CN110020135B

CN110020135B - Demand determination method, resource recommendation method and related device

Info

Publication number: CN110020135B
Application number: CN201711099014.3A
Authority: CN
Inventors: 戴铮; 苏丁; 叶苏俐; 李江
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2017-11-09
Filing date: 2017-11-09
Publication date: 2023-01-31
Anticipated expiration: 2037-11-09
Also published as: CN110020135A

Abstract

The embodiment of the invention discloses a demand determining method, which can divide a plurality of user groups through resource obtaining records, wherein users in the same user group have the same resource quality demand, the attention degree of the user group to the target resource relative to other groups to the same type of resource can be determined according to the target attention proportion of the users in one user group to the target resource and the target attention proportion of other user groups to the target resource, when the attention degree of one user group to the target resource is higher, the user group can be determined to have the demand on the target resource, so that the demand of the users in the user group to the target resource can be determined, namely the target resource can be more concerned, and the demand of the users in the user group can be effectively determined in a computing mode taking the user group as a unit. In addition, the embodiment of the application also discloses a resource recommendation method and a related device.

Description

Demand determination method, resource recommendation method and related device

Technical Field

The present invention relates to the field of data processing, and in particular, to a demand determining method, a resource recommending method, and a related apparatus.

Background

With the development of internet technology, a large amount of resources are provided on the internet, and a user can browse or acquire resources meeting the needs of the user through the internet.

For a resource provider or an internet platform server for providing resource access to a user, if the requirement of the user using the internet can be accurately determined, resources can be provided and recommended for the user in a targeted manner, so that the use efficiency of the internet is improved.

However, there is currently no way to efficiently determine the needs of a user.

Disclosure of Invention

In order to solve the technical problem, the invention provides a requirement determining method, a resource recommending method and a related device, which can effectively determine the requirements of users in a user group.

The embodiment of the invention discloses the following technical scheme:

in a first aspect, an embodiment of the present application provides a demand determining method, where the method includes:

acquiring a plurality of user groups divided according to resource acquisition records of users, wherein users in the same user group have the same resource quality requirement;

calculating target attention proportions corresponding to each user group respectively according to the attention condition of users in the user groups to target resources, wherein the target attention proportions comprise the proportion of users in one user group who have attention to the target resources to the total number of the users in the user group;

calculating the attention degree of each user group to the target resource according to the target attention proportion corresponding to each user group;

and taking the user group with the attention degree to the target resource meeting the preset condition as the user group with the requirement on the target resource.

Optionally, the concern condition includes browsing, collecting, shopping cart adding, purchasing or evaluating behaviors of a target resource, where the target resource is a resource or a category of resources.

Optionally, the calculating the attention degree of each user group to the target resource according to the target attention ratio respectively corresponding to each user group includes:

aiming at a first user group, calculating attention difference indexes between the first user group and other user groups according to a target attention ratio of the first user group and target attention ratios of the other user groups, wherein the first user group is one of the user groups, and the other user groups are user groups except the first user group;

obtaining a linear fitting slope of the first user group according to attention difference indexes between the first user group and other user groups;

and performing hierarchical clustering according to the linear fitting slope of each user group to obtain the attention degree of each user group to the target resource.

Optionally, the obtaining is divided into a plurality of user groups according to the resource obtaining record of the user, and includes:

aiming at a first user, acquiring a resource acquisition record of the first user, wherein the resource acquisition record of the first user comprises transfer characteristic values when the first user acquires different resources;

for a first resource acquired by the first user, calculating a resource acquisition score of the first user for the first resource according to the transfer characteristic value of the first user for the first resource and the average transfer characteristic value for the first resource;

obtaining scores according to the resources of the first user aiming at the different resources, and determining the comprehensive score of the first user;

and carrying out group division according to the comprehensive scores of all the users to obtain the plurality of user groups.

Optionally, the method further includes:

determining a second user group to which a second user belongs, wherein the second user group is one of the user groups;

and recommending resources to the second user according to the resources required by the second user group, wherein the resources required by the second user group comprise the target resources.

Optionally, the method further includes:

determining a third user group to which a third user belongs;

and recommending resources to the third user according to resources required by a fourth user group, wherein the fourth user group is one of the user groups, the resource quality requirement of the fourth user group is closest to the resource quality requirement of the third user group, the resource quality requirement of the fourth user group is greater than the resource quality requirement of the third user group, and the resources required by the fourth user group comprise the target resources.

In a second aspect, an embodiment of the present application provides a demand determination apparatus, which includes an obtaining unit, a first calculating unit, a second calculating unit, and a confirming unit:

the acquisition unit is used for acquiring a plurality of user groups divided according to the resource acquisition records of the users, and the users in the same user group have the same resource quality requirements;

the first calculating unit is used for calculating a target attention proportion corresponding to each user group according to the attention condition of users in the user groups to target resources, wherein the target attention proportion comprises the proportion of users in one user group who have attention to the target resources to the total number of the users in the user group;

the second calculating unit is used for calculating the attention degree of each user group to the target resource according to the target attention proportion corresponding to each user group;

the confirming unit is used for taking the user group with the attention degree to the target resource meeting the preset condition as the user group with the requirement on the target resource.

Optionally, the second computing unit is further configured to:

calculating, for a first user group, attention difference indicators between the first user group and other user groups according to a target attention ratio of the first user group and target attention ratios of the other user groups, where the first user group is one of the user groups, and the other user groups are user groups other than the first user group;

Optionally, the obtaining unit is further configured to:

and carrying out group division according to the comprehensive scores of all the users to obtain a plurality of user groups.

Optionally, the apparatus further includes a recommendation unit:

the recommending unit is configured to determine a second user group to which a second user belongs, where the second user group is one of the multiple user groups; and recommending resources to the second user according to the resources required by the second user group, wherein the resources required by the second user group comprise the target resources.

Optionally, the recommending unit is further configured to determine a third user group to which a third user belongs; and recommending resources to the third user according to resources required by a fourth user group, wherein the fourth user group is one of the user groups, the resource quality requirement of the fourth user group is closest to the resource quality requirement of the third user group, the resource quality requirement of the fourth user group is greater than the resource quality requirement of the third user group, and the resources required by the fourth user group comprise the target resources.

