target preference score for other resource contents.

[0014] In one aspect, an embodiment of the present application provides a resource content push device, characterized in that the resource content push device includes an input interface and an output interface, and further includes:

[0015] a processor adapted to implement one or more instructions; and

[0016] A computer storage medium stores one or more instructions, wherein the one or more instructions are suitable for being loaded by the processor and executing the above-mentioned resource content pushing method.

[0017] On the one hand, an embodiment of the present application provides a computer storage medium, characterized in that the computer storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, they are used to execute the above-mentioned resource content push method.

[0018] On the one hand, an embodiment of the present application provides a computer program product, which includes a computer program stored in a computer storage medium; the processor of the resource content push device reads the computer program from the computer storage medium, and the processor executes the computer program, so that the resource content push device executes the above-mentioned resource content push method.

[0019] In an embodiment of the present application, multiple resource contents searched by the target object in a first historical time period and the search time corresponding to each resource content can be obtained, and a first number of reference objects that generated search behavior in a second historical time period and a second number of reference objects that searched for each resource content in the second historical time period can be obtained; and then, for any resource content among the multiple resource contents, the target object's preference for the any resource content can be scored based on the search time corresponding to the any resource content, the first number and the second number corresponding to the any resource content, to obtain the target object's time decay score and importance preference score for the any resource content; based on the time decay score and Importance preference score, determines the target preference score of the target object for any resource content; then, the preferred resource content can be determined from multiple resource contents, and the preferred resource content can be pushed to the target object; wherein, the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents; the time decay score can be used to indicate: the search time corresponding to any resource content, the influence of the target object's preference for any resource content, the importance preference score can be used to indicate: the importance of any resource content to the target object; the target preference score can be used to indicate: the target object's preference for any resource content within the first historical time period.

[0020] Generally speaking, the target object's preference for any resource content is positively correlated with the time difference between the current system time and the search time corresponding to the resource content. That is, the farther the search time corresponding to the resource content is from the current system time, the lower the target object's preference for the resource content is generally, and the closer the search time corresponding to the resource content is to the current system time, the higher the target object's preference for the resource content is generally. Therefore, by introducing a time decay score that can indicate the impact of the search time corresponding to the resource content on the target object's preference for the resource content, the impact of the target object's search time for the resource content on the target object's preference for the resource content can be fully introduced, and then based on the importance preference score that can indicate the importance of the resource content to the target object, the target object's preference for each resource content can be accurately characterized, and then personalized push of resource content can be achieved based on the target object's preference for each resource content. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1 This is a schematic diagram of pushing resource content provided by an embodiment of the present application;

[0023] Figure 2 This is a flow chart of a resource content push method provided in an embodiment of the present application;

[0024] Figure 3 This is a schematic diagram of search time corresponding to resource content provided by an embodiment of the present application;

[0025] Figure 4 This is a flowchart of another resource content push method provided by an embodiment of the present application;

[0026] Figure 5 This is a flowchart of another resource content push method provided by an embodiment of the present application;

[0027] Figure 6 This is a schematic diagram of the structure of the resource content push device provided in an embodiment of the present application;

[0028] Figure 7 It is a structural diagram of the resource content push device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0030] An embodiment of the present application provides a resource content push solution, which can obtain multiple resource contents searched by a target object in a first historical time period and the search time corresponding to each resource content, and obtain a first number of reference objects that generated search behavior in a second historical time period, and a second number of reference objects that searched for each resource content in the second historical time period; then, for any resource content among the multiple resource contents, based on the search time, first number and second number corresponding to the any resource content, the target object's preference for the any resource content can be scored to obtain a time decay score and an importance preference score for the any resource content; based on the time decay score and the importance preference score, the target object's target preference score for the any resource content is determined; the preferred resource content is determined from the multiple resource contents, and the preferred resource content is pushed to the target object.

[0031] Among them, the second historical time period includes the first historical time period. The time decay score of the target object for any resource content is used to indicate: the influence of the search time corresponding to the any resource content on the target object's preference for the any resource content. The importance preference score of the target object for any resource content is used to indicate: the importance of the any resource content to the target object. The target preference score of the target object for any resource content is used to indicate: the target object's preference for the any resource content within the first historical time period. The target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents.

[0032] Furthermore, the first historical time period and the second historical time period can be set according to specific needs. The embodiment of the present application is explained by taking the first historical time period and the second historical time period as the same time period as an example; the reference object that generates the search behavior in the second historical time period includes the target object. The target preference score of the target object for any resource content is positively correlated with the time decay score and importance preference score of the target object for any resource content; the closer the search time corresponding to any resource content is to the current system time, the greater the time decay score of the target object for any resource content, indicating that the search time corresponding to any resource content has a greater impact on the target object's preference for any resource content, and also indicating that the search time corresponding to any resource content has a greater impact on determining the target object's target preference score for any resource content.

[0033] In a specific implementation, the above-mentioned resource content push scheme can be executed by a resource content push device, which can be a terminal device or a server; the terminal device here can include but is not limited to: computers, smart phones, tablet computers, laptops, intelligent voice interaction devices, smart home appliances, car terminals, smart wearable devices, etc.; the server here can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), as well as big data and artificial intelligence platforms and other basic cloud computing services. Further optionally, the above-mentioned resource content push scheme can also be executed individually or collaboratively by any electronic device with computing power, which is not limited in the embodiments of the present application. The embodiments of the present application will be explained later using the resource content push device as an example.

[0034] The above resource content push solution can be used to push any type of resource content, for example, it can be used to push resource content such as fund products and insurance products. Figure 1, is a schematic diagram of pushing resource content provided by an embodiment of the present application. If the target object searches for multiple resource contents including resource content 1, resource content 2, resource content 3, resource content 4, and resource content 5 through the target client running in its terminal device in the first historical time period, the target client is a client for pushing resource contents; the target client can send the search time corresponding to each resource content searched by the target object in the first historical time period to the resource content pushing device; the resource content pushing device generates a first number of reference objects for search behavior in the second historical time period based on the search time corresponding to each resource content searched by the target object in the first historical time period, and searches for each resource content in the second historical time period. The second number of reference objects of the source content, after determining the target preference score of the target object for each resource content, can determine the preferred resource content from multiple resource contents, wherein the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents; and then the preferred resource content can be pushed to the target object, specifically to the target client running in the terminal device of the target object; if resource content 1, resource content 2 and resource content 3 are determined as preferred resource content, the resource content push device can push resource content 1, resource content 2 and resource content 3 to the target client running in the terminal device of the target object, so that the target object can view the corresponding resource content. Further optionally, after the resource content push device determines the preferred resource content from multiple resource contents, it can perform push-related operations only when it is necessary to push the preferred resource content, such as when the target object cold-starts the target client, when the target object refreshes the page in the target client, and so on.

