CN111899047A

CN111899047A - Resource recommendation method and device, computer equipment and computer-readable storage medium

Info

Publication number: CN111899047A
Application number: CN202010674091.2A
Authority: CN
Inventors: 耿晓东
Original assignee: Rajax Network Technology Co Ltd
Current assignee: Rajax Network Technology Co Ltd
Priority date: 2020-07-14
Filing date: 2020-07-14
Publication date: 2020-11-06

Abstract

The invention discloses a resource recommendation method, a resource recommendation device, computer equipment and a computer readable storage medium, and relates to the technical field of Internet. The method comprises the following steps: when detecting that a user to be recommended accesses a target subsystem, acquiring real-time behavior data and historical behavior data of the user to be recommended; extracting at least one target resource supply scheme from a plurality of preset resource supply schemes according to the real-time behavior data and the historical behavior data; and determining the resources to be recommended based on at least one target resource supply scheme, and recommending the resources to be recommended to the users to be recommended.

Description

Resource recommendation method and device, computer equipment and computer-readable storage medium

Technical Field

The present invention relates to the field of internet technologies, and in particular, to a resource recommendation method and apparatus, a computer device, and a computer-readable storage medium.

Background

With the coming of the internet era, terminal devices such as smart phones, tablet computers, smart watches and the like are in endlessly, and become essential communication and entertainment tools in daily life of people gradually. In order to adapt to the current rhythm of life and enable the life of people to be more intelligent, a platform for maintaining the relationship between customers and stores is produced. Any entity store can be accessed into the platform, the platform can integrate the resources of the accessed entity store, and the resources which are possibly interested by the user are recommended to the user in the platform, so that the daily requirements of the user are met, the user is not required to passively browse the content which is disliked by the user, the exposure rate of the entity store can be increased, and the sales volume of the entity store is increased.

In the related art, the platform is divided into a plurality of subsystems, and each subsystem has different functions, for example, some subsystems provide purchase of comprehensive products, some subsystems provide purchase of various coupons and consumption tickets in physical stores, and some subsystems provide takeout services. When the resource recommendation is carried out for the user, the platform can count historical behaviors of the user in each subsystem in the past, determine the interest and hobbies of the user according to the counted historical behaviors, customize corresponding resources for the user according to the interest and hobbies, and recommend the resources to the user.

In the process of implementing the invention, the inventor finds that the related art has at least the following problems:

some users may only generate historical behaviors in some subsystems in the platform and consume the historical behaviors once, but for other subsystems, the users may never consume the historical behaviors, belong to new users of other subsystems, the historical behaviors of the users in other subsystems are blank, if the users use functions provided by other subsystems, the platform can only manually recommend resources for the users, the manual work not only involves a large amount of workload, but also causes that the resources recommended to the new users are not accurate enough, and has huge deviation with the actual needs and intentions of the new users, so that the new users cannot be really transformed, and the loss of the users is caused.

Disclosure of Invention

In view of the above, the present invention provides a resource recommendation method, device, computer device, and computer readable storage medium, and mainly aims to solve the problem that the current platform can only manually recommend resources for the user, and the manual work not only involves a large amount of workload, but also causes inaccurate resources recommended to a new user, has a huge deviation from the actual needs and intentions of the new user, and cannot truly convert the new user, thereby causing the loss of the user.

According to a first aspect of the present invention, there is provided a resource recommendation method, including:

when it is detected that a user to be recommended accesses a target subsystem, acquiring real-time behavior data and historical behavior data of the user to be recommended, wherein the real-time behavior data is behavior data generated when the user to be recommended currently accesses the target subsystem, and the historical behavior data is behavior data generated by the user to be recommended in other subsystems except the target subsystem;

extracting at least one target resource supply scheme from a plurality of preset resource supply schemes according to the real-time behavior data and the historical behavior data, wherein the at least one target resource supply scheme is a preset resource supply scheme matched with the real-time behavior data and the historical behavior data in the plurality of preset resource supply schemes;

and determining resources to be recommended based on the at least one target resource supply scheme, and recommending the resources to be recommended to the user to be recommended.

In another embodiment, the determining a resource to be recommended based on the at least one target resource supply scheme, and recommending the resource to be recommended to the user to be recommended includes:

reading the resources indicated by the at least one target resource supply scheme as the resources to be recommended;

classifying the resources to be recommended to obtain at least one resource group;

for each resource group in the at least one resource group, determining a meeting place theme corresponding to the resource group, and acquiring an entrance identifier of the meeting place theme;

establishing an incidence relation between a resource identifier of the resource to be recommended included in the resource group and the entrance identifier, and linking the resource to be recommended included in the resource group into a target position indicated by the entrance identifier based on the incidence relation;

and determining the display sequence of the resources to be recommended included in the resource group, and displaying the resources to be recommended included in the resource group at the target position according to the display sequence.

In another embodiment, the extracting at least one target resource supply scheme from a plurality of preset resource supply schemes according to the real-time behavior data and the historical behavior data includes:

acquiring a plurality of preset resource supply schemes, and determining a plurality of preset behaviors included in the plurality of preset resource supply schemes, wherein each preset resource supply scheme in the plurality of preset resource supply schemes at least includes a preset behavior and a preset resource type;

determining a target preset behavior consistent with the real-time behavior data and the historical behavior data in the plurality of preset behaviors, and taking a preset resource supply scheme to which the target preset behavior belongs as a candidate resource supply scheme;

determining a policy priority for each preset resource supply scheme included in the candidate resource supply scheme based on the real-time behavior data and the historical behavior data;

and sorting preset resource supply schemes included in the candidate resource supply schemes according to the order of the strategy priorities from high to low, and taking a preset number of preset resource supply schemes ranked at the top in the candidate resource supply schemes as the at least one target resource supply scheme.