In a third aspect, an embodiment of the present application provides a processing device for requirement determination, where the processing device includes a processor and a memory, where:

the memory is used for storing program codes and transmitting the program codes to the processor;

the processor is used for executing the following method according to the instructions in the program code:

Optionally, the processor is further configured to execute the following method according to the instructions in the program code:

determining a second user group to which a second user belongs, wherein the second user group is one of the plurality of user groups;

determining a third user group to which a third user belongs;

In a fourth aspect, an embodiment of the present application provides a resource recommendation method, where the method includes:

determining a resource acquisition record of a target user, determining a first user group where the target user is located, wherein users in the same user group have the same resource quality requirement;

and recommending resources to the target user according to the resources which are required by a second user group, wherein the resource quality requirement of the second user group is close to the resource quality requirement of the first user group, and the resource quality requirement of the second user group is greater than the resource quality requirement of the first user group.

Optionally, the determining the resource obtaining record of the target user determines a first user group where the target user is located, including:

acquiring a resource acquisition record of the target user, wherein the resource acquisition record of the target user comprises transfer characteristic values of the target user when acquiring different resources;

aiming at target resources acquired by the target user, calculating a resource acquisition score of the target user for the target resources according to the transfer characteristic value of the target user for the target resources and the average transfer characteristic value of the target user for the target resources;

obtaining scores according to the resources of the target user aiming at the different resources, and determining the comprehensive score of the target user;

and determining that the target user is in the first user group according to the comprehensive score of the target user, wherein the comprehensive score of the target user is in the comprehensive score range of the first user group.

Optionally, the method further includes:

and recommending resources to the target user according to the resources required by the first user group.

In a fifth aspect, an embodiment of the present application provides a resource recommendation apparatus, where the apparatus includes a determining unit and a recommending unit:

the determining unit is used for determining a resource acquisition record of a target user to determine a first user group where the target user is located, and users in the same user group have the same resource quality requirement;

the recommendation unit is configured to recommend resources to the target user according to resources that a second user group has a demand, where a resource quality demand of the second user group is close to a resource quality demand of the first user group, and the resource quality demand of the second user group is greater than the resource quality demand of the first user group.

Optionally, the determining unit is further configured to:

Optionally, the recommending unit is further configured to recommend resources to the target user according to the resources that the first user group has a demand.

In a sixth aspect, an embodiment of the present application provides a processing device for resource recommendation, where the processing device includes a processor and a memory, where:

acquiring a resource acquisition record of the target user, wherein the resource acquisition record of the target user comprises transfer characteristic values of the target user when the target user acquires different resources;

aiming at the target resources acquired by the target user, calculating the resource acquisition scores of the target user aiming at the target resources according to the transfer characteristic values of the target user aiming at the target resources and the average transfer characteristic values aiming at the target resources;

and recommending resources to the target user according to the resources which are required by the first user group.

According to the technical scheme, a plurality of user groups can be divided through the resource acquisition records, users in the same user group have the same resource quality requirement, the attention degree of the user group to the target resource, which is different from that of other groups to the same type of resource, can be determined according to the target attention proportion of the users in one user group to the target resource and the target attention proportion of other user groups to the target resource, when the attention degree of one user group to the target resource is higher, the user group can be determined to have the requirement on the target resource, so that the requirement of the users in the user group to the target resource can be determined, namely the target resource can be more concerned, and the requirement of the users in the user group can be effectively determined in a computing mode taking the user group as a unit.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a flowchart of a demand determination method according to an embodiment of the present invention;

fig. 2 is a schematic diagram of resources required by different user groups according to an embodiment of the present invention;

fig. 3 is a schematic diagram of resources required by different user groups according to another embodiment of the present invention;

FIG. 4 is a flowchart of another demand determination method according to an embodiment of the present invention;

FIG. 5 is a flowchart of another demand determination method according to an embodiment of the present invention;

fig. 6 is a flowchart of a resource recommendation method according to an embodiment of the present invention;

fig. 7 is a device structure diagram of a demand determination device according to an embodiment of the present application;

fig. 8 is a block diagram of an apparatus of a processing device for demand determination according to an embodiment of the present application;

fig. 9 is a device structure diagram of a resource recommendation device according to an embodiment of the present application;

fig. 10 is a device structure diagram of a processing apparatus for resource recommendation according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments, but not all embodiments, of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The resources provided on the internet may include physical resources or may include virtual resources. The physical resource may be an article having a physical structure, such as home appliances, clothes, cosmetics, and the like. The virtual resource can be a virtualized product, such as a virtual article, a game coin, an electronic voucher, and the like in a game, and the virtual resource can also be a service, such as a door cleaning service, a door oil smoke cleaning service, and the like provided by a cleaning worker.

Because the resources provided on the internet are many, the user can conveniently browse or search the required resources, the user time is saved, the resources provided on the internet can be classified, and the resources with the same attribute can be classified into a category. Categories are understood to be generic to resources that can be classified as a class, all of which have at least one attribute in common. For example, "shoes" may be a category that may include various resources having the attribute of "shoes", such as athletic shoes, casual shoes, sandals, canvas shoes, or more detailed specific resources of various brand sizes, etc. For example, "canvas shoes" may also be a category that may include various resources having the attribute of "canvas shoes", such as specific resources of brand a canvas shoes, brand b canvas shoes, flannel, high top canvas shoes, and the like. How to classify the resources to obtain the categories is not limited in the present invention, and the resources may be classified according to the classification accuracy, or according to the specific requirements,

A feasible resource recommendation method can be that according to the historical transaction records of the users, an automatic personalized recommendation algorithm is adopted to recommend products with prices similar to those in the historical transaction records to the users. However, in the actual life demand, the user pays more attention to the resources because the resources are not only based on the price, for example, the user purchases a shampoo, and the user may pay attention to the efficacy of the shampoo, such as oil control, dandruff removal and the like, but also not only because of the price of the product. The price may be a factor that affects the user's attention to the resource, but is not a decisive factor that affects the user's demand, so the above-mentioned method of recommending the resource based on the price is not good enough to meet the actual demand of the user.

Therefore, embodiments of the present invention provide a method for determining a requirement, a method for recommending a resource, and a related device, which can divide a plurality of user groups by using a resource acquisition record, where users in the same user group have the same resource quality requirement, and according to a target attention ratio of users in one user group to a target resource and a target attention ratio of other user groups to the target resource, a degree of attention of the user group to the target resource, which is different from that of other groups to the same type of resource, can be determined, and when the degree of attention of one user group to the target resource is higher, the user group can be determined to have a requirement for the target resource, so that it can be determined that users in the user group have a requirement for the target resource, that is, the target resource is paid more attention, and it can be seen that a calculation manner using the user group as a unit can effectively determine the requirement of users in the user group.

The following describes the requirement determining method provided by the embodiment of the invention in detail. As shown in fig. 1, a flowchart of a demand determination method provided in an embodiment of the present invention is shown, where the method includes:

s101: and acquiring a plurality of user groups divided according to the resource acquisition records of the users, wherein the users in the same user group have the same resource quality requirement.