[0035] It is particularly important to note that in the specific implementation of this application, object-related data is involved, such as multiple resource contents searched by the target object within the first historical time period and the search time corresponding to each resource content. When the embodiment of this application is applied to a specific product or technology, it is necessary to obtain the object's permission or consent, and the collection, use and processing of relevant data must comply with local laws, regulations and standards.

[0036] Based on the above resource content push solution, the embodiment of the present application provides a resource content push method. Figure 2 , which is a flow chart of a resource content pushing method provided in an embodiment of the present application. Figure 2 The resource content pushing method shown can be executed by a resource content pushing device. Figure 2 The resource content push method shown may include the following steps:

[0037] S201, obtain multiple resource contents searched by the target object in the first historical time period and the search time corresponding to each resource content, and obtain the first number of reference objects that generate search behavior in the second historical time period, and the second number of reference objects that search for each resource content in the second historical time period.

[0038] Among them, the second historical time period includes the first historical time period; the first historical time period and the second historical time period can be set according to specific needs. The embodiment of the present application is explained by taking the first historical time period and the second historical time period as the same time period as an example; based on this, it can be seen that the reference object that generates the search behavior in the second historical time period includes the target object.

[0039] In one embodiment, if the target object has multiple search behaviors for a certain resource content within the first historical time period, that is, the target object searches for a certain resource content multiple times within the first historical time period, then the number of search times corresponding to the resource content is multiple; a search time corresponding to the resource content can be: the time when the target object generates the corresponding search behavior for the resource content within the first historical time period, that is, the time when the target object searches for the resource content once within the first historical time period.

[0040] For example, see Figure 3, which is a schematic diagram of the search time corresponding to a resource content provided in an embodiment of the present application. If the search behavior generated by the target object in the first historical time period and the time when the corresponding search behavior is generated are represented by a first sequence, it can be: target object: timestamp 1|search behavior 1; timestamp 2|search behavior 2; timestamp 3|search behavior 3; timestamp 4|search behavior 4; timestamp 5|search behavior 5; timestamp 6|search behavior 6; timestamp 7|search behavior 7; timestamp 8|search behavior 8; if the target object generates a search behavior for resource content 1 at the time indicated by timestamp 1, the time indicated by timestamp 4, and the time indicated by timestamp 8, respectively, that is, the target object searches for resource content 1 at the time indicated by timestamp 1, the time indicated by timestamp 4, and the time indicated by timestamp 8, respectively; if At the time indicated by timestamp 2 and the time indicated by timestamp 7, the target object generated a search behavior for resource content 2; if at the time indicated by timestamp 3, the target object generated a search behavior for resource content 3, at the time indicated by timestamp 5, the target object generated a search behavior for resource content 4, and at the time indicated by timestamp 6, the target object generated a search behavior for resource content 5; then the multiple resource contents searched by the target object in the first historical time period and the search time corresponding to each resource content are represented by a second sequence, which can be: target object: timestamp 1|resource content 1; timestamp 2|resource content 2; timestamp 3|resource content 3; timestamp 4|resource content 1; timestamp 5|resource content 4; timestamp 6|resource content 5; timestamp 7|resource content 2; timestamp 8|resource content 1.

[0041] Furthermore, the search times corresponding to the resource contents searched by the target object within the first historical time period are: resource content 1: the time indicated by timestamp 1, the time indicated by timestamp 4, and the time indicated by timestamp 8, a total of 3; resource content 2: the time indicated by timestamp 2 and the time indicated by timestamp 7, a total of 2; resource content 3: the time indicated by timestamp 3, a total of 1; resource content 4: the time indicated by timestamp 5, a total of 1; resource content 5: the time indicated by timestamp 6, a total of 1.

[0042] S202, for any resource content among multiple resource contents, based on the search time, the first quantity and the second quantity corresponding to any resource content, the target object's preference for any resource content is scored to obtain the target object's time decay score and importance preference score for any resource content.

[0043] The second number corresponding to the resource content is the second number of reference objects that searched for the resource content within the second historical time period. The time decay score indicates the impact of the search time corresponding to the resource content on the target object's preference for the resource content, and the importance preference score indicates the importance of the resource content to the target object. Furthermore, the closer the search time corresponding to the resource content is to the current system time, the greater the target object's time decay score for the resource content, indicating a greater impact of the search time on the target object's preference for the resource content and a greater influence on determining the target object's target preference score for the resource content. It also indicates that the search behavior generated at the search time for the resource content has a greater impact on the target object's decision-making after the current system time, i.e., a greater impact on the target object's decision-making after the current system time, such as which resource content the target object will be interested in, which resource content the target object will trigger, which resource content the target object will subscribe to, and which resource content the target object will purchase. Generally speaking, the greater the target object's importance preference score for the resource content, the greater the importance of the resource content to the target object.

[0044] In one embodiment, the resource content push device scores the target object's preference for any resource content based on the search time, first quantity and second quantity corresponding to any resource content, and obtains the target object's importance preference score for any resource content, which may include: determining the target object's time decay score for any resource content based on the time difference between the current system time and the search time corresponding to the any resource content; the time decay score is negatively correlated with the time difference; counting the number of first behaviors of the target object generating search behaviors within the first historical time period, and the number of second behaviors of the target object searching for the any resource content within the first historical time period; determining the target object's importance preference score for the any resource content based on the first number of behaviors, the second number of behaviors, the first quantity and the second quantity corresponding to any resource content. Taking the above-mentioned second sequence as an example, the number of first behaviors is 8. If any resource content is resource content 1, the corresponding number of second behaviors is 3. If any resource content is resource content 2, the corresponding number of second behaviors is 2. If any resource content is resource content 3, the corresponding number of second behaviors is 1. If any resource content is resource content 4, the corresponding number of second behaviors is 1. If any resource content is resource content 5, the corresponding number of second behaviors is 1.

[0045] S203: Determine the target preference score of the target object for any resource content based on the time decay score and the importance preference score.

[0046] Among them, the target preference score is used to indicate: the target object's preference for any resource content within the first historical time period; the larger the target preference score of the target object for any resource content, the greater the target object's preference for any resource content within the first historical time period; the target object's target preference score for any resource content is positively correlated with the target object's time attenuation score and importance preference score for any resource content.

[0047] S204: Determine preferred resource content from multiple resource contents, and push the preferred resource content to the target object.

[0048] Among them, the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource content. In a feasible implementation, the resource content push device can determine the resource content indicated by the target preference scores ranked in the top K (i.e., Top K) among the target preference scores of the target object for each resource content as the preferred resource content, wherein K can be set according to specific needs. In another feasible implementation, the resource content push device can determine the resource content indicated by the target preference scores greater than the preference score threshold among the target preference scores of the target object for each resource content as the preferred resource content. Among them, the preference score threshold can be set according to specific needs.