In another embodiment, the determining the policy priority of each preset resource provisioning scheme included in the candidate resource provisioning scheme based on the real-time behavior data and the historical behavior data includes:

determining at least one related resource of the real-time behavior data and the historical behavior data respectively, wherein the at least one related resource is a resource browsed or consumed or collected or shared in the real-time behavior data and the historical behavior data;

for each relevant resource in the at least one relevant resource, counting a first triggering time of the relevant resource in the real-time behavior data and the historical behavior data, wherein the first triggering time is used for indicating the time of browsing or consuming or collecting or sharing the relevant resource in the real-time behavior data and the historical behavior data;

and determining a preset resource supply scheme matched with the related resource in the candidate resource supply schemes, and taking the first triggering times as the strategy priority of the preset resource supply scheme matched with the related resource, wherein the preset behavior of the preset resource supply scheme matched with the related resource comprises the text content of the related resource.

In another embodiment, the reading the resource indicated by the at least one target resource supply scheme as the resource to be recommended includes:

reading at least one preset resource type included in the at least one target resource supply scheme, and taking a resource indicated by the at least one preset resource type as the resource to be recommended; and/or the presence of a gas in the gas,

when the resource style specified in the at least one target resource supply scheme is read, extracting a specified resource conforming to the resource style from the resources indicated by the at least one preset resource type, and taking the specified resource as the resource to be recommended, wherein the resource style at least comprises a commodity style, a virtual card style and a multimedia style.

In another embodiment, the classifying the resources to be recommended to obtain at least one resource group includes:

determining the resource type of the resource to be recommended, dividing the resource to be recommended with consistent resource types into the same resource group, and obtaining at least one resource group; or the like, or, alternatively,

and inquiring the activity type bound to the resource to be recommended, and dividing the resource to be recommended with consistent activity types into the same resource group to obtain the at least one resource group.

In another embodiment, the determining the presentation order of the resources to be recommended included in the resource group includes:

querying a second triggering frequency of the resource to be recommended included in the resource group in the real-time behavior data and the historical behavior data, wherein the second triggering frequency is used for indicating the frequency of browsing, consuming, collecting or sharing the resource to be recommended included in the resource group in the real-time behavior data and the historical behavior data;

and sequencing the resources to be recommended included in the resource group according to the sequence of the second triggering times from high to low to obtain the display sequence.

In another embodiment, the method further comprises:

counting, for each preset resource supply plan of the plurality of preset resource supply plans, a number of orders of a resource recommended based on the preset resource supply plan and at least one order time within a specified historical time period;

deleting the preset resource supply scheme from the plurality of preset resource supply schemes when the order-placing times are lower than a time threshold value;

and when the order placing times are not lower than the time threshold, acquiring a time parameter specified in a preset behavior included in the preset resource supply scheme, constructing a predicted order placing time period by using the at least one order placing time, and updating the content of the time parameter to the predicted order placing time period.

According to a second aspect of the present invention, there is provided a resource recommendation apparatus, comprising:

the system comprises an acquisition module, a recommendation module and a recommendation module, wherein the acquisition module is used for acquiring real-time behavior data and historical behavior data of a user to be recommended when the user to be recommended is detected to access a target subsystem, the real-time behavior data is behavior data generated when the user to be recommended currently accesses the target subsystem, and the historical behavior data is behavior data generated by the user to be recommended in other subsystems except the target subsystem;

an extracting module, configured to extract at least one target resource supply scheme from a plurality of preset resource supply schemes according to the real-time behavior data and the historical behavior data, where the at least one target resource supply scheme is a preset resource supply scheme that matches the real-time behavior data and the historical behavior data in the plurality of preset resource supply schemes;

and the recommending module is used for determining the resource to be recommended based on the at least one target resource supply scheme and recommending the resource to be recommended to the user to be recommended.

In another embodiment, the recommendation module includes:

a reading unit, configured to read a resource indicated by the at least one target resource provisioning scheme as the resource to be recommended;

the classification unit is used for classifying the resources to be recommended to obtain at least one resource group;

the acquisition unit is used for determining a meeting place theme corresponding to the resource group for each resource group in the at least one resource group and acquiring an entrance identifier of the meeting place theme;

the establishing unit is used for establishing an incidence relation between the resource identifier of the resource to be recommended included in the resource group and the entrance identifier, and linking the resource to be recommended included in the resource group into the target position indicated by the entrance identifier based on the incidence relation;

and the display unit is used for determining the display sequence of the resources to be recommended included in the resource group and displaying the resources to be recommended included in the resource group at the target position according to the display sequence.

In another embodiment, the extraction module includes:

an obtaining unit, configured to obtain the multiple preset resource supply schemes, and determine multiple preset behaviors included in the multiple preset resource supply schemes, where each preset resource supply scheme in the multiple preset resource supply schemes at least includes a preset behavior and a preset resource type;

a first determining unit, configured to determine, among the plurality of preset behaviors, a target preset behavior that is consistent with the real-time behavior data and the historical behavior data, and use a preset resource supply scheme to which the target preset behavior belongs as a candidate resource supply scheme;

a second determining unit, configured to determine, based on the real-time behavior data and the historical behavior data, a policy priority of each preset resource supply scheme included in the candidate resource supply scheme;

and the sorting unit is used for sorting preset resource supply schemes included in the candidate resource supply schemes according to the order of the strategy priorities from high to low, and taking a preset number of preset resource supply schemes which are arranged at the head in the candidate resource supply schemes as the at least one target resource supply scheme.

In another embodiment, the second determining unit is configured to determine at least one related resource of the real-time behavior data and the historical behavior data, where the at least one related resource is a resource browsed or consumed or collected or shared in the real-time behavior data and the historical behavior data; for each relevant resource in the at least one relevant resource, counting a first triggering time of the relevant resource in the real-time behavior data and the historical behavior data, wherein the first triggering time is used for indicating the time of browsing or consuming or collecting or sharing the relevant resource in the real-time behavior data and the historical behavior data; and determining a preset resource supply scheme matched with the related resource in the candidate resource supply schemes, and taking the first triggering times as the strategy priority of the preset resource supply scheme matched with the related resource, wherein the preset behavior of the preset resource supply scheme matched with the related resource comprises the text content of the related resource.