For example, in the embodiment of the present invention, the resource obtaining record may be a record of a user obtaining a resource on the internet, for example, a transaction record of a user purchasing a resource.

Because the resource acquisition record of one user can reflect the quality requirement of the user on the resource to a certain extent, the resource quality requirement of the user can be analyzed according to the resource acquisition record of the user, so that the user can be divided into a plurality of user groups according to the difference of the resource quality requirements of the user, and the users in the same user group can have the same or similar resource quality requirements.

S102: and calculating the target attention proportion corresponding to each user group according to the attention condition of the users in the user group to the target resource.

For example, the target resource may be a resource or a category of resource, and the form of the target resource may be determined according to a specific computing requirement, which is not limited in the embodiment of the present invention.

The attention condition of the user to the target resource reflects whether the user has a demand for the target resource and the level of the demand degree, the attention condition of the user to the target resource can be reflected by browsing, collecting, shopping cart adding, purchasing or evaluating behaviors of the user to the target resource, taking the target resource as a shampoo as an example, the attention condition of the user to the shampoo can include that the user browses a page of the shampoo, and can also include that the user purchases the shampoo through the internet.

The attention condition of each user in the user group to the target resource can reflect the demand trend of the user group to the target resource to a certain extent, and in order to facilitate the intuitive understanding of the attention condition of the user group to the target resource, the attention ratio can be expressed by a target attention ratio, wherein the target attention ratio comprises the ratio of users in the user group who have attention to the target resource to the total number of users in the user group. The higher the target attention proportion of one user group is, the higher the proportion of the users concerned about the target resource in the user group to the total users in the user group is, the better the demand trend of the user group to the target resource is. Next, a description is given for one of the behaviors, and taking a browsing behavior as an example, the target attention ratios respectively corresponding to the user groups can be calculated according to the ratio of the users who have browsed the target resource in the user group to the total number of users in the user group. For example, the user group a includes 100 users, and for one target resource, if 80 users have browsed the target resource, the browsing ratio is 80/100=0.8, and the calculated browsing ratio of 0.8 may be used as the target attention ratio of the user group a to the target resource. The target attention ratio corresponding to the acquired behavior is similar to the target attention ratio calculation method corresponding to the browsing behavior, and is not described herein again.

Through S102, the target attention ratios corresponding to the same target resource for the multiple user groups may be determined. Assuming that the plurality of user groups are specifically a group a, a group b, and a group c, and the target resource is a shampoo, a target attention ratio (e.g., 0.2) of the group a to the shampoo, a target attention ratio (e.g., 0.3) of the group b to the shampoo, and a target attention ratio (e.g., 0.8) of the group c to the shampoo may be determined.

S103: and calculating the attention degree of each user group to the target resource according to the target attention proportion corresponding to each user group.

Taking one of the user groups as an example, when calculating the attention degree of the user group to the target resource, the attention degree of the user group to the target resource needs to be comprehensively considered, and the attention degree of the user group to the target resource is calculated according to the target attention ratio of the user group to the target resource, so that the attention degree of the user group to the target resource can reflect the attention degree of the user group to the target resource relative to other user groups in the user groups. The higher the attention degree of the user group to the target resource is, the higher the requirement of the user group to the target resource is compared with other user groups, that is, the target resource more conforms to the requirement of the user group.

S104: and taking the user group with the attention degree to the target resource meeting the preset condition as the user group with the requirement on the target resource.

For example, if a user group has a higher attention degree to the target resource, it indicates that the user group has a higher demand for the target resource than other user groups having a lower attention degree, that is, the target resource better meets the overall demand of the user group, or that users belonging to the user group are more likely to have a demand for the target resource, that is, more likely to browse or acquire the target resource. For example, when the plurality of user groups are specifically 10, which are user groups L0 to L9, respectively, and the attention degree of the user group L3 to the target resource is higher than the attention degree of other user groups to the target resource, it may be determined that the user group having a demand for the target resource is L3.

There may be various ways to determine which user group or user groups have a demand for the target resource, for example, the user group with the highest attention degree for the target resource in the plurality of user groups may be determined as the user group having a demand for the target resource; determining a user group with a degree of attention to a target resource exceeding a preset threshold among the plurality of user groups as a user group having a demand for the target resource; the user group with the higher attention degree to the target resource in the plurality of user groups may also be determined as the user group having the demand for the target resource. That is, the preset condition may be set according to specific requirements, may be set to have the highest attention degree, may also be set to be a preset threshold of the attention degree, may also be set to be the top n pieces with higher attention degree, and the like.

For different target resources, by implementing the above-mentioned S101 to S104, a user group having a requirement for each target resource can be determined respectively.

When the target resource is one resource, for example, the shampoo is divided into eight types, i.e., oil control, dandruff removal, amino acid, damage repair, organic pure natural, essential oil shampoo, sea salt particle, silicone oil-free shampoo, and the like according to different effects, and each type of shampoo is one resource. By taking the oil control shampoo as a target resource, a user group with the requirement on the oil control shampoo can be determined according to the method for determining the requirement. By analogy with this method, a user group with a requirement for each resource can be determined, as shown in fig. 2, the plurality of user groups specifically include 5 user groups L1 to L5, respectively, where the user group L1 has a requirement for shampoo for controlling oil and removing dandruff, the user group L2 has a requirement for shampoo for amino acid and repairing damage, the user group L3 has a requirement for shampoo of pure natural organic nature, the user group L4 has a requirement for shampoo of essential oil and sea salt particles, and the user group L5 has a requirement for shampoo of no silicone oil.

When the target resource is a category of resources, for example, apparel may belong to a category, which may include sweaters, pants, T-shirts, or more detailed individual brand size specific resources, and the like. The method comprises the steps of taking clothing as a target resource, determining user groups with requirements on the clothing according to the method for determining the requirements, and determining the user groups with requirements on each resource by analogy with the method, wherein as shown in fig. 3, the user groups are 4 specifically and are user groups L1-L4 respectively, the user group L1 has requirements on popular food, clothing and shoes, the user group L2 has requirements on movie tickets, peripheral tourism and household automobiles, the user group L3 has requirements on organic food, outbound tourism and silicon-free shampoo, and the user group L4 has requirements on treadmills, vacation real estate and yachts.

As can be seen from the above two examples, the requirements of different user groups on resources are different, which mainly reflects the difference in resource quality, that is, the requirements of different user groups on resource quality are different.