[0049] In an embodiment of the present application, multiple resource contents searched by the target object in a first historical time period and the search time corresponding to each resource content can be obtained, and a first number of reference objects that generated search behavior in a second historical time period and a second number of reference objects that searched for each resource content in the second historical time period can be obtained; and then, for any resource content among the multiple resource contents, the target object's preference for the any resource content can be scored based on the search time corresponding to the any resource content, the first number and the second number corresponding to the any resource content, to obtain the target object's time decay score and importance preference score for the any resource content; based on the time decay score and Importance preference score, determines the target preference score of the target object for any resource content; then, the preferred resource content can be determined from multiple resource contents, and the preferred resource content can be pushed to the target object; wherein, the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents; the time decay score can be used to indicate: the search time corresponding to any resource content, the influence of the target object's preference for any resource content, the importance preference score can be used to indicate: the importance of any resource content to the target object; the target preference score can be used to indicate: the target object's preference for any resource content within the first historical time period.

[0050] Generally speaking, the target object's preference for any resource content is positively correlated with the time difference between the current system time and the search time corresponding to the resource content. That is, the farther the search time corresponding to the resource content is from the current system time, the lower the target object's preference for the resource content is generally, and the closer the search time corresponding to the resource content is to the current system time, the higher the target object's preference for the resource content is generally. Therefore, by introducing a time decay score that can indicate the impact of the search time corresponding to the resource content on the target object's preference for the resource content, the impact of the target object's search time for the resource content on the target object's preference for the resource content can be fully introduced, and then based on the importance preference score that can indicate the importance of the resource content to the target object, the target object's preference for each resource content can be accurately characterized, and then personalized push of resource content can be achieved based on the target object's preference for each resource content.

[0051] Based on the above-mentioned related embodiments of the resource content push method, the present application embodiment provides another resource content push method. Figure 4 , which is a flow chart of another resource content pushing method provided in an embodiment of the present application. Figure 4 The resource content pushing method shown can be executed by a resource content pushing device. Figure 4 The resource content push method shown may include the following steps:

[0052] S401, obtain multiple resource contents searched by the target object in the first historical time period and the search time corresponding to each resource content, and obtain the first number of reference objects that generate search behavior in the second historical time period, and the second number of reference objects that search for each resource content in the second historical time period.

[0053] The second historical time period includes the first historical time period.

[0054] In one embodiment, a method for obtaining any resource content from a plurality of resource contents, i.e., a resource content push device obtaining any resource content searched by a target object within a first historical time period, may include: obtaining a search text entered into an input box by the target object within the first historical time period; determining the degree of difference between the search text and each of a plurality of preset resource contents; and determining the preset resource content indicated by the minimum degree of difference among the plurality of preset resource contents as the any resource content. The input box may be an input box provided by the target client for inputting a search text to search for resource content, and the search text may be obtained by typing text or by performing voice recognition on a recorded voice; the preset resource content is resource content that can be provided by the target client, such as resource content such as fund products and insurance products. Furthermore, based on the minimum difference, any one of the resource contents determined from multiple preset resource contents matches the search text; any one of the resource contents is the preset resource content matched from multiple preset resource contents, which can represent the target object's search intention this time, and is mainly used to complete the search text data to reduce the noise contained in the search text; for example, if the search text input by the target object is "A Insurance Intention", and the preset resource content matched from multiple preset resource contents is "A Insurance Accident Insurance", then the resource content matching the search text is "A Insurance Accident Insurance", and then the resource content pushed to the target object based on the resource content "A Insurance Accident Insurance" may be "A Insurance Accident Insurance (Student Edition)", "A Insurance Accident Insurance (Adult Edition)", "A Insurance Accident Insurance (1 Day)", and so on.

[0055] Furthermore, after the target object enters the search text, the time when the search behavior is executed based on the search text is determined as a search time for the any resource content. The time when the search behavior is executed based on the search text can be, for example, the time when the search execution component is triggered after the target object enters the search text. Furthermore, different search texts entered into the input box by the target object within the first historical time period may match the same preset resource content from multiple preset resource contents as the corresponding resource content; after the target object enters each search text within the first historical time period, the time when the search behavior is executed based on each search text is determined as each search time for the any resource content.

[0056] For example, if the search behavior generated by the target object in the first historical time period and the time when the corresponding search behavior was generated are represented by a search text sequence, taking the above-mentioned first sequence as an example, if the corresponding search text sequence is: target object: timestamp 1|search text 1; timestamp 2|search text 2; timestamp 3|search text 3; timestamp 4|search text 4; timestamp 5|search text 5; timestamp 6|search text 6; timestamp 7|search text 7; timestamp 8|search text 8; if the preset resource contents matched by each search text from multiple preset resources are: preset resource content 1, preset resource content 2, preset resource content 3, preset resource content 1, preset resource content 4, preset resource content 5. Preset resource content 2, preset resource content 1; the multiple resource contents searched by the target object in the first historical time period and the search time corresponding to each resource content can be expressed in the second sequence as: target object: timestamp 1|resource content 1; timestamp 2|resource content 2; timestamp 3|resource content 3; timestamp 4|resource content 1; timestamp 5|resource content 4; timestamp 6|resource content 5; timestamp 7|resource content 2; timestamp 8|resource content 1; among them, resource content 1 is the preset resource content 1, resource content 2 is the preset resource content 2, resource content 3 is the preset resource content 3, resource content 4 is the preset resource content 4, and resource content 5 is the preset resource content 5.

[0057] In a specific implementation, the resource content pushing device determines the difference degree between the search text and any one of the multiple preset resource contents, which may include: determining the edit distance between the search text and the content identifier of any one of the preset resource contents; using the edit distance as the difference degree between the search text and any one of the preset resource contents; wherein, the edit distance is used to indicate: the minimum number of editing operations required to convert the search text into the content identifier of any one of the preset resource contents. Optionally, the content identifier of any one of the preset resource contents may be the name, description text, etc. of any one of the preset resource contents that can uniquely indicate any one of the preset resource contents. In the embodiments of the present application, the name is used as an example for elaboration. Further, the minimum number of editing operations required to convert the search text into the content identifier of any one of the preset resource contents may be: the minimum number of editing operations required to execute the editing operations required to convert the search text into the content identifier of any one of the preset resource contents; optionally, the editing operation may include a character insertion operation, a character deletion operation, a character replacement operation, etc. For example, if the multiple preset resource contents include: "A Insurance Accident Insurance", "A Insurance Accident Insurance (Student Version)", "A Insurance Accident Insurance (Adult Version)", "A Insurance Accident Insurance (1 Day)", "A Insurance Life Insurance", "B Insurance Accident Insurance", etc., if the search text input by the target object is "A Insurance Accident", then the number of editing operations required to convert "A Insurance Accident" into "A Insurance Accident Insurance" is 2, that is, execute a character insertion operation, insert "Out", execute a character insertion operation, insert "Insurance", and convert "A Insurance Accident" into "A Insurance Life Insurance" The number of editing operations required is 3, that is, execute a character replacement operation, replace "Accident" with "Person", execute a character insertion operation, insert "Life", execute a character insertion operation, insert "Insurance", and so on.