In another embodiment, the reading unit is configured to read at least one preset resource type included in the at least one target resource provisioning scheme, and use a resource indicated by the at least one preset resource type as the resource to be recommended; and/or when the resource style specified in the at least one target resource supply scheme is read, extracting a specified resource conforming to the resource style from the resources indicated by the at least one preset resource type, and taking the specified resource as the resource to be recommended, wherein the resource style at least comprises a commodity style, a virtual card style and a multimedia style.

In another embodiment, the classification unit is configured to determine a resource type of the resource to be recommended, and divide the resource to be recommended with the consistent resource type into the same resource group to obtain the at least one resource group; or inquiring the activity type bound to the resource to be recommended, and dividing the resource to be recommended with consistent activity type into the same resource group to obtain the at least one resource group.

In another embodiment, the presentation unit is configured to query a second number of times of triggering, in the real-time behavior data and the historical behavior data, of a resource to be recommended included in the resource group, where the second number of times of triggering is used to indicate a number of times that the resource to be recommended included in the resource group is browsed or consumed or collected or shared in the real-time behavior data and the historical behavior data; and sequencing the resources to be recommended included in the resource group according to the sequence of the second triggering times from high to low to obtain the display sequence.

In another embodiment, the apparatus further comprises:

the counting module is used for counting the order placing times and at least one order placing time of the recommended resources based on the preset resource supply schemes in a specified historical time period for each preset resource supply scheme in the plurality of preset resource supply schemes;

a deletion module configured to delete the preset resource supply scheme from the plurality of preset resource supply schemes when the number of times of ordering is lower than a number threshold;

and the updating module is used for acquiring a time parameter specified in a preset behavior included in the preset resource supply scheme when the order placing times are not lower than the time threshold, constructing a predicted order placing time period by adopting the at least one order placing time, and updating the content of the time parameter into the predicted order placing time period.

According to a third aspect of the present invention, there is provided a computer device comprising a memory storing a computer program and a processor implementing the steps of the method of the first aspect when the processor executes the computer program.

According to a fourth aspect of the present invention, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of the first aspect described above.

By the technical scheme, the invention provides a resource recommendation method, a device, computer equipment and a computer readable storage medium, when detecting that a user to be recommended accesses a target subsystem, the invention obtains real-time behavior data and historical behavior data of the user to be recommended, determines a target resource supply scheme more suitable for the current resource recommendation of the user in a plurality of preset resource supply schemes by combining the current real-time behavior data of the user in the subsystem and the historical behavior data of the user in other subsystems, and executes the resource recommendation of the user according to the target resource supply scheme without manually interfering in the resource recommendation process, thereby saving a large amount of workload, ensuring that the resource recommended for the user is closer to the current actual needs and intentions of the user, successfully attracting the user when the user accesses, and the new user is really converted, so that the user loss is avoided.

The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

FIG. 1 is a flowchart illustrating a resource recommendation method according to an embodiment of the present invention;

FIG. 2A is a flowchart illustrating a resource recommendation method according to an embodiment of the present invention;

FIG. 2B is a schematic diagram illustrating an interaction of a resource recommendation method according to an embodiment of the present invention;

FIG. 3A is a schematic structural diagram illustrating a resource recommendation apparatus according to an embodiment of the present invention;

FIG. 3B is a schematic structural diagram illustrating a resource recommendation apparatus according to an embodiment of the present invention;

FIG. 3C is a schematic structural diagram of a resource recommendation apparatus according to an embodiment of the present invention;

fig. 3D is a schematic structural diagram illustrating a resource recommendation apparatus according to an embodiment of the present invention;

fig. 4 shows a schematic device structure diagram of a computer apparatus according to an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

An embodiment of the present invention provides a resource recommendation method, as shown in fig. 1, the method includes:

101. when it is detected that a user to be recommended accesses a target subsystem, real-time behavior data and historical behavior data of the user to be recommended are obtained, the real-time behavior data are behavior data generated when the user to be recommended currently accesses the target subsystem, and the historical behavior data are behavior data generated by the user to be recommended in other subsystems except the target subsystem.

102. And extracting at least one target resource supply scheme from the plurality of preset resource supply schemes according to the real-time behavior data and the historical behavior data, wherein the at least one target resource supply scheme is a preset resource supply scheme matched with the real-time behavior data and the historical behavior data in the plurality of preset resource supply schemes.

103. And determining the resources to be recommended based on at least one target resource supply scheme, and recommending the resources to be recommended to the users to be recommended.

According to the method provided by the embodiment of the invention, when the fact that the user to be recommended accesses the target subsystem is detected, the real-time behavior data and the historical behavior data of the user to be recommended are obtained, the real-time behavior data of the user to be recommended in the subsystem and the historical behavior data of the user in other subsystems are combined, the target resource supply scheme more suitable for the current resource recommendation of the user is determined in a plurality of preset resource supply schemes, the resource recommendation of the user is executed according to the target resource supply scheme, manual intervention in the resource recommendation process is not needed, a large amount of workload is saved, the fact that the resource recommended for the user is closer to the current actual needs and intentions of the user is ensured, the user is successfully attracted when the user accesses, the new user is really converted, and the user loss is avoided.

An embodiment of the present invention provides a resource recommendation method, as shown in fig. 2A, the method includes:

201. when it is detected that a user to be recommended accesses a target subsystem, real-time behavior data and historical behavior data of the user to be recommended are obtained.