It can be seen from the above-mentioned method that, through the resource acquisition record, a plurality of user groups can be divided, users in the same user group have the same resource quality requirement, and according to the target attention proportion of users in one user group to the target resource and the target attention proportion of other user groups to the target resource, the attention degree of the user group to the target resource, which is different from that of other groups to the same type of resource, can be determined, when the attention degree of one user group to the target resource is higher, the user group can be determined to have a requirement on the target resource, so that the user in the user group can be clear of having a requirement on the target resource, that is, paying more attention to the target resource, and the requirement of users in the user group can be effectively determined by using a calculation mode in which the user group is a unit.

It can be known from the above description that, in S103, the attention degree of one user group to the target resource is calculated, and it needs to consider the target attention ratio of other user groups in the multiple user groups to the target resource, that is, the attention degree of one user group to the target resource needs to reflect the difference between the user group and other user groups in the attention condition of the target resource, in the embodiment of the present invention, the difference between the user group and the target resource in the attention condition of other user groups can be represented by a linear fitting slope, so that the attention degree of each user group to the target resource can be obtained according to the linear fitting slope of each user group, and then a manner of obtaining the attention degree of the user group to the target resource according to the linear fitting slope provided by the embodiment of the present invention is specifically described, as shown in fig. 4, specific steps are as follows:

s201: and aiming at a first user group, calculating attention difference indexes between the first user group and other user groups according to the target attention proportion of the first user group and the target attention proportion of other user groups.

For example, the first user group may be any one of the user groups acquired in S101, and the other user groups may be user groups other than the first user group. In S201 and S202, the first user group is taken as an example for description, and the same processing procedure may be applied to the other user groups to obtain the linear fitting slopes corresponding to the other user groups.

In the embodiment of the present invention, a user group having a demand for a target resource needs to be determined, and the target attention proportion of the user group calculated in S102 may reflect the attention situation of the user group to the target resource, that is, the demand trend of the target resource, but for the same target resource, which user group or user groups pay more attention to the target resource needs to compare the attention situation of each user group to the target resource.

Taking the first user group as an example, the attention difference index between the first user group and the other user groups can be obtained by comparing the target attention ratio of the first user group with the target attention ratio of the other user groups. The attention difference index may reflect a difference of the attention situation of the first user group to the target resource compared to the other user groups, and the attention situation may be used to represent a demand trend of the target resource, so the difference of the attention situation of the target resource may also be understood as a difference of the demand trend of the target resource. For example, according to the target attention ratio of the user group L0 and the user group L1, the attention difference index between the user group L0 and the user group L1, which is calculated, may represent the difference of the demand trend of the user group L0 for the target resource relative to the user group L1.

The above-mentioned ways of comparing the target attention ratios may be various, the target attention ratio of the first user group may be subtracted from the target attention ratios of the other user groups, respectively, to obtain corresponding attention difference indicators, or the target attention ratio of the first user group may be divided from the target attention ratios of the other user groups, respectively, to obtain corresponding attention difference indicators, for example, the number of the user groups is 5, which are respectively user groups L0-L4, for the same target resource, the target attention ratios corresponding to the user groups L0-L4 are a0, a1, a2, a3, and a4, respectively, the user group L0 may be divided from the target attention ratios of the other user groups L1-L4, to obtain the attention difference indicators between the user group L0 and the other user groups L1-L4 are a0/a1, a0/a2, a0/a3, and a0/a4, respectively.

S202: and obtaining a linear fitting slope of the first user group according to the attention difference indexes between the first user group and other user groups.

Taking the first user group as an example, the other user groups may be user groups other than the first user group in the plurality of user groups, and the other user groups may be one user group or a plurality of user groups. When the other user groups are multiple user groups, the attention difference indicators between the first user group and the other user groups calculated in S201 may be multiple, for example, the other user groups include a user group a, a user group b, and a user group c, the attention difference indicator A1 between the first user group and the user group a may be obtained according to the target attention ratio of the first user group and the target attention ratio of the user group a, the attention difference indicator A2 between the first user group and the user group b may be obtained according to the target attention ratio of the first user group and the target attention ratio of the user group b, the attention difference indicator A3 between the first user group and the user group c may be obtained according to the target attention ratio of the first user group and the target attention ratio of the user group c, and A1, A2, and A3 are the attention difference indicators between the first user group and the other user groups.

One attention difference index of the first user group can reflect the difference of the demand tendency of the first user group on the target resource compared with one user group in other user groups, when a plurality of user groups exist in the other user groups, the calculated attention difference indexes between the first user group and the other user groups are also multiple, and in order to intuitively reflect the overall difference of the demand tendency of the first user group on the target resource compared with the other user groups, the obtained multiple target difference indexes can be subjected to linear fitting to obtain the linear fitting slope of the first user group. The linear fit slope for a user group may reflect the overall difference in the trend of demand for target resources for that user group as compared to other user groups.

The attention difference index calculated in S201 may be a specific numerical value, and a linear fitting slope obtained by linearly fitting these numerical values is often a curve, and through a variation trend of the curve, the overall difference of the demand trend of the first user group to the target resource compared to other user groups may be more intuitively reflected, for example, the number of the user groups is specifically 5, which are respectively user groups L0 to L4, and for the same target resource, the target attention ratios corresponding to the user groups L0 to L4 are respectively 0.8, 0.6, 0.5, 0.4, and 0.1, the target attention ratios of the user group L0 and the other user groups L1 to L4 may be divided by each other to obtain attention difference indicators between the user group L0 and the other user groups L1 to L4, which are 0.8/0.6=4/3, 0.8/0.5=8/5, 0.8/0.4=2, and 0.8/0.1=8, respectively, and the attention difference indicators are linearly fitted to obtain a linear fitting slope of the user group L0, which is a curve showing an increasing trend, and through the increasing trend of the linear fitting slope, it can be seen that the demand trend of the target resource is higher for the user group L0 than for the other user groups L1 to L4.

For the same target resource, each user group may obtain a corresponding linear fit slope, for example, 10 user groups are respectively user groups L0-L9, and for the same target resource, corresponding target interest ratios are respectively a0-a9, the target interest ratios may be divided to calculate interest variance indicators, the interest variance indicators between the user group L0 and the other user groups L1-L9 are respectively a [0-1] = a0/a1, a [0-2] = a0/a2, a [0-3] = a0/a3, a [0-4] = a0/a4, a [0-5] = a0/a5, a [0-6] = a0/a6, a [0-7] = a0/a7, a [0-8] = a0/a8, and a [0-9] = a 0/9, after linear fitting these attention differential indicators, a linear fitting slope r0 of the user group L0 can be obtained, and similarly, the attention differential indicators between the user group L1 and the other user groups (the user groups except L1 in L0-L9) are respectively B [1-0] = a1/a0, B [1-2] = a1/a2, B [1-3] = a1/a3, B [1-4] = a1/a4, B [1-5] = a1/a5, B [1-6] = a1/a6, B [1-7] = a1/a7, B [1-8] = a1/a8 and B [1-9] = a1/a9, after linear fitting these attention differential indicators, the linear fitting slope r1 of the user group L1 can be obtained, and by analogy, the linear fitting slope corresponding to each user group can be calculated.