[0058] In another feasible implementation manner, after the resource content pushing device obtains the search text input by the target object into the input box within the first historical time period, it may call a text intention recognition classifier to execute operations of determining the difference degree between the search text and each of the multiple preset resource contents, and determining the preset resource content indicated by the minimum difference degree among the multiple preset resource contents as any one of the relevant resource contents. In a specific implementation, when the resource content pushing device calls the text intention recognition classifier to determine the difference degree between the search text and any one of the multiple preset resource contents, it may include: extracting the semantic features of the search text and the semantic features of the content identifier of any one of the preset resource contents through the text intention recognition classifier; determining the feature distance between the semantic features of the search text and the semantic features of the content identifier of any one of the preset resource contents; using the feature distance as the difference degree between the search text and any one of the preset resource contents; optionally, the feature distance may be an Euclidean distance, a Manhattan distance, etc., and can be adaptively selected according to specific requirements.

[0059] Optionally, the text intent recognition classifier can be any neural network model capable of classifying search text into content identifiers for multiple pre-defined resource contents. The construction of the text intent recognition classifier relies on machine learning technology within the field of artificial intelligence. Artificial intelligence (AI) is the theory, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive field of computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess perception, reasoning, and decision-making capabilities. AI technology is an interdisciplinary discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. Artificial intelligence software technologies primarily encompass computer vision (CV), speech processing, natural language processing, and machine learning (ML) / deep learning (DL). Machine learning is the core of AI and the fundamental way to make computers intelligent. Its applications span all areas of AI.

[0060] S402 : For any resource content among the multiple resource contents, determine a time decay score of the target object for any resource content based on a time difference between the current system time and the search time corresponding to the resource content.

[0061] Among them, the time decay score is used to indicate: the impact of the search time corresponding to any resource content on the target object's preference for any resource content; the time decay score is negatively correlated with the time difference. In other words, the larger the time difference, that is, the farther the search time corresponding to any resource content is from the current system time, the smaller the time decay score.

[0062] In one embodiment, if the target object searches for the second behavior of any resource content once in the first historical time period, that is, the target object searches for the resource content once in the first historical time period, and the number of search times corresponding to the resource content is one; then the time decay score of the target object for the resource content can be expressed by the following formula 1.1:

[0063] V ui =e -αt (1.1)

[0064] Among them, u represents the target object, i represents the resource content; α is a parameter that can be set according to specific needs, for example, it can be set to 0.00006, 0.00007, etc.; t represents the time difference between the current system time and the search time corresponding to the resource content. If the current system time is t n Indicates the search time t corresponding to any resource content s If t=t n -t s .

[0065] In one embodiment, if the target object searches for the second behavior of any resource content multiple times within the first historical time period, that is, the target object searches for the any resource content multiple times within the first historical time period, and the number of search times corresponding to the any resource content is multiple; then the resource content push device determines the time decay score of the target object for any resource content based on the time difference between the current system time and the search time corresponding to any resource content, which may include: determining the time decay sub-score of the target object for the any resource content at each search time based on the time difference between the current system time and each search time corresponding to the any resource content; and determining the time decay score of the target object for the any resource content according to the time decay sub-score of the target object for the any resource content at each search time.

[0066] In a feasible implementation, the resource content push device may sum the time decay sub-scores of the target object for the resource content at each search time, and use the summed result as the time decay score of the target object for the resource content. In this case, the time decay score of the target object for the resource content may be expressed by the following formula 1.2:

[0067]

[0068] Where u represents the target object, i represents any resource content; α is a parameter that can be set according to specific needs, for example, it can be set to 0.00006, 0.00007, etc.; j is an independent variable, H is the number of search times corresponding to any resource content, t j It represents the time difference between the current system time and the j-th search time corresponding to any resource content; when H is 1, formula 1.2 can be unified with the above formula 1.1.

[0069] In a feasible implementation, the resource content push device determines the time decay score of the target object for any resource content based on the time decay sub-score of the target object for the any resource content at each search time. The maximum time decay sub-score among the time decay sub-scores of the target object for the any resource content at each search time can be determined as the time decay score of the target object for the any resource content. In this case, it is equivalent to only using the search time closest to the current system time among the multiple search times corresponding to the any resource content to determine the time decay score of the target object for the any resource content.

[0070] S403: Count the number of first search behaviors generated by the target object within the first historical time period, and the number of second search behaviors generated by the target object within the first historical time period.

[0071] In one embodiment, the number of search times corresponding to any resource content can be used as the second number of times the target object searches for the any resource content within the first historical time period; taking the above-mentioned first sequence and second sequence as an example, the first number of times the target object generates a search behavior within the first historical time period is statistically 8. If the any resource content is resource content 1, the second number of times the target object searches for the any resource content within the first historical time period is statistically 3.

[0072] S404 : Determine the importance preference score of the target object for any resource content based on the first behavior number, the second behavior number, the first quantity, and the second quantity corresponding to any resource content.

[0073] The importance preference score is used to indicate the importance of any resource content to the target object.

[0074] In one embodiment, the resource content push device determines the importance preference score of the target object for any resource content based on the number of first behaviors, the number of second behaviors, the first quantity and the second quantity corresponding to any resource content, which may include: determining the target term frequency (TF) corresponding to any resource content according to the number of first behaviors and the second behavior; the target term frequency is used to indicate: the frequency with which the target object searches for the any resource content in the first historical time period; determining the reference search frequency corresponding to the any resource content according to the first quantity and the second quantity corresponding to the any resource content; the reference search frequency is used to indicate: the frequency with which the any resource content is searched by the reference object in the second historical time period; determining the importance preference score of the target object for the any resource content according to the target term frequency and the reference search frequency corresponding to the any resource content; the importance preference score is positively correlated with the target term frequency, and the importance preference score is negatively correlated with the reference search frequency.

[0075] In a specific implementation, the resource content push device may determine the ratio of the second behavior count to the first behavior count as the target word frequency corresponding to any resource content; the target word frequency corresponding to any resource content may be expressed by the following formula 2:

[0076]

[0077] Among them, N2 represents the second number of times the target object searches for the content of any resource in the first historical time period, and N1 represents the first number of times the target object generates a search behavior in the first historical time period.