At present, a plurality of platforms providing an online service function for an entity store are divided into a plurality of subsystems, each subsystem has different functions, for example, some subsystems can provide purchase of comprehensive products, some subsystems can provide purchase of various coupons and consumption tickets in the entity store, and some subsystems can provide take-out services, and the subsystems are presented on a terminal of a user by different front-end applications, so that the user can determine which front-end application to access according to the current requirement. The inventors have recognized that while these subsystems are managed by platform integration, it is likely that a user will only use the functionality provided by one or more of the subsystems, not use the functionality of other subsystems in addition to, or even have no access to, the other subsystems. These other subsystems, if it is desired to provide a user with resources that the user may be interested in when the user first accesses the system, may not be able to do so because the user's historical behavior data in the system is blank. The current means for solving the problem is to determine the resource recommendation which may be interested in the subsystem which is accessed by the user for the first time by means of the historical behavior data of the user in other subsystems of the platform, but since the user accesses the subsystem which has never been accessed, it is indicated that the current requirement of the user may not be solved by the subsystems which have been accessed by the user for the first time, and the resource recommendation of the user by means of the historical behavior data of the user in other subsystems of the platform is likely to deviate from the current purpose of the user greatly. The process needs to be completed manually, the workload is huge, manual subjective judgment is involved, errors of recommended resources are increased, the subsystem cannot be helped to convert the users who visit for the first time, and loss of the users who visit for the first time can be caused. Therefore, the invention provides a resource recommendation method, which presets a plurality of different preset resource supply schemes, determines which preset resource supply scheme is more suitable for the current resource recommendation of a user by combining the real-time behavior data of the user in a subsystem and the historical behavior data of the user in other subsystems when detecting that the user accesses the subsystem for the first time, and executes the resource recommendation of the user according to the preset strategy, so that the resource recommended for the user is closer to the current requirement of the user, the user is successfully attracted when the user accesses the subsystem for the first time, and the conversion rate of a new user of the system is improved. It should be noted that, in the embodiment of the present invention, a user to be recommended may be a user who accesses a target subsystem for the first time, may also be a user who registers in the target subsystem for the first time, or may also be a user who accesses a platform for the first time or registers in the platform, and the like.

Therefore, when it is detected that the user to be recommended accesses the target subsystem, the real-time behavior data and the historical behavior data of the user to be recommended need to be acquired, so that resource recommendation of the user to be recommended is realized on the basis of the real-time behavior data and the historical behavior data in the following. The real-time behavior data is behavior data generated when the user to be recommended accesses the target subsystem currently, and specifically may include commodities, articles, advertisements and the like browsed, collected and shared when the user accesses the target subsystem currently. In the embodiment of the present invention, the real-time behavior data is taken as an example of the behavior data that the user currently generates in the target system, and in the actual application process, the real-time behavior data may be the very short-term behavior of the user, such as today, yesterday, and the data in the target subsystem last day, and generally does not exceed 3 days. The historical behavior data is behavior data generated by the user to be recommended in other subsystems except the target subsystem, and specifically may include commodities, articles, advertisements and the like that the user browses, collects and shares in other subsystems, and may also include commodities that the user purchases in other subsystems. Considering that the historical behavior data of the user in other subsystems is more, if the historical behavior data are referred to, the historical behavior data of the user in other subsystems in the past week, the past month, the past half year or the past year can be extracted for subsequent operations. The history period for acquiring the historical behavior data is not particularly limited.

202. The method comprises the steps of obtaining a plurality of preset resource supply schemes, determining a plurality of preset behaviors included in the plurality of preset resource supply schemes, determining a target preset behavior consistent with real-time behavior data and historical behavior data in the plurality of preset behaviors, and taking the preset resource supply scheme to which the target preset behavior belongs as a candidate resource supply scheme.

In the embodiment of the invention, in order to determine how to determine which resources are recommended to a user to be recommended, a plurality of preset resource supply schemes are preset in a platform, and then according to the acquired real-time behavior data and historical behavior data of the user to be recommended, which preset resource supply scheme is adopted to determine the recommended resources for the user to be recommended is determined in the plurality of preset resource supply schemes. Each preset resource supply scheme in the plurality of preset resource supply schemes at least comprises a preset behavior and a preset resource type, for example, the preset behavior included in the preset resource supply scheme a may be that "the user has searched the fast food XX in the target subsystem on the current day", and the preset resource type may be that "the fast food resource"; the preset behavior included in the preset resource supply scheme B may be "the user purchased the fitness resource in other subsystems", and the preset resource type may be "salad type dish or fitness type resource". It should be noted that, in consideration of the large number of the preset resource supply schemes, in order to uniformly manage the preset resource supply schemes, the preset resource supply schemes may be numbered, and a plurality of preset resource supply schemes may be stored in a table manner, which is specifically referred to in table 1 below.

The reference value of the preset resource supply scheme during resource recommendation subsequently exists only if the acquired real-time behavior data and historical behavior data of the user to be recommended hit preset behaviors specified in the preset resource supply scheme, and the preset resource supply scheme in which the specified preset behaviors do not appear in the real-time behavior data and the historical behavior data of the user to be recommended at all does not help the resource recommendation process, so that the user to be recommended can be screened out. Therefore, it is required to obtain a plurality of preset resource supply schemes, determine a plurality of preset behaviors included in the plurality of preset resource supply schemes, determine a target preset behavior consistent with the real-time behavior data and the historical behavior data among the plurality of preset behaviors, and use the preset resource supply scheme to which the target preset behavior belongs as a candidate resource supply scheme, so as to determine which preset resource supply scheme is suitable for recommending resources for a current user to be recommended in the candidate resource supply scheme.

203. The method comprises the steps of determining the strategy priority of each preset resource supply scheme included in candidate resource supply schemes based on real-time behavior data and historical behavior data, sequencing the preset resource supply schemes included in the candidate resource supply schemes according to the sequence of the strategy priority from high to low, and taking the preset resource supply scheme with the preset number arranged at the head in the candidate resource supply schemes as at least one target resource supply scheme.