S203: and performing hierarchical clustering according to the linear fitting slope of each user group to obtain the attention degree of each user group to the target resource.

In analogy to the determination of the linear fit slope of the first user group, the linear fit slope of each user group can be determined by repeating the operations S201 to S202.

For the same target resource, each user group has a corresponding linear fitting slope, and how many user groups can obtain how many corresponding linear fitting slopes, for example, the number of the user groups is 10 specifically, for the same target resource, each user group has a corresponding linear fitting slope, and thus, 10 corresponding linear fitting slopes can be obtained.

The linear fit slope of one user group may reflect the overall difference of the demand trend of the user group for the target resource compared to other user groups, but for the same target resource, which user group or user groups have demands on the same target resource needs to be comprehensively compared with the linear fit slopes of the user groups. In the embodiment of the invention, a hierarchical clustering mode can be adopted to comprehensively compare the linear fitting slopes of all user groups, so that the attention degree of each user group to the target resource is obtained. The level of the attention degree to the target resource may be used to indicate the level of the demand degree of the user group for the target resource, and taking the first user group as an example, the higher the attention degree of the first user group for the target resource is, which indicates that the first user group has a higher demand for the target resource compared to other user groups.

In the embodiment of the present invention, the acquired plurality of user groups may have been divided in advance according to the resource acquisition record. The acquired user groups can also be obtained by scoring the resource quality requirements of the users according to the resource acquisition records and dividing the user groups according to different comprehensive scores. Next, a method for dividing the user group according to the composite score according to the embodiment of the present invention is described in detail, as shown in fig. 5. The specific steps include S301-S304. When the user group is divided, the larger the number of users is, the more representative the divided user group is. The calculation method of the scores of the users is similar, and the description is given by taking one user as an example.

S301: and aiming at a first user, acquiring a resource acquisition record of the first user, wherein the resource acquisition record of the first user comprises transfer characteristic values when the first user acquires different resources.

For example, the first user may be a user having a resource acquisition record, and the method and the system divide a large number of users into different user groups by analyzing and processing resource acquisition records of the large number of users, where the first user is any one of the large number of users. In S301 to S303, the first user is taken as an example for description, and the same processing flow may be applied to different users to obtain respective composite scores of the different users.

The resource acquisition record may be a record of the user acquiring the resource through the internet, such as a historical transaction record of the resource on the internet platform, specifically, for example, a transaction record of the user a purchasing goods on the elutriation platform all the year round, where the resource may be the goods purchased by the user a.

The transfer characteristic value may be a characteristic value transferred from a characteristic value belonging to the first user to another subject (e.g., a subject providing the resource) in order to obtain the resource, for example, for a commodity transaction through the internet, the first user may transfer a transfer to a merchant providing the commodity in order to obtain the commodity, and the selling price of the commodity is paid. In this example, the transfer characteristic value is the selling price of the article.

S302: and for a first resource acquired by the first user, calculating a resource acquisition score of the first user for the first resource according to the transfer characteristic value of the first user for the first resource and the average transfer characteristic value of the first user for the first resource.

The number of the resources acquired by the first user through the internet may be one or more, and the first resource may be any one of the resources acquired by the first user.

Taking the first resource as an example, the resource acquisition score may be used to indicate a degree of resource quality requirement of the user for the first resource, and a higher resource acquisition score indicates a higher resource quality requirement of the user for the first resource.

In calculating the resource acquisition score of the first resource, the average transfer characteristic value of the first resource may be used as a reference value, and the average transfer characteristic value of the first resource may be understood as an average of transfer characteristic values of the first resource transferred when the first resource is acquired by the user in the internet. For example, when the first resource is the anti-dandruff shampoo, the average amount spent by users on purchasing the anti-dandruff shampoo in the internet can be the average transfer characteristic value of the anti-dandruff shampoo, and the amount spent by one user on purchasing the anti-dandruff shampoo can be the transfer characteristic value of the user for the anti-dandruff shampoo.

When the transfer characteristic value of the first resource is higher than the average transfer characteristic value, which indicates that the resource quality requirement of the first user on the first resource is higher, the score of the corresponding resource acquisition score is higher, and a higher resource acquisition score indicates that the resource quality requirement of the first user on the first resource is higher. Or it can be understood that, when the first user wants to use a transfer characteristic value far higher than the average transfer characteristic value to obtain the first resource, the resource quality of the first resource required by the first user is obviously higher than that of most users, and besides the first resource needs to serve a basic function, a higher user experience is desired, for example, shampoo can serve the basic function of washing hair, and also wants to have additional functions of not damaging hair quality, moistening hair quality, smoothing hair, fragrance and the like during hair washing. However, the shampoo with these functions is obviously more expensive than the common shampoo in general. When the transfer characteristic value of the first resource is lower than or close to the average transfer characteristic value, it indicates that the resource quality demand of the first user on the first resource is relatively low, the score of the corresponding resource acquisition score is relatively low, and the lower the resource acquisition score is, the lower the resource quality demand of the first user on the first resource is. Or, it can be understood that, when the basic and average transfer characteristic values provided by the first user for obtaining the first resource are even and even lower, the requirement of the first user on the resource quality of the first resource is obviously about as much as or even lower than that of most users, for example, only the first resource is expected to have a basic function, for example, shampoo can have a basic function of washing hair.

Taking a historical transaction record of a commodity purchased by a user as an example, the transaction amount of a shampoo (first resource) purchased by the user a (first user) is 200 yuan (transfer characteristic value), the average price of the shampoo purchased by the user a (average transfer characteristic value) can be obtained according to the historical transaction record, the transfer characteristic value is higher than the average transfer characteristic value, it can be shown that the resource quality requirement of the user on the first resource is higher, and the corresponding resource acquisition score is also higher.

S303: and obtaining scores according to the resources of the first user aiming at the different resources, and determining the comprehensive score of the first user.

The resource acquisition scores of other resources are similar to the resource acquisition score calculation method of the first resource, and are not described herein any more, and according to the calculation method of S302, the resource acquisition scores of different resources by the first user can be calculated.