[0078] Furthermore, the resource content pushing device may determine the ratio of the second number corresponding to any resource content to the first number as the reference search frequency corresponding to the any resource content.

[0079] In an optional embodiment, the resource content push device determines the importance preference score of the target object for any resource content based on the target word frequency and reference search frequency corresponding to any resource content, which may include: determining the inverse-document-frequency (IDF) corresponding to any resource content based on the reference search frequency corresponding to the any resource content; the inverse-document frequency is negatively correlated with the reference search frequency; determining the word frequency-inverse document frequency (TF-IDF) corresponding to any resource content based on the target word frequency and inverse document frequency corresponding to the any resource content, and using the word frequency-inverse document frequency as the importance preference score of the target object for the any resource content.

[0080] Among them, the inverse document frequency is usually used to measure the universal importance of words in a document. The more common a certain word in the target document is in multiple documents in the document library, the smaller the universal importance of the word, and the smaller the inverse document frequency corresponding to the word; similarly, the more frequently the reference object searches for any resource content in the second historical time period, the more reference objects are interested in the resource content, the higher the popularity of the resource content, and the smaller the universal importance of the resource content. Specifically, the inverse document frequency corresponding to the resource content can be shown by the following formula 3:

[0081]

[0082] Among them, M1 represents the first number of reference objects that generate search behavior in the second historical time period, and M2 represents the second number of reference objects that search for the any resource content in the second historical time period.

[0083] Furthermore, the term frequency-inverse document frequency corresponding to any resource content can be expressed by the following formula 4:

[0084] tf-idf ui =tf ui *idf i (4)

[0085] S405 : Determine the target preference score of the target object for any resource content based on the time decay score and the importance preference score.

[0086] Among them, the target preference score is used to indicate: the target object's preference for any resource content within the first historical time period; the target object's target preference score for any resource content is positively correlated with the target object's time decay score and importance preference score for any resource content. In an optional embodiment, the resource content push device can use the product of the target object's time decay score and importance preference score for any resource content as the target object's target preference score for any resource content; that is, the target object's time decay score for any resource content can be used as a weight parameter to perform weighted processing on the target object's importance preference score for any resource content to obtain the target object's target preference score for any resource content, which can fully introduce the influence of the target object's search time for any resource content on the target object's preference for any resource content. Specifically, the target object's target preference score for any resource content can be shown by the following formula 5:

[0087] Score ui =V ui *(tf-idf ui )=V ui*tf ui *idf i (5)

[0088] S406: Determine preferred resource content from the multiple resource contents, and push the preferred resource content to the target object.

[0089] The target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource content.

[0090] See also Figure 5 , which is a flow chart of another resource content push method provided in real time by this application. The resource content push device can execute the following steps: S501, obtain a search text sequence that can indicate the search behavior generated by the target object in the first historical time period and the time when the corresponding search behavior was generated; S502, from multiple preset resource contents, determine the resource content that matches each search text in the search text sequence; S503, obtain multiple resource contents searched by the target object in the first historical time period and the search time corresponding to each resource content. A search time corresponding to any resource content is the time when the target object generates the corresponding search behavior for any resource content.

[0091] S504, obtain the first number of reference objects that generate search behavior in the second historical time period, and the second number of reference objects that search for each resource content in the second historical time period; S505, for any resource content among multiple resource contents, based on the search time, the first number and the second number corresponding to the any resource content, score the target object's preference for the any resource content to obtain the target object's time decay score and importance preference score for the any resource content; S506, based on the time decay score and the importance preference score, determine the target object's target preference score for the any resource content; S507, determine the preferred resource content from multiple resource contents, and push the preferred resource content to the target object, and the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents.

[0092] In one embodiment, the resource content push solution proposed in the embodiments of the present application can be reused in any scenario where historical search data that can be obtained and can indicate the historical preferences of the target object is available. That is, when the historical search data used is data related to search behavior, the historical search data can include: multiple resource contents searched by the target object in a first historical time period and the search time corresponding to each resource content, a first number of reference objects that generated search behavior in a second historical time period, and a second number of reference objects that searched for each resource content in the second historical time period. When the historical search data used is data related to click behavior, the historical search data can include: multiple resource contents clicked by the target object in the first historical time period and the click time corresponding to each resource content, a first number of reference objects that generated click behavior in the second historical time period, and a second number of reference objects that clicked each resource content in the second historical time period. When the historical search data used is data related to payment behavior, the historical search data can include: multiple resource contents paid by the target object in the first historical time period and the payment time corresponding to each resource content, a first number of reference objects that generated payment behavior in the second historical time period, and a second number of reference objects that paid for each resource content in the second historical time period.

[0093] In an embodiment of the present application, a time decay score of the target object for any resource content can be introduced, and then when characterizing the target object's preference for any resource content, the influence of the time when the target object generates a search behavior for the any resource content can be fully considered; and by taking the word frequency-inverse document frequency corresponding to the any resource content as the importance preference score of the target object for the any resource content, on the basis of the target word frequency corresponding to the any resource content, the inverse document frequency corresponding to the any resource content is also introduced. The higher the popularity of the resource content, the smaller the corresponding inverse document frequency, and the lower the popularity of the resource content, the larger the corresponding inverse document frequency. The target preference score determined based on the time decay score and the importance preference score can not only accurately characterize the target object's preference for each resource content, and realize personalized push of resource content based on the target object's preference for each resource content, but also alleviate the problem of overexposure of high-hot resource content and underexposure of low-hot resource content.

[0094] Based on the above-mentioned embodiments related to the resource content push method, the present application embodiment provides a resource content push device. Figure 6 , is a structural diagram of a resource content pushing device provided in an embodiment of the present application. The resource content pushing device may include an acquisition unit 601 and a processing unit 602. Figure 6 The resource content pushing device shown can be used to perform the following operations:

[0095] An acquisition unit 601 is configured to acquire multiple resource contents searched by a target object within a first historical time period and the search time corresponding to each resource content, and acquire a first number of reference objects that generated search behavior within a second historical time period, and a second number of reference objects that searched for each resource content within the second historical time period; the second historical time period includes the first historical time period;

[0096] Processing unit 602 is configured to score the target object's preference for any resource content among the multiple resource contents based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content, to obtain a time decay score and an importance preference score for the target object for the any resource content; the time decay score is used to indicate the effect of the search time corresponding to the any resource content on the target object's preference for the any resource content, and the importance preference score is used to indicate the importance of the any resource content to the target object;

[0097] The processing unit 602 is further configured to determine a target preference score of the target object for the any resource content based on the time decay score and the importance preference score; the target preference score is used to indicate the degree of preference of the target object for the any resource content within the first historical time period;

[0098] The processing unit 602 is further configured to determine preferred resource content from the multiple resource contents and push the preferred resource content to the target object; the target object's target preference score for the preferred resource content is greater than the target object's target preference score for other resource contents.