In the embodiment of the invention, the candidate resource supply scheme is not completely suitable for resource recommendation of the current user to be recommended, if the candidate resource supply scheme comprises two preset resource supply schemes, one of the two preset resource supply schemes is summarized that the user browses fast food XX currently, fast food resources are recommended for the user; the other is that the user previously purchases fitness products in other subsystems, and salad dishes are recommended for the user. And the user to be recommended browses the fast food XX only when the user wants to buy the fast food XX today, so that if the real-time behavior data of the user browses the fast food XX, the possibility that the user to be recommended wants to buy fast food resources today is higher, so that the two preset resource supply schemes are compared, the former preset resource supply scheme is more advantageous and has higher priority than the latter preset resource supply scheme, therefore, after the candidate resource supply scheme is determined, the policy priority of each preset resource supply scheme included in the candidate resource supply scheme needs to be determined based on the real-time behavior data and the historical behavior data, the preset resource supply schemes included in the candidate resource supply scheme are sorted according to the order of the policy priority from high to low, the preset resource supply scheme with the preset number ranked at the top in the candidate resource supply scheme is taken as at least one target resource supply scheme, therefore, the target resource supply scheme with the highest strategy priority is used as the resource supply scheme which is referred to first, and the accuracy of the resources recommended to the user to be recommended is guaranteed to the greatest extent.

When the policy priority of each preset resource supply scheme included in the candidate resource supply scheme is determined, at least one related resource of the real-time behavior data and the historical behavior data is determined respectively, wherein the at least one related resource is a resource which is browsed or consumed or collected or shared in the real-time behavior data and the historical behavior data. For example, if the real-time behavior data shows that the user to be recommended is browsing fast food XX, and the historical behavior data shows that the user to be recommended has purchased fitness ZZ in other subsystems, both the fast food XX and the fitness ZZ can be used as related resources. Then, for each related resource in the at least one related resource, counting a first triggering time of the related resource in the real-time behavior data and the historical behavior data, wherein the first triggering time is used for indicating the time of browsing or consuming or collecting or sharing the related resource in the real-time behavior data and the historical behavior data. For example, if it is displayed in the real-time behavior data that the user to be recommended is browsing the fast food XX, and the historical behavior data shows that the user to be recommended has purchased the fast food XX in other subsystems, the first trigger time of the fast food XX is 2 times. And finally, determining a preset resource supply scheme matched with the related resource in the candidate resource supply schemes, and taking the first triggering times as the strategy priority of the preset resource supply scheme matched with the related resource, wherein the preset behavior of the preset resource supply scheme matched with the related resource comprises the text content of the related resource. For example, assuming that the relevant resource is "fast food XX", and the preset resource supply scheme a includes a preset behavior that "the user has searched fast food XX in the target subsystem on the current day", the preset resource supply scheme matching the fast food XX is the preset resource supply scheme a, the first trigger time of the fast food XX is 2 times, and the policy priority of the preset resource supply scheme a may be determined to be 2.

Through the above process, a policy priority is determined for each preset resource provisioning scheme included in the candidate resource provisioning schemes, and the determination of the policy priority is substantially to select the target resource provisioning scheme according to the cold start rule. In the actual application process, the policy priority of the preset resource supply scheme matched with the relevant resources extracted from the real-time behavior data can be directly set to be higher than the policy priority of the preset resource supply scheme matched with the relevant resources extracted from the historical behavior data, the preset resource supply schemes corresponding to the relevant resources extracted from the real-time behavior data and the historical behavior data are distinguished, and the policy priorities are respectively determined. Or the matching degree of each preset resource supply scheme included in the candidate resource supply scheme with the real-time behavior data and the historical behavior data of the user to be recommended can be respectively calculated based on a matching degree algorithm, and the matching degree is used as the strategy priority. Or the policy priority of each preset resource supply scheme can be manually set in advance, and the candidate resource supply schemes are directly marked according to the set policy priority. The present invention does not specifically limit the manner of determining the policy priority of each preset resource provisioning scheme included in the candidate resource provisioning scheme.

204. And determining the resource to be recommended based on at least one target resource supply scheme.

In the embodiment of the present invention, after determining at least one target resource supply scheme, the resource to be recommended may be determined based on the at least one target resource supply scheme. Each preset resource supply scheme at least comprises a preset behavior and a preset resource type, so that at least one preset resource type included in at least one target resource supply scheme can be read, and the resource indicated by the at least one preset resource type is taken as the resource to be recommended. For example, assuming that at least one preset resource type included in the at least one target resource supply scheme is salad vegetables and fitness products, resources such as salmon salad, vegetable salad, dumbbell, yoga mat and the like may be used as the resource to be recommended.

It should be noted that the types of the resources to be recommended are diversified, for example, the resources may be commercial resources such as salmon salad, vegetable salad, dumbbell, etc., virtual card resources such as red envelope, coupon, membership card, etc., or multimedia resources such as bought order, food video, fitness music, etc., so some target resource supply schemes may limit the resource types, for example, some target resource supply schemes may limit the resources that only recommend music types to the user. Therefore, in the process of practical application, when the resource style specified in at least one target resource supply scheme is read, the specified resource conforming to the resource style is extracted from the resources indicated by at least one preset resource type, and the specified resource is taken as the resource to be recommended, wherein the resource style at least comprises a commodity style, a virtual card style and a multimedia style. If any target resource supply scheme does not limit the resource style, all the determined resources can be directly used as the resources to be recommended.

In the practical application process, the number of items of the target resource supply scheme determined according to the real-time behavior data and the historical behavior data of the user to be recommended is likely to be large, the amount of the resource to be recommended determined based on the target resource supply scheme is large, and the user is puzzled by the intrusive resource recommendation of the user to be recommended, so that the resource to be recommended can be determined by extracting the target resource supply scheme with the priority of the strategy ranked in the first few places, the resource amount of the resource to be recommended is reduced, and the resource to be recommended is refined. For example, the target resource supply schemes with the policy priorities ranked in the first three or five are extracted to determine the resources to be recommended, and the like.