One resource acquisition score only reflects the degree of resource quality requirement of the user on the resource, and the comprehensive score obtained by integrating the degrees of resource quality requirement of the user on different resources can more comprehensively reflect the overall resource quality requirement degree of the user.

Therefore, the resource acquisition scores of different resources of the user can be integrated to determine a comprehensive score, and the comprehensive score is used for representing the resource quality demand degree of the user. The composite score of the first user may be an average score of resource acquisition scores obtained by the first user for different resources, or may be an average score of resource acquisition scores given to some important resources, and a composite score calculated by using the resource acquisition scores and corresponding weights. S304: and carrying out group division according to the comprehensive scores of all the users to obtain a plurality of user groups.

In analogy to the determination of the composite score of the first user, the composite score of each user can be determined by repeatedly performing S301 to S303. When the group division is performed according to the comprehensive scores of the users, the users with the comprehensive scores in the same scoring interval can be divided into a user group. For example, the users with higher composite scores are classified into one user group, and the users with lower composite scores are classified into another user group. Or a group of users with closer comprehensive scores can be divided into a user group. The present invention is not limited to how the user groups are divided according to the composite score.

For example, the transaction amount of a user a purchasing a piece of scarves is 2000 yuan, the average price of the user for purchasing the scarves is 20 yuan according to the historical transaction record, and the transaction amount is higher than the average price, which can indicate that the resource quality requirement of the user a is high to a certain extent, and the transaction amount of the user a for other commodities is basically equal to the average transaction amount. The transaction amount of a car bought by the user B is 50 ten thousand yuan, the average price of the car bought by the user B is 15 ten thousand yuan according to the historical transaction records, the transaction amount is higher than the average price, the requirement of the user B on the quality of resources can be indicated to a certain extent, and the transaction amount of the user B for other commodities is basically equal to the average transaction amount. In this case, although the material income level, the consumption level, etc. of the user a and the user B may not be equivalent, for example, the user a may not have enough income to purchase the car purchased by the user B, but the user a and the user B may show higher resource quality requirement on some levels (for scarves and cars), when determining the composite score for the user a and the user B, the composite score of the user a and the composite score of the user B may be in a similar interval, for example, a percentage may be in an interval of 80-90 minutes, when performing the user group division according to the composite score, the user a and the user B may be divided into a user group, and the user group may show higher resource quality requirement or requirement.

Therefore, the comprehensive score of the user reflects the degree of the user for the resource quality requirement, and the higher the comprehensive score is, the higher the resource quality requirement of the user is, and the lower the comprehensive score is, the lower the resource quality requirement of the user is. In the above steps, the group division is performed according to the comprehensive scores of the users, so that a plurality of user groups with step performance (reaction quality and reaction market trend) can be divided. In a specific implementation, the groups may be divided according to the range of the composite score, that is, each user group may have a corresponding score range, for example, the composite score of each user has a different score value in the range of 1-100, the users may be divided into 10 user groups (L0-L9), each user group from L0 to L9 has a corresponding score range of 1-10, 11-20, 21-30, 31-40, 41-50, 51-60, 61-70, 71-80, 81-90 and 91-100, a higher score indicates a higher demand of the user for the resource quality, and when the composite score of the user a is 50, the user a belongs to the user group L4.

According to the resource recommendation method and device, the user group with the resource requirements can be determined, and the resources required by the user group can be used as the basis for recommending the resources to the user.

In a traditional resource recommendation mode, resources are often recommended to a user based on price, but the price is not a decisive factor influencing the user demand, that is, the recommended resources may not meet the actual demand of the user, so that the resource recommendation efficiency is low.

By the above embodiment of the present application, it can be determined which user groups have requirements for target resources. By determining user groups that have requirements for different resources, it may be determined that each user group has the required resources. When the resources are recommended to the user, the resources can be recommended to the user according to the resources required by the user group where the user is located or according to the resources required by the user group which does not include the user, so that the accuracy and the effectiveness of resource recommendation are improved.

The embodiment of the application provides at least two resource recommendation modes for recommending resources to the user according to the resources required by the user group, wherein the first resource recommendation mode is to recommend the user according to the resources required by the user group where the user is located, and the second resource recommendation mode is to recommend the user according to the resources required by the user group which does not include the user. The following will be described in detail, respectively.

Aiming at a first resource recommendation mode:

the scene of recommending resources may be a provider of resources, or an internet platform service for providing resource access to a user may present recommended resources to the user, for example, when the user logs in a panning platform, the resources recommended to the user are presented in a presentation page. Or recommending resources to the user when the user searches resources, for example, recommending resources related to shampoo to the user according to the search of the user when the user searches "shampoo" on the panning platform.

Taking the second user as an example, a second user group to which the second user belongs may be determined first; and recommending resources to the second user according to the resources required by the second user group.

For example, the second user group may be any one of the user groups acquired in S101, and the resource that the second user group has a requirement includes the target resource.

The users in the same user group have the same or similar resource quality requirements, when a user belongs to a certain user group, it is indicated that the user and the user group have the same or similar resource quality requirements, and then when recommending resources to the user, the resources which the user group to which the user belongs has requirements can be recommended to the user. For example, when the user a searches for shampoo on the panning platform, the user group a has a need for shampoo with oil control and anti-dandruff effects (such as the target resource), and when the user a belongs to the user group a, the shampoo has a high possibility of meeting the actual need of the user a, and the user a can be recommended with the shampoo.

Before resource recommendation is performed, a user group to which a user belongs needs to be determined, taking a second user as an example, when the second user belongs to a user in the user group, the user group to which the second user belongs is already determined when the user group is obtained; when the second user does not belong to the user in the user group, a feasible manner may be to calculate a composite score of the second user according to the operations in S301 to S303, and determine the user group to which the second user belongs according to the composite score.

After the user group to which the user belongs is determined, when resource recommendation is performed, resources that the user group has needs can be recommended to the user, for example, the user a belongs to the user group L2, which indicates that the user a and the user group L2 have the same or similar resource quality needs, and when the user a searches for shampoo on the panning platform, referring to fig. 2, the user group L2 is more biased toward shampoo of amino acid and damage repairing type (such as the target resources), that is, the two types of shampoo better meet the needs of the user group L2, and since the user a and the user group L2 have the same or similar resource quality needs, shampoo of amino acid and damage repairing type can be recommended to the user a.

The resource recommendation is carried out by the method, the recommended resource is the resource meeting the requirements of the user group, so when the resource is recommended to the user belonging to the user group, the resource recommended by the method can better meet the requirements of the user compared with the recommendation based on price because the user and the user group have the same or similar resource quality requirements, thereby being more easily accepted by the user and improving the resource recommendation efficiency.