[0099] In one embodiment, the processing unit 602 scores the target object's preference for any resource content based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content. When obtaining the target object's importance preference score for the any resource content, the processing unit 602 specifically performs the following operations:

[0100] Counting the number of first search behaviors generated by the target object within the first historical time period, and the number of second search behaviors generated by the target object within the first historical time period;

[0101] An importance preference score of the target object for the any resource content is determined based on the first behavior number, the second behavior number, the first quantity, and the second quantity corresponding to the any resource content.

[0102] In one embodiment, when the processing unit 602 determines the importance preference score of the target object for any resource content based on the first behavior number, the second behavior number, the first quantity, and the second quantity corresponding to any resource content, the processing unit 602 specifically performs the following operations:

[0103] Determining a target word frequency corresponding to the any resource content based on the first behavior count and the second behavior count; the target word frequency is used to indicate: a frequency with which the target object searches for the any resource content within the first historical time period;

[0104] Determining a reference search frequency corresponding to any resource content based on the first number and the second number corresponding to any resource content; the reference search frequency is used to indicate: a frequency at which any resource content is searched by a reference object within the second historical time period;

[0105] Based on the target word frequency and reference search frequency corresponding to any resource content, the importance preference score of the target object for any resource content is determined; the importance preference score is positively correlated with the target word frequency, and negatively correlated with the reference search frequency.

[0106] In one embodiment, when the processing unit 602 determines the importance preference score of the target object for any resource content based on the target word frequency and the reference search frequency corresponding to any resource content, it specifically performs the following operations:

[0107] Determining an inverse document frequency corresponding to any resource content according to a reference search frequency corresponding to any resource content; wherein the inverse document frequency is negatively correlated with the reference search frequency;

[0108] According to the target word frequency and inverse document frequency corresponding to any resource content, the word frequency-inverse document frequency corresponding to any resource content is determined, and the word frequency-inverse document frequency is used as the importance preference score of the target object for any resource content.

[0109] In one embodiment, the processing unit 602 scores the target object's preference for any resource content based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content. When obtaining the target object's time decay score for the any resource content, the processing unit 602 specifically performs the following operations:

[0110] Based on the time difference between the current system time and the search time corresponding to the any resource content, a time decay score of the target object for the any resource content is determined; the time decay score is negatively correlated with the time difference.

[0111] In one embodiment, the target object searches for the second behavior of any resource content multiple times within the first historical time period, and the number of search times corresponding to any resource content is multiple;

[0112] When the processing unit 602 determines the time decay score of the target object for the any resource content based on the time difference between the current system time and the search time corresponding to the any resource content, it specifically performs the following operations:

[0113] determining, based on time differences between the current system time and respective search times corresponding to the respective resource contents, a time decay sub-score of the target object for the respective resource contents at the respective search times;

[0114] The time decay score of the target object for the any resource content is determined according to the time decay sub-score of the target object for the any resource content at the respective search times.

[0115] In one embodiment, a method for obtaining any resource content among the multiple resource contents includes:

[0116] The acquisition unit 601 is configured to acquire the search text entered into the input box by the target object within the first historical time period;

[0117] The acquisition unit 601 is further configured to determine a difference between the search text and each of the plurality of preset resource contents;

[0118] The acquisition unit 601 is further configured to determine the preset resource content indicated by the minimum difference among the plurality of preset resource contents as the any one resource content.

[0119] In one embodiment, when the obtaining unit 601 determines the difference between the search text and any preset resource content among the plurality of preset resource contents, it specifically performs the following operations:

[0120] Determining an edit distance between the search text and the content identifier of any preset resource content; the edit distance is used to indicate: a minimum number of edits required to convert the search text into the content identifier of any preset resource content;

[0121] The edit distance is used as the difference between the search text and any preset resource content.

[0122] According to one embodiment of the present application, Figure 2 、 Figure 4 as well as Figure 5 The steps involved in the resource content push method shown can be performed by Figure 6 The resource content push device shown in FIG. Figure 2 Step S201 shown can be performed by Figure 6 The acquisition unit 601 in the resource content push device shown is executed, Figure 2 Steps S202 to S204 shown in FIG. Figure 6 The processing unit 602 in the resource content push device shown in FIG. Figure 4 Step S401 shown can be performed by Figure 6 The acquisition unit 601 in the resource content push device shown is executed, Figure 4 Steps S402 to S406 shown in FIG. Figure 6 The processing unit 602 in the resource content push device shown in FIG. Figure 5 Steps S501 to S504 shown can be performed by Figure 6 The acquisition unit 601 in the resource content push device shown is executed, Figure 5 Steps S505 to S507 shown in FIG. Figure 6 The processing unit 602 in the resource content pushing device shown is executed.

[0123] According to another embodiment of the present application, Figure 6 The various units in the resource content push device shown can be individually or all combined into one or several other units to form, or one (or some) of the units can be further divided into multiple functionally smaller units to form, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above-mentioned units are divided based on logical functions. In actual applications, the functions of one unit can also be implemented by multiple units, or the functions of multiple units can be implemented by one unit. For example, the functions implemented by the above-mentioned units can be implemented by a processing unit. In other embodiments of the present application, the resource content push device divided based on logical functions can also include other units. In actual applications, these functions can also be implemented with the assistance of other units, and can be implemented by the collaboration of multiple units.

[0124] According to another embodiment of the present application, the program can be executed by running on a general computing device such as a computer including a central processing unit (CPU), a random access memory (RAM), a read-only memory (ROM) and other processing elements and storage elements. Figure 2 、 Figure 4 as well as Figure 5 A computer program (including program code) for each step of the corresponding method shown in FIG. Figure 6 The resource content push device shown in and the resource content push method of the embodiment of the present application are implemented. The computer program can be recorded on, for example, a computer readable storage medium, and loaded into the above-mentioned computing device through the computer readable storage medium and run therein.

[0125] In an embodiment of the present application, multiple resource contents searched by the target object in a first historical time period and the search time corresponding to each resource content can be obtained, and a first number of reference objects that generated search behavior in a second historical time period and a second number of reference objects that searched for each resource content in the second historical time period can be obtained; and then, for any resource content among the multiple resource contents, the target object's preference for the any resource content can be scored based on the search time corresponding to the any resource content, the first number and the second number corresponding to the any resource content, to obtain the target object's time decay score and importance preference score for the any resource content; based on the time decay score and Importance preference score, determines the target preference score of the target object for any resource content; then, the preferred resource content can be determined from multiple resource contents, and the preferred resource content can be pushed to the target object; wherein, the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents; the time decay score can be used to indicate: the search time corresponding to any resource content, the influence of the target object's preference for any resource content, the importance preference score can be used to indicate: the importance of any resource content to the target object; the target preference score can be used to indicate: the target object's preference for any resource content within the first historical time period.