205. And recommending the resources to be recommended to the user to be recommended.

In the embodiment of the invention, after the resources to be recommended are determined, the resources to be recommended can be recommended to the user. When resources to be recommended are recommended to a user to be recommended, for a front-end application on a user terminal to be recommended, the resources to be recommended need to be displayed at a certain position of a page, and the user to be recommended is attracted to trigger the resources to be recommended. However, considering that if the resources to be recommended are rubbed together and displayed to the user to be recommended, for example, if the first few displayed resources to be recommended are salmon salad and vegetable salad, and then the next vegetable salad is changed into fried chicken, the display logic changes suddenly, so that the user may be immersed in the selection of salad and may ignore even if the user is very interested in fried chicken, so that the resources to be recommended can be classified, the resources to be recommended with the same characteristics are arranged in the same resource group and displayed together, and the consistency of the display logic is ensured. When the resources to be recommended are classified to obtain at least one resource group, two ways can be adopted. One way is to classify the resources to be recommended according to the resource types of the resources to be recommended, determine the resource types of the resources to be recommended, divide the resources to be recommended with consistent resource types into the same resource group, and obtain at least one resource group. For example, salmon salad, vegetable salad, and fruit salad are all divided into resource groups with salad-like dish resource types, and fried chicken, fried potato chips, fried meat shashlik, and the like are divided into resource groups with fried food resource types. The other mode is that the resources to be recommended are classified according to the activity types of the current binding activities of the resources to be recommended, the activity types bound by the resources to be recommended are inquired, the resources to be recommended with consistent activity types are divided into the same resource group, and at least one resource group is obtained. For example, assuming that the activities bound by salmon salad, fried chicken and french fries are all "new user exemption", salmon salad, fried chicken and french fries are divided into the same resource group.

After at least one resource group is obtained through division, in order to express the characteristics of the resource groups and draw the attention of a user to be recommended, for each resource group in the at least one resource group, a meeting place theme corresponding to the resource group needs to be determined, and an entrance identifier of the meeting place theme is obtained. And then, establishing an incidence relation between the resource identifier of the resource to be recommended included in the resource group and the entrance identifier, and linking the resource to be recommended included in the resource group into the target position indicated by the entrance identifier based on the incidence relation. For example, for a resource group whose bound activities are all "new user exempt from list", the meeting place theme of the resource group may be "new user exempt from list", the entry ID of the meeting place theme is obtained as the entry identifier, the association relationship between the entry ID and the resource identifier of salmon salad, fried chicken and fried potato included in the resource group is established, and based on the association relationship, the destination position indicated by the entry ID of salmon salad, fried chicken and fried potato is entered into the page. The purpose of establishing the association relationship is to directly determine which resources are displayed for the user to be recommended when the user to be recommended triggers the target entrance corresponding to the entrance ID, and in the actual application process, some resources may be pre-stored in a certain position of some other meeting place in advance, so that the meeting place ID of the resource can also be determined, and the association relationship between the entrance ID and the meeting place ID of the resource is established, so that the user to be recommended can directly enter the meeting place where the resource is located to access when triggering the resource, and can also indirectly recommend the content in the meeting place. Specifically, the established association relationship may be stored using table 2.

It should be noted that, in the above description, the resource to be recommended is divided into at least one resource group for display, and in the process of practical application, a meeting place theme of "new person explosives" may also be added under the catalog of "new person home page", and the determined resource to be recommended is linked to the meeting place theme for display, so that the user to be recommended can view the resource customized for the user in a unified manner.

In addition, when resources included in a resource group are displayed, the matching degrees of different resources and the current requirements of the user to be recommended also have gaps, some resources may be mistaken for the user to be recommended and are not actually interested by the user to be recommended, so for each resource group, when the resources to be recommended in the resource group are displayed, the display sequence of the resources to be recommended included in the resource group also needs to be determined, and the resources to be recommended included in the resource group are displayed at the target position according to the display sequence. When the display sequence is determined, the second triggering times of the resources to be recommended included in the resource group in the real-time behavior data and the historical behavior data can be inquired, and the resources to be recommended included in the resource group are sorted according to the sequence from high to low of the second triggering times to obtain the display sequence. Specifically, the second triggering number is used for indicating the number of times that the resource to be recommended included in the resource group is browsed or consumed or collected or shared in the real-time behavior data and the historical behavior data. That is, considering that the user browses, purchases, collects, or shares only the resource of interest for a plurality of times, the resource with the largest number of times related to the real-time behavior data and the historical behavior data of the user to be recommended is displayed at the top. Or, in practical application, considering that the real-time behavior data is more convincing than the historical behavior data, the resource to be recommended determined according to the real-time behavior data may be arranged at the front position of the display sequence, and when a plurality of resources to be recommended determined according to the real-time behavior data exist, the resource to be recommended with a high number of triggering times is arranged at the front position of the display sequence with reference to the number of triggering times of the resource to be recommended in the real-time behavior data. The present invention does not specifically limit the manner of determining the display order.

Through the process, the resource recommendation operation of the user to be recommended is realized. However, for the user and the platform, the interests and the data are constantly changed along with the time, and the constant preset resource supply scheme is likely to gradually adapt to the interests and the change of the data, so an update optimization mechanism for a plurality of preset resource supply schemes is further provided in the present invention, and how to update or optimize the preset resource supply scheme is determined by counting the operation quality of each preset resource supply scheme. Specifically, for each of the plurality of preset resource supply schemes, the number of times of placing orders of the resource recommended based on the preset resource supply scheme and at least one time of placing orders within a specified historical period may be counted. When the next time count is lower than the time threshold, the preset resource supply scheme is deleted from the plurality of preset resource supply schemes, that is, if the recommended resource determined for the user based on a certain preset resource supply scheme is rarely triggered by the user, even if the triggering time reaches the minimum, it may be determined that the existence sense of the preset resource supply scheme is very low, and the preset resource supply scheme may be filtered. In the process of practical application, a preset number can be set, and the preset resource supply scheme with the lowest order-placing frequency is deleted, for example, the preset number can be 2, and then two preset resource supply schemes with the lowest order-placing frequency are deleted. The present invention does not specifically limit the timing of deleting the preset resource supply scheme.

And when the next time is not lower than the time threshold, acquiring a time parameter specified in a preset behavior included in the preset resource supply scheme, constructing a predicted order placing time period by adopting at least one order placing time, and updating the content of the time parameter into the predicted order placing time period. That is, the resources recommended according to the preset resource supply scheme are often purchased by the user within at least one ordering time, and the time parameters specified in the preset behaviors included in the preset resource supply scheme need to be updated according to the at least one ordering time, so that the preset resource supply scheme is refined. For example, assuming that the preset resource supply scheme Q includes a preset behavior that "the user accesses the system at 7 to 9 points" and the preset resource type is "breakfast class", but it is statistically determined that the predicted next time period constructed based on at least one next time corresponding to the bread, porridge, steamed stuffed bun, etc. resources recommended to the user by the preset resource supply scheme Q is substantially between 7 to 8 points, the preset resource type recommended to the user when the user accesses the system at 7 to 8 points is "breakfast class" by updating "7 to 9 points" in the preset behavior included in the preset resource supply scheme Q to "7 to 8 points". It should be noted that, if the preset behavior included in the preset resource supply scheme does not have a specified time parameter, the predicted next time period may not be constructed for updating, and the present invention is not particularly limited to this. In practical application, the above-mentioned ways of updating the preset resource supply schemes can be summarized to establish a horse racing mechanism, and the recommendation effect of each preset resource supply scheme is supervised and updated through the horse racing mechanism, so that the timeliness of updating is ensured.

In the embodiment of the invention, an interaction process exists, no matter real-time behavior data or historical behavior data are collected by a terminal, the whole process of determining the resource supply scheme and determining the resource to be recommended is completed by a server, and the resource to be recommended is further displayed in a meeting place or is executed by the terminal, so the whole interaction process is summarized as follows:

referring to fig. 2B, when detecting that the user to be recommended accesses the target subsystem, the terminal acquires real-time behavior data and historical behavior data of the user to be recommended, and transmits the real-time behavior data and the historical behavior data to the server. After receiving the real-time behavior data and the historical behavior data, the server determines candidate resource supply schemes matched with the real-time behavior data and the historical behavior data in a plurality of preset resource supply schemes, and sequences the candidate resource supply schemes to obtain at least one target resource supply scheme. And then, the server starts to determine the resources to be recommended based on at least one target resource supply scheme and pushes the resources to be recommended to the terminal. After receiving the resources to be recommended, the terminal groups the resources to be recommended to obtain at least one resource group, determines a meeting place theme corresponding to the resource group, obtains an entrance identifier of the meeting place theme, and links the resources to be recommended included in the resource group into a target position indicated by the entrance identifier, so that the resources to be recommended really reach the users to be recommended.

Further, as a specific implementation of the method shown in fig. 1, an embodiment of the present invention provides a resource recommendation apparatus, and as shown in fig. 3A, the apparatus includes: an acquisition module 301, an extraction module 302 and a recommendation module 303.

The obtaining module 301 is configured to, when it is detected that a user to be recommended accesses a target subsystem, obtain real-time behavior data and historical behavior data of the user to be recommended, where the real-time behavior data is behavior data generated when the user to be recommended currently accesses the target subsystem, and the historical behavior data is behavior data generated by the user to be recommended in other subsystems except the target subsystem;

the extracting module 302 is configured to extract at least one target resource supply scheme from a plurality of preset resource supply schemes according to the real-time behavior data and the historical behavior data, where the at least one target resource supply scheme is a preset resource supply scheme that matches the real-time behavior data and the historical behavior data in the plurality of preset resource supply schemes;

the recommending module 303 is configured to determine a resource to be recommended based on the at least one target resource supply scheme, and recommend the resource to be recommended to the user to be recommended.

In a specific application scenario, as shown in fig. 3B, the recommending module 303 includes: a reading unit 3031, a classification unit 3032, an acquisition unit 3033, an establishment unit 3034 and a presentation unit 3035.

The reading unit 3031 is configured to read a resource indicated by the at least one target resource provisioning scheme as the resource to be recommended;

the classification unit 3032 is configured to classify the resources to be recommended to obtain at least one resource group;

the obtaining unit 3033 is configured to determine, for each resource group in the at least one resource group, a meeting place topic corresponding to the resource group, and obtain an entry identifier of the meeting place topic;

the establishing unit 3034 is configured to establish an association relationship between the resource identifier of the resource to be recommended included in the resource group and the entry identifier, and link the resource to be recommended included in the resource group to the target position indicated by the entry identifier based on the association relationship;

the display unit 3035 is configured to determine a display order of the resources to be recommended included in the resource group, and display the resources to be recommended included in the resource group in the target position according to the display order.

In a specific application scenario, as shown in fig. 3C, the extracting module 302 includes: an acquisition unit 3021, a first determination unit 3022, a second determination unit 3023, and a sorting unit 3024.

The obtaining unit 3021 is configured to obtain the multiple preset resource supply schemes, and determine multiple preset behaviors included in the multiple preset resource supply schemes, where each preset resource supply scheme in the multiple preset resource supply schemes at least includes a preset behavior and a preset resource type;

the first determining unit 3022 is configured to determine a target preset behavior that is consistent with the real-time behavior data and the historical behavior data among the plurality of preset behaviors, and use a preset resource supply scheme to which the target preset behavior belongs as a candidate resource supply scheme;

the second determining unit 3023, configured to determine a policy priority of each preset resource supply scheme included in the candidate resource supply scheme based on the real-time behavior data and the historical behavior data;

the sorting unit 3024 is configured to sort the preset resource supply schemes included in the candidate resource supply schemes in order from high to low in the policy priority, and use a preset number of preset resource supply schemes ranked first in the candidate resource supply schemes as the at least one target resource supply scheme.

In a specific application scenario, the second determining unit 3023 is configured to determine at least one related resource of the real-time behavior data and the historical behavior data, where the at least one related resource is a resource browsed or consumed or collected or shared in the real-time behavior data and the historical behavior data; for each relevant resource in the at least one relevant resource, counting a first triggering time of the relevant resource in the real-time behavior data and the historical behavior data, wherein the first triggering time is used for indicating the time of browsing or consuming or collecting or sharing the relevant resource in the real-time behavior data and the historical behavior data; and determining a preset resource supply scheme matched with the related resource in the candidate resource supply schemes, and taking the first triggering times as the strategy priority of the preset resource supply scheme matched with the related resource, wherein the preset behavior of the preset resource supply scheme matched with the related resource comprises the text content of the related resource.

In a specific application scenario, the reading unit 3031 is configured to read at least one preset resource type included in the at least one target resource provisioning scheme, and use a resource indicated by the at least one preset resource type as the resource to be recommended; and/or when the resource style specified in the at least one target resource supply scheme is read, extracting a specified resource conforming to the resource style from the resources indicated by the at least one preset resource type, and taking the specified resource as the resource to be recommended, wherein the resource style at least comprises a commodity style, a virtual card style and a multimedia style.

In a specific application scenario, the classifying unit 3032 is configured to determine a resource type of the resource to be recommended, and divide the resource to be recommended with the consistent resource type into the same resource group to obtain the at least one resource group; or inquiring the activity type bound to the resource to be recommended, and dividing the resource to be recommended with consistent activity type into the same resource group to obtain the at least one resource group.

In a specific application scenario, the presentation unit 3035 is configured to query a second triggering time of the resource to be recommended included in the resource group in the real-time behavior data and the historical behavior data, where the second triggering time is used to indicate a time that the resource to be recommended included in the resource group is browsed or consumed or collected or shared in the real-time behavior data and the historical behavior data; and sequencing the resources to be recommended included in the resource group according to the sequence of the second triggering times from high to low to obtain the display sequence.

In a specific application scenario, as shown in fig. 3D, the apparatus further includes: a statistics module 304, a deletion module 305, and an update module 306.

The counting module 304 is configured to count, for each preset resource supply scheme in the plurality of preset resource supply schemes, the number of orders of the recommended resources based on the preset resource supply scheme and at least one order time within a specified historical time period;

the deleting module 305 is configured to delete the preset resource supply scheme from the plurality of preset resource supply schemes when the number of orders is lower than a threshold number;

the updating module 306 is configured to, when the order placing frequency is not lower than the frequency threshold, obtain a time parameter specified in a preset behavior included in the preset resource supply scheme, construct a predicted order placing time period by using the at least one order placing time, and update the content of the time parameter to the predicted order placing time period.

According to the device provided by the embodiment of the invention, when the fact that the user to be recommended accesses the target subsystem is detected, the real-time behavior data and the historical behavior data of the user to be recommended are obtained, the real-time behavior data of the user to be recommended in the subsystem and the historical behavior data of the user in other subsystems are combined, the target resource supply scheme more suitable for the current resource recommendation of the user is determined in a plurality of preset resource supply schemes, the resource recommendation of the user is executed according to the target resource supply scheme, manual intervention in the resource recommendation process is not needed, a large amount of workload is saved, the fact that the resource recommended for the user is closer to the current actual needs and intentions of the user is ensured, the user is successfully attracted when the user accesses, a new user is really converted, and the user loss is avoided.

It should be noted that other corresponding descriptions of the functional units related to the resource recommendation device provided in the embodiment of the present invention may refer to the corresponding descriptions in fig. 1 and fig. 2A to fig. 2B, and are not described herein again.

In an exemplary embodiment, referring to fig. 4, there is further provided a device, where the device 400 includes a communication bus, a processor, a memory, and a communication interface, and may further include an input/output interface and a display device, where the functional units may communicate with each other through the bus. The memory stores computer programs, and the processor is used for executing the programs stored in the memory and executing the resource recommendation method in the embodiment.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the resource recommendation method.

Through the above description of the embodiments, those skilled in the art will clearly understand that the present application can be implemented by hardware, and also by software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.), and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the implementation scenarios of the present application.

Those skilled in the art will appreciate that the figures are merely schematic representations of one preferred implementation scenario and that the blocks or flow diagrams in the figures are not necessarily required to practice the present application.

Those skilled in the art will appreciate that the modules in the devices in the implementation scenario may be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or may be located in one or more devices different from the present implementation scenario with corresponding changes. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.

The above application serial numbers are for description purposes only and do not represent the superiority or inferiority of the implementation scenarios.

The above disclosure is only a few specific implementation scenarios of the present application, but the present application is not limited thereto, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present application.

Claims

1. A resource recommendation method, comprising:

2. The method according to claim 1, wherein the determining a resource to be recommended based on the at least one target resource supply scheme, and recommending the resource to be recommended to the user to be recommended comprises:

3. The method according to claim 1, wherein the extracting at least one target resource supply scheme from a plurality of preset resource supply schemes according to the real-time behavior data and the historical behavior data comprises:

4. The method according to claim 3, wherein the determining the policy priority of each preset resource provisioning scheme included in the candidate resource provisioning scheme based on the real-time behavior data and the historical behavior data comprises:

5. The method according to claim 2, wherein the reading the resource indicated by the at least one target resource supply scheme as the resource to be recommended comprises:

6. The method according to claim 2, wherein the classifying the resources to be recommended to obtain at least one resource group comprises:

7. The method of claim 2, wherein the determining the presentation order of the resources to be recommended included in the resource group comprises:

8. The method of claim 1, further comprising:

9. A resource recommendation device, comprising:

10. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 8 when executing the computer program.