Aiming at a second resource recommendation mode:

in practical applications, resource providers or internet platform service providers for providing resource access to users often recommend resources to users in a low-price recommendation manner in order to increase the consumption capacity of users, for example, if a user a purchases 20-yuan shampoo, shampoo with a price of less than 20 yuan can be recommended to the users, so that more users can be attracted to purchase the product in a low-price manner.

However, the user pays more attention to resources because the resource quality is not only based on the price but also based on the price, and the recommended resource style is gradually lowered due to the adoption of a low-price recommendation mode, so that the actual requirements of the user cannot be well met, and even the user experience is influenced. In order to avoid the above problem, an alternative way is to increase the price of the recommended goods to increase the recommended format, but in the recommendation method, the recommended resources may exceed the consumption capability of the user and do not meet the actual needs of the user, thereby reducing the efficiency of resource recommendation.

In view of the above problems, cross-user group recommendation may be performed according to an obtained user group, specifically, a third user group to which a third user belongs may be determined, where the third user group is one of the multiple user groups; and recommending resources to the third user according to resources required by a fourth user group, wherein the fourth user group is one of the plurality of user groups, the resource quality requirement of the fourth user group is closest to the resource quality requirement of the third user group, and the resource quality requirement of the fourth user group is greater than the resource quality requirement of the third user group. The fourth group of users has the required resources including the target resource.

The method for determining the third user group to which the third user belongs is similar to the method for determining the second user group to which the second user belongs, and details are not repeated here.

In the embodiment of the invention, a plurality of user groups can be divided in sequence according to the resource acquisition scores, the resource acquisition scores of the adjacent user groups are the closest, and the resource acquisition scores can be used for expressing the degree of the resource quality requirements of the users, so that the resource quality requirements of the adjacent user groups are also the closest. The fourth user group is a user group adjacent to the third user group, and the resource quality requirement of the fourth user group is closest to the resource quality requirement of the third user group compared to other user groups, so that when the resource (such as the target resource) having the requirement of the fourth user group is recommended to the users of the third user group, the recommended resource has a higher possibility of meeting the actual requirement of the user.

And because the resource quality requirement of the fourth user group is greater than the resource quality requirement of the third user group, and the higher the resource quality requirement of the user group is, it indicates that the quality pursuit of the users in the user group for the resource is higher, that is, the tone of the resource required by the user group is also relatively higher, so that the resource required by the fourth group is recommended to the users belonging to the third user group, and the tone of the recommended resource can be improved.

By the cross-user group recommendation mode, the resource recommendation efficiency is guaranteed, the style of recommended resources can be improved, recommended users can know resources concerned by users with higher resource quality requirements, and the attention level of the recommended users is improved.

It should be noted that the second user and the third user may be the same user or different users. When the second user and the third user are the same user, there are various ways of recommending resources. For example, if the user a belongs to the user group L1, and the user group L2 adjacent to the user group L1 (the resource quality requirement of the user group L2 is greater than the resource quality requirement of the user group L1), when recommending the resource to the user a, the resource that the user group L1 has a requirement may be recommended to the user a, the resource that the user group L2 has a requirement may also be recommended to the user a, and the resource that the user group L1 has a requirement and the resource that the user group L2 has a requirement may also be recommended to the user a at the same time.

When the resources of two user groups are recommended to the user at the same time, the weight value of each user group may also be set, for example, if the user a belongs to the user group L1, the user group L2 adjacent to the user group L1 (the resource quality requirement of the user group L2 is greater than the resource quality requirement of the user group L1), the weight value of the user group L1 may be set to 70%, and the weight value of the user group L1 may be set to 30%, then in the recommended resources, the resources belonging to the user group L1 account for 70% of the recommended resources, and the resources belonging to the user group L2 account for 30% of the recommended resources.

The two resource recommendation manners can be implemented in the embodiments corresponding to fig. 1 to 5, and can also be implemented after determining the user group where the user to be recommended is located. The following describes a scheme for implementing resource recommendation after determining a user group where a user to be recommended is located.

Fig. 6 is a flowchart of a method of a resource recommendation method provided in an embodiment of the present application, where the method includes:

s601: determining a resource acquisition record of a target user, determining a first user group where the target user is located, wherein users in the same user group have the same resource quality requirement.

S602: and recommending resources to the target user according to the resources which are required by a second user group, wherein the resource quality requirement of the second user group is close to the resource quality requirement of the first user group, and the resource quality requirement of the second user group is greater than the resource quality requirement of the first user group.

The resource acquisition record may be a record of the user acquiring the resource on the internet, and may be a transaction record of the user purchasing the resource, for example.

In this embodiment, how to determine the resource quality requirement of a user according to the resource acquisition record of the user is not limited. However, in order to improve the accuracy of determining the user group where a user is located, optionally, S601 may be implemented by:

and acquiring a resource acquisition record of the target user, wherein the resource acquisition record of the target user comprises transfer characteristic values of the target user when acquiring different resources.

And aiming at the target resources acquired by the target user, calculating the resource acquisition scores of the target user aiming at the target resources according to the transfer characteristic values of the target user aiming at the target resources and the average transfer characteristic values aiming at the target resources.

And obtaining scores according to the resources of the target user aiming at the different resources, and determining the comprehensive score of the target user.

For the description of the features in the above determination method, reference may be made to the related description in the foregoing embodiments, and details are not described here.

Therefore, by the cross-user group recommendation mode, the resource recommendation efficiency is guaranteed, the style of recommended resources can be improved, recommended users can know resources concerned by users with higher resource quality requirements, and the attention level of the recommended users is improved.

In this embodiment, in addition to performing resource recommendation for the target user across user groups, resources required in the user group where the target user is located may also be recommended to the target user. After S601, resource recommendation may be performed to the target user according to the resource that the first user group has a demand.

Therefore, when the resource is recommended to the user belonging to the user group, the resource recommended according to the method can better meet the requirement of the user compared with the recommendation based on price because the user has the same or similar resource quality requirement as the user group, so that the resource is more easily accepted by the user and the resource recommendation efficiency is improved.

Based on the requirement determining method provided by the embodiment, the embodiment of the application further provides a requirement determining device. Fig. 7 is a device structure diagram of a demand determination device according to an embodiment of the present application, where the device includes an obtaining unit 701, a first calculating unit 702, a second calculating unit 702, and a confirming unit 704:

the acquiring unit 701 is configured to acquire a plurality of user groups divided according to the resource acquisition record of the user, where users in the same user group have the same resource quality requirement;

the first calculating unit 702 is configured to calculate, according to the attention situation of the users in the user groups to the target resource, a target attention proportion corresponding to each user group, where the target attention proportion includes a proportion of users in one user group who have an attention to the target resource to a total number of users in the user group;

the second calculating unit 703 is configured to calculate, according to the target attention ratio respectively corresponding to each user group, the attention degree of each user group to the target resource respectively;

the confirming unit 704 is configured to use a user group whose attention degree to the target resource meets a preset condition as a user group having a demand for the target resource.