[0126] Generally speaking, the target object's preference for any resource content is positively correlated with the time difference between the current system time and the search time corresponding to the resource content. That is, the farther the search time corresponding to the resource content is from the current system time, the lower the target object's preference for the resource content is generally, and the closer the search time corresponding to the resource content is to the current system time, the higher the target object's preference for the resource content is generally. Therefore, by introducing a time decay score that can indicate the impact of the search time corresponding to the resource content on the target object's preference for the resource content, the impact of the target object's search time for the resource content on the target object's preference for the resource content can be fully introduced, and then based on the importance preference score that can indicate the importance of the resource content to the target object, the target object's preference for each resource content can be accurately characterized, and then personalized push of resource content can be achieved based on the target object's preference for each resource content.

[0127] Based on the above-mentioned related embodiments of the resource content push method and the resource content push device embodiment, the present application also provides a resource content push device. Figure 7 , is a structural diagram of a resource content push device provided in an embodiment of the present application. Figure 7 The resource content push device shown may include at least a processor 701, an input interface 702, an output interface 703, and a computer storage medium 704. The processor 701, the input interface 702, the output interface 703, and the computer storage medium 704 may be connected via a bus or other means.

[0128] Computer storage medium 704 may be stored in the memory of the resource content push device. Computer storage medium 704 is used to store a computer program, which includes program instructions. Processor 701 is used to execute the program instructions stored in computer storage medium 704. Processor 701 (also known as a CPU (Central Processing Unit)) is the computing and control core of the resource content push device. It is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the above-mentioned resource content push method process or corresponding functions.

[0129] The embodiment of the present application also provides a computer storage medium (Memory), which is a memory device in a resource content push device for storing programs and data. It is understandable that the computer storage medium here can include both the built-in storage medium in the terminal and, of course, the extended storage medium supported by the terminal. The computer storage medium provides a storage space that stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor 701 are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer storage medium here can be a high-speed random access memory (RAM) memory, or a non-volatile memory (non-volatile memory), such as at least one disk storage; optionally, it can also be at least one computer storage medium located away from the aforementioned processor.

[0130] In one embodiment, the processor 701 may load and execute one or more instructions stored in a computer storage medium to implement the above-mentioned Figure 2 、 Figure 4 as well as Figure 5 In the corresponding steps of the method in the embodiment of the resource content push method, in a specific implementation, one or more instructions in the computer storage medium are loaded by the processor 701 and the following steps are executed:

[0131] Obtaining multiple resource contents searched by a target object within a first historical time period and the search time corresponding to each resource content, and obtaining a first number of reference objects that generated search behavior within a second historical time period, and a second number of reference objects that searched for each resource content within the second historical time period; the second historical time period includes the first historical time period;

[0132] For any resource content among the multiple resource contents, based on the search time corresponding to the any resource content, the first quantity and the second quantity corresponding to the any resource content, the target object's preference for the any resource content is scored to obtain a time decay score and an importance preference score for the target object for the any resource content; the time decay score is used to indicate the influence of the search time corresponding to the any resource content on the target object's preference for the any resource content, and the importance preference score is used to indicate the importance of the any resource content to the target object;

[0133] Determining a target preference score of the target object for the any resource content based on the time decay score and the importance preference score; the target preference score is used to indicate: the degree of preference of the target object for the any resource content within the first historical time period;

[0134] Determine preferred resource content from the multiple resource contents, and push the preferred resource content to the target object; the target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource contents.

[0135] In one embodiment, the processor 701 scores the target object's preference for any resource content based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content. When obtaining the target object's importance preference score for the any resource content, the processor 701 specifically performs the following operations:

[0136] Counting the number of first search behaviors generated by the target object within the first historical time period, and the number of second search behaviors generated by the target object within the first historical time period;

[0137] An importance preference score of the target object for the any resource content is determined based on the first behavior number, the second behavior number, the first quantity, and the second quantity corresponding to the any resource content.

[0138] In one embodiment, when the processor 701 determines the importance preference score of the target object for any resource content based on the first behavior number, the second behavior number, the first quantity, and the second quantity corresponding to any resource content, the processor 701 specifically performs the following operations:

[0139] Determining a target word frequency corresponding to the any resource content based on the first behavior count and the second behavior count; the target word frequency is used to indicate: a frequency with which the target object searches for the any resource content within the first historical time period;

[0140] Determining a reference search frequency corresponding to any resource content based on the first number and the second number corresponding to any resource content; the reference search frequency is used to indicate: a frequency at which any resource content is searched by a reference object within the second historical time period;

[0141] Based on the target word frequency and reference search frequency corresponding to any resource content, the importance preference score of the target object for any resource content is determined; the importance preference score is positively correlated with the target word frequency, and negatively correlated with the reference search frequency.

[0142] In one embodiment, when the processor 701 determines the importance preference score of the target object for any resource content based on the target word frequency and the reference search frequency corresponding to any resource content, it specifically performs the following operations:

[0143] Determining an inverse document frequency corresponding to any resource content according to a reference search frequency corresponding to any resource content; wherein the inverse document frequency is negatively correlated with the reference search frequency;

[0144] According to the target word frequency and inverse document frequency corresponding to any resource content, the word frequency-inverse document frequency corresponding to any resource content is determined, and the word frequency-inverse document frequency is used as the importance preference score of the target object for any resource content.

[0145] In one embodiment, the processor 701 scores the target object's preference for any resource content based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content. When obtaining the target object's time decay score for the any resource content, the processor 701 specifically performs the following operations:

[0146] Based on the time difference between the current system time and the search time corresponding to the any resource content, a time decay score of the target object for the any resource content is determined; the time decay score is negatively correlated with the time difference.

[0147] In one embodiment, the target object searches for the second behavior of any resource content multiple times within the first historical time period, and the number of search times corresponding to any resource content is multiple;

[0148] When the processor 701 determines the time decay score of the target object for the any resource content based on the time difference between the current system time and the search time corresponding to the any resource content, it specifically performs the following operations:

[0149] determining, based on time differences between the current system time and respective search times corresponding to the respective resource contents, a time decay sub-score of the target object for the respective resource contents at the respective search times;

[0150] The time decay score of the target object for the any resource content is determined according to the time decay sub-score of the target object for the any resource content at the respective search times.

[0151] In one embodiment, a method for obtaining any resource content among the multiple resource contents includes:

[0152] The processor 701 is configured to obtain a search text input into the input box by the target object within the first historical time period;

[0153] The processor 701 is further configured to determine a difference between the search text and each of the plurality of preset resource contents;

[0154] The processor 701 is further configured to determine the preset resource content indicated by the minimum difference among the plurality of preset resource contents as the any one resource content.

[0155] In one embodiment, when the processor 701 determines the difference between the search text and any preset resource content among the plurality of preset resource contents, it specifically performs the following operations:

[0156] Determining an edit distance between the search text and the content identifier of any preset resource content; the edit distance is used to indicate: a minimum number of edits required to convert the search text into the content identifier of any preset resource content;

[0157] The edit distance is used as the difference between the search text and any preset resource content.

[0158] The embodiment of the present application provides a computer program product, which includes a computer program stored in a computer storage medium; a processor of a resource content push device reads the computer program from the computer storage medium, and the processor executes the computer program, so that the resource content push device performs the above-mentioned Figure 2 、 Figure 4 as well as Figure 5 The computer-readable storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0159] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A resource content push method, characterized in that: include: Obtaining multiple resource contents searched by a target object within a first historical time period and the search time corresponding to each resource content, and obtaining a first number of reference objects that generated search behavior within a second historical time period, and a second number of reference objects that searched for each resource content within the second historical time period; the second historical time period includes the first historical time period; For any resource content among the multiple resource contents, scoring the target object's preference for the resource content based on the search time corresponding to the resource content, the first quantity, and the second quantity corresponding to the resource content, to obtain a time decay score and an importance preference score for the target object for the resource content; The time decay score is used to indicate the influence of the search time corresponding to any resource content on the preference of the target object for the resource content, and the importance preference score is used to indicate the importance of any resource content to the target object; Determining a target preference score of the target object for the any resource content based on the time decay score and the importance preference score; The target preference score is used to indicate: the target object's preference for any resource content within the first historical time period; Determine preferred resource content from the multiple resource contents, and push the preferred resource content to the target object; The target preference score of the target object for the preferred resource content is greater than the target preference score of the target object for other resource content.

2. The method according to claim 1, wherein Scoring the target object's preference for any resource content based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content to obtain the target object's importance preference score for the any resource content includes: Counting the number of first search behaviors generated by the target object within the first historical time period, and the number of second search behaviors generated by the target object within the first historical time period; An importance preference score of the target object for the any resource content is determined based on the first behavior number, the second behavior number, the first quantity, and the second quantity corresponding to the any resource content.

3. The method according to claim 2, wherein The determining, based on the first behavior number, the second behavior number, the first quantity, and the second quantity corresponding to the any resource content, of the importance preference score of the target object for the any resource content includes: Determining a target word frequency corresponding to the any resource content based on the first behavior count and the second behavior count; the target word frequency is used to indicate: a frequency with which the target object searches for the any resource content within the first historical time period; Determining a reference search frequency corresponding to any resource content based on the first number and the second number corresponding to any resource content; the reference search frequency is used to indicate: a frequency at which any resource content is searched by a reference object within the second historical time period; Based on the target word frequency and reference search frequency corresponding to any resource content, the importance preference score of the target object for any resource content is determined; the importance preference score is positively correlated with the target word frequency, and negatively correlated with the reference search frequency.

4. The method according to claim 3, wherein The determining, based on the target word frequency and the reference search frequency corresponding to the any resource content, the importance preference score of the target object for the any resource content includes: Determining an inverse document frequency corresponding to any resource content according to a reference search frequency corresponding to any resource content; wherein the inverse document frequency is negatively correlated with the reference search frequency; According to the target word frequency and inverse document frequency corresponding to any resource content, the word frequency-inverse document frequency corresponding to any resource content is determined, and the word frequency-inverse document frequency is used as the importance preference score of the target object for any resource content.

5. The method according to claim 1, wherein Scoring the target object's preference for the any resource content based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content to obtain a time decay score of the target object for the any resource content includes: Based on the time difference between the current system time and the search time corresponding to the any resource content, a time decay score of the target object for the any resource content is determined; the time decay score is negatively correlated with the time difference.

6. The method according to claim 5, wherein The target object searches for the second behavior of the any resource content multiple times within the first historical time period, and the number of search times corresponding to the any resource content is multiple; The determining, based on the time difference between the current system time and the search time corresponding to the any resource content, a time decay score of the target object for the any resource content includes: determining, based on time differences between the current system time and respective search times corresponding to the respective resource contents, a time decay sub-score of the target object for the respective resource contents at the respective search times; The time decay score of the target object for the any resource content is determined according to the time decay sub-score of the target object for the any resource content at the respective search times.

7. The method according to claim 1, wherein The method of obtaining any resource content among the multiple resource contents includes: Obtaining the search text entered into the input box by the target object within the first historical time period; Determining a degree of difference between the search text and each of the plurality of preset resource contents; The preset resource content indicated by the minimum difference among the plurality of preset resource contents is determined as the any one resource content.

8. The method according to claim 7, wherein Determining the difference between the search text and any preset resource content among the plurality of preset resource contents includes: Determining an edit distance between the search text and the content identifier of any preset resource content; the edit distance is used to indicate: a minimum number of edits required to convert the search text into the content identifier of any preset resource content; The edit distance is used as the difference between the search text and any preset resource content.

9. A resource content push device, characterized in that: include: an acquisition unit, configured to acquire multiple resource contents searched by a target object within a first historical time period and the search time corresponding to each resource content, and acquire a first number of reference objects that generated search behavior within a second historical time period, and a second number of reference objects that searched for each resource content within the second historical time period; the second historical time period includes the first historical time period; a processing unit configured to score the target object's preference for any resource content among the multiple resource contents based on the search time corresponding to the any resource content, the first quantity, and the second quantity corresponding to the any resource content, to obtain a time decay score and an importance preference score of the target object for the any resource content; The time decay score is used to indicate the influence of the search time corresponding to any resource content on the preference of the target object for the resource content, and the importance preference score is used to indicate the importance of any resource content to the target object; The processing unit is further configured to determine a target preference score of the target object for the any resource content based on the time decay score and the importance preference score; The target preference score is used to indicate: the target object's preference for any resource content within the first historical time period; The processing unit is further configured to determine preferred resource content from the multiple resource contents and push the preferred resource content to the target object; the target object's target preference score for the preferred resource content is greater than the target object's target preference score for other resource contents.

10. A resource content push device, characterized in that: The resource content push device includes an input interface and an output interface, and also includes: a processor adapted to implement one or more instructions; and A computer storage medium storing one or more instructions, wherein the one or more instructions are suitable for being loaded by the processor and executing the resource content pushing method according to any one of claims 1 to 8.

11. A computer storage medium, characterized in that The computer storage medium stores computer program instructions, which, when executed by a processor, are used to execute the resource content pushing method according to any one of claims 1 to 8.

12. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, it is used to load and execute the resource content pushing method according to any one of claims 1 to 8.

Images