Optionally, the second calculating unit is further configured to:

Optionally, the obtaining unit is further configured to:

Optionally, the apparatus further includes a recommendation unit:

the recommending unit is used for determining a second user group to which a second user belongs, wherein the second user group is one of the user groups; and recommending resources to the second user according to the resources required by the second user group, wherein the resources required by the second user group comprise the target resources.

It can be seen that, through the resource obtaining record, a plurality of user groups can be divided, users in the same user group have the same resource quality requirement, according to the target attention proportion of users in one user group to the target resource and the target attention proportion of other user groups to the target resource, the attention degree of the user group to the target resource, which is different from that of other groups to the same type of resources, can be determined, when the attention degree of one user group to the target resource is higher, the user group can be determined to have the requirement on the target resource, so that the user in the user group can be determined to have the requirement on the target resource, that is, the target resource can be more concerned, and the requirement of users in the user group can be effectively determined in a computing mode taking the user group as a unit.

The requirement determining apparatus in the embodiment of the present application is described above from the perspective of a modular functional entity, and a processing device for requirement determination is also provided in the embodiment of the present application, and the processing device in the embodiment of the present application is described below from the perspective of hardware processing.

Fig. 8 is a device structure diagram of a processing device for requirement determination according to an embodiment of the present application, where the processing device 800 includes a processor 802 and a memory 801, where:

the memory 801 is used for storing program codes and transmitting the program codes to the processor 802;

the processor 802 is configured to execute the following method according to instructions in the program code:

determining a third user group to which a third user belongs;

According to the target attention rate of users in one user group to target resources and the target attention rate of other user groups to the target resources, the attention degree of the user group to the target resources, which is different from that of other groups to the same type of resources, can be determined, when the attention degree of one user group to the target resources is higher, the user group can be determined to have the demand on the target resources, so that the user in the user group can be determined to have the demand on the target resources, namely the target resources are concerned more, and the demand of users in the user group can be effectively determined in a computing mode taking the user group as a unit.

Based on the resource recommendation method provided by the embodiment, the embodiment of the application further provides a resource recommendation device. Fig. 9 is a device structure diagram of a resource recommendation device according to an embodiment of the present application, where the device includes a determining unit 901 and a recommending unit 902:

the determining unit 901 is configured to determine a resource obtaining record of a target user, and determine a first user group where the target user is located, where users in the same user group have the same resource quality requirement;

the recommending unit 902 is configured to recommend resources to the target user according to resources that a second user group has a demand, where the resource quality demand of the second user group is close to the resource quality demand of the first user group, and the resource quality demand of the second user group is greater than the resource quality demand of the first user group.

Optionally, the determining unit is further configured to:

The resource recommendation device in the embodiment of the present application is described above from the perspective of the modular functional entity, and a processing device for resource recommendation is further provided in the embodiment of the present application, and the processing device in the embodiment of the present application is described below from the perspective of hardware processing.

Fig. 10 is a device structure diagram of a processing device for resource recommendation according to an embodiment of the present application, where the processing device 1000 includes a processor 1002 and a memory 1001, where:

the memory 1001 is configured to store a program code and transmit the program code to the processor 1001;

the processor 1002 is configured to execute the following method according to instructions in the program code:

Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium may be at least one of the following media: various media that can store program codes, such as a read-only memory (ROM), a RAM, a magnetic disk, or an optical disk.

It should be noted that, in the present specification, all the embodiments are described in a progressive manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus and system embodiments, since they are substantially similar to the method embodiments, they are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, and units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement without inventive effort.

While the invention has been described with reference to specific preferred embodiments, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the following claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for demand determination, the method comprising:

taking a user group with the attention degree to the target resource meeting a preset condition as a user group with a requirement on the target resource;

wherein the method comprises the following steps: determining a third user group to which a third user belongs; and recommending resources to the third user according to resources required by a fourth user group, wherein the fourth user group is one of the user groups, the resource quality requirement of the fourth user group is closest to the resource quality requirement of the third user group, the resource quality requirement of the fourth user group is greater than the resource quality requirement of the third user group, and the resources required by the fourth user group comprise the target resources.

2. The method of claim 1, wherein the concerns include browsing, favoring, shopping cart adding, purchasing, or rating behavior for a target resource, the target resource being a resource or a category of resources.

3. The method according to claim 1, wherein the calculating the attention degree of each user group to the target resource according to the target attention ratio corresponding to each user group comprises:

4. The method of claim 1, wherein the obtaining is divided into a plurality of user groups according to resource obtaining records of users, and comprises:

for a first resource acquired by the first user, calculating a resource acquisition score of the first user for the first resource according to a transfer characteristic value of the first user for the first resource and an average transfer characteristic value of the first user for the first resource;

5. The method of any one of claims 1 to 4, further comprising:

6. A demand determination apparatus, characterized in that the apparatus comprises an acquisition unit, a first calculation unit, a second calculation unit, and a confirmation unit:

the confirming unit is used for taking a user group with the attention degree to the target resource meeting a preset condition as a user group with a requirement on the target resource;

the requirement determining device is further used for determining a third user group to which a third user belongs; and recommending resources to the third user according to resources required by a fourth user group, wherein the fourth user group is one of the user groups, the resource quality requirement of the fourth user group is closest to the resource quality requirement of the third user group, the resource quality requirement of the fourth user group is greater than the resource quality requirement of the third user group, and the resources required by the fourth user group comprise the target resources.

7. A processing device for demand determination, the processing device comprising a processor and a memory, wherein:

the processor is further used for determining a third user group to which a third user belongs; and recommending resources to the third user according to resources required by a fourth user group, wherein the fourth user group is one of the user groups, the resource quality requirement of the fourth user group is closest to the resource quality requirement of the third user group, the resource quality requirement of the fourth user group is greater than the resource quality requirement of the third user group, and the resources required by the fourth user group comprise the target resources.

8. A method for resource recommendation, the method comprising:

9. The method of claim 8, wherein determining the resource acquisition record of the target user determines a first group of users where the target user is located, comprising:

10. The method of claim 8, further comprising:

11. A resource recommendation device, characterized in that the device comprises a determination unit and a recommendation unit:

12. A processing device for resource recommendation, the processing device comprising a processor and a memory, wherein: