CN111859117B

CN111859117B - Information recommendation method and device, electronic equipment and readable storage medium

Info

Publication number: CN111859117B
Application number: CN202010567968.8A
Authority: CN
Inventors: 赵琳琳
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-06-19
Filing date: 2020-06-19
Publication date: 2024-03-26
Anticipated expiration: 2040-06-19
Also published as: CN111859117A

Abstract

The application relates to the technical field of Internet and discloses an information recommendation method, an information recommendation device, electronic equipment and a readable storage medium, wherein the information recommendation method comprises the following steps: acquiring terminal use records of a plurality of users respectively; the terminal use record comprises the use time of each user for a plurality of application programs of the user terminal and the type of the application programs which are correspondingly used; determining a fuzzy relationship between the plurality of users based on the terminal usage records of the plurality of users; classifying a plurality of users based on the fuzzy relation to obtain a classification result; and based on the classification result, sending recommendation information to user terminals of a plurality of users. The information recommendation method provided by the application can improve the accuracy of classification, so that proper information can be recommended to a corresponding user in a proper time period, and accurate recommendation is realized.

Description

Information recommendation method and device, electronic equipment and readable storage medium

Technical Field

The application relates to the technical field of internet, in particular to an information recommendation method, an information recommendation device, electronic equipment and a readable storage medium.

Background

With the development of internet technology, the personalized demands of users are further improved, and users can be classified, so that recommendation information is sent to the users in a personalized way, for example, the users are classified, the office workers in the nine night are recommended to get in the afternoon tea, and the night is recommended to people who stay all night.

At present, preset conditions are generally defined directly, and users are classified according to whether the users meet the preset conditions, for example, five days of night sleep are defined as people who stay up frequently, four days of night sleep are excluded, and the classification mode cannot accurately describe subtle differences on the user behavior level, so that classification results are inaccurate, and time and content of recommending information to the users may not meet user requirements.

Disclosure of Invention

The present application aims to solve at least one of the above technical drawbacks, and specifically proposes the following technical solutions:

in a first aspect, an information recommendation method is provided, including:

acquiring terminal use records of a plurality of users respectively; the terminal use record comprises the use time of each user for a plurality of application programs of the user terminal and the type of the application programs correspondingly used;

determining a fuzzy relationship between the plurality of users based on the terminal usage records of the plurality of users;

Classifying a plurality of users based on the fuzzy relation to obtain a classification result;

and sending recommendation information to user terminals of the plurality of users based on the classification result.

In an alternative embodiment of the first aspect, determining the fuzzy relationship between the plurality of users based on the terminal usage records of the plurality of users includes:

determining the active frequency distribution data of the user in a preset period according to the terminal use record of the user for any user;

a fuzzy relationship between the plurality of users is determined based on the number of active times distribution data of the plurality of users.

In an optional embodiment of the first aspect, the activity number distribution data includes an accumulated activity number of the user terminal for each preset time period within a preset period;

for any user, determining the active frequency distribution data of the user in a preset period based on the terminal use record of the user, wherein the method comprises the following steps:

determining, for any user, the number of activations of the user for each preset time period of each sub-period in a preset period based on a terminal usage record of the user;

based on the number of activities of the user in each preset time period of each sub-period in the preset period, the accumulated number of activities of the user in each preset time period in the preset period is determined.

In an optional embodiment of the first aspect, for any user, determining, based on a terminal usage record of the user, active number distribution data of the user in a preset period includes:

determining initial activity frequency distribution data of the user in a preset period according to a terminal use record of the user for any user;

and carrying out signal denoising on the initial activity frequency distribution data of the user to obtain the activity frequency distribution data of the user.

In an optional embodiment of the first aspect, signal denoising is performed on the initial activity number distribution data of the user to obtain activity number distribution data of the user, including:

determining the median value of each active time value in a preset neighborhood of the active time value according to any active time value in the initial active time distribution data of the user;

replacing the active number value with the determined median value;

and acquiring the activity number distribution data based on each replaced activity number value.

In an alternative embodiment of the first aspect, the fuzzy relationship comprises a similarity fuzzy relationship matrix;

determining a fuzzy relationship between the plurality of users based on the activity number distribution data of the plurality of users, comprising:

Generating an liveness fuzzy set based on the liveness times distribution data of a plurality of user terminals;

a similarity fuzzy relation matrix is determined based on a fuzzy similarity relation between any two predefined activity times distribution data and an activity fuzzy set.

In an optional embodiment of the first aspect, classifying the plurality of users based on the fuzzy relation to obtain a classification result includes:

obtaining a truncated matrix of the similarity fuzzy relation matrix;

a classification result of the plurality of users is determined based on the truncated matrix.

In an optional embodiment of the first aspect, obtaining a truncated matrix of the similarity fuzzy relation matrix comprises:

determining a range of category numbers corresponding to the classification result;

determining a threshold value of the truncated matrix based on the range of the category number;

and obtaining a truncated matrix of the similarity fuzzy relation matrix corresponding to the threshold value.

In an alternative embodiment of the first aspect, determining classification results for a plurality of users based on the truncated matrix includes:

if the truncated matrix is an equivalent matrix, determining a classification result based on the truncated matrix;

if the truncated matrix is not the equivalent matrix, the truncated matrix is adjusted until the adjusted truncated matrix is the equivalent matrix, and a classification result is determined based on the adjusted truncated matrix.

In an optional embodiment of the first aspect, based on the classification result, sending recommendation information to user terminals of the plurality of users comprises:

determining at least one sample user corresponding to each category based on the classification result;

aiming at any sample user, acquiring an information pushing time period of the sample user;

determining recommendation information corresponding to the information push time period;

and in the information pushing time period, sending recommendation information to the users with the same category as the sample user.

In a second aspect, there is provided an information recommendation apparatus including:

the acquisition module is used for respectively acquiring terminal use records of a plurality of users; the terminal use record comprises the use time of each user for a plurality of application programs of the user terminal and the type of the application programs correspondingly used;

the determining module is used for determining fuzzy relations among the plurality of users based on the terminal use records of the plurality of users;

the classification module is used for classifying the plurality of users based on the fuzzy relation to obtain classification results;

and the sending module is used for sending the recommendation information to the user terminals of the plurality of users based on the classification result.

In an optional embodiment of the second aspect, the determining module is specifically configured to, when determining the fuzzy relationship between the plurality of users based on the terminal usage records of the plurality of users:

In an optional embodiment of the second aspect, the activity number distribution data comprises an accumulated activity number of the user terminal for each preset time period within a preset period;

the determining module is specifically configured to, when determining, for any user, the active frequency distribution data of the user in a preset period based on a terminal usage record of the user:

In an optional embodiment of the second aspect, for any user, the determining module is specifically configured to, when determining, for any user, the activity number distribution data of the user in a preset period based on the terminal usage record of the user:

In an optional embodiment of the second aspect, the determining module is specifically configured to, when performing signal denoising on the initial activity number distribution data of the user to obtain the activity number distribution data of the user:

replacing the active number value with the determined median value;

In an alternative embodiment of the second aspect, the fuzzy relationship comprises a similarity fuzzy relationship matrix;

the determining module is specifically configured to, when determining the fuzzy relationship between the plurality of users based on the activity number distribution data of the plurality of users:

In an optional embodiment of the second aspect, the classification module is specifically configured to, when classifying the plurality of users based on the fuzzy relation to obtain a classification result:

obtaining a truncated matrix of the similarity fuzzy relation matrix;

In an optional embodiment of the second aspect, the classification module is specifically configured to, when acquiring a truncated matrix of the similarity fuzzy relation matrix:

In an alternative embodiment of the second aspect, the classification module is specifically configured to, when determining classification results for the plurality of users based on the truncated matrix:

In an optional embodiment of the second aspect, the sending module is specifically configured to, when sending the recommendation information to the user terminals of the plurality of users based on the classification result:

In a third aspect, there is provided an electronic device including a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the information recommendation method shown in the first aspect of the present application when executing the program.

In a fourth aspect, a computer readable storage medium is provided, on which a computer program is stored, which program, when being executed by a processor, implements the information recommendation method according to the first aspect of the present application.

In a fifth aspect, there is provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium; the processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the information recommendation method shown in the first aspect.

The beneficial effects that this application provided technical scheme brought are:

the fuzzy relation among the plurality of users is determined through the terminal usage records of the plurality of users, the users are classified based on the fuzzy relation, the classification result is obtained, the recommendation information corresponding to the classification result is sent, and for users with subtle differences, the users can still be classified into corresponding categories based on the fuzzy relation, so that the classification accuracy can be improved, and the appropriate information can be recommended to the corresponding users in an appropriate time period.

Further, when abnormal data interference may occur in the sub-period, the active times of the user in each preset time period of each sub-period in the preset period are respectively and correspondingly overlapped to obtain the accumulated active times of the user in each preset time period in the preset period, so that the influence of the abnormal interference data in the sub-period can be reduced, the classification accuracy is improved, and the accuracy of information recommendation is improved.

Further, for the accumulated active times of the user in each preset time period in the preset period, the value of one point in the accumulated active times distribution is replaced by the median of each point in the neighborhood of the point through median filtering, so that the influence of noise data and abnormal values of the behavior of the user can be further removed, and the accuracy of information recommendation is improved.

Further, different thresholds are set for different classification numbers, so that a plurality of users are classified based on the truncated matrix of the similarity fuzzy relation matrix, and the different classification numbers can be applied to recommended scenes of different recommended information.

Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application.

Drawings

The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

fig. 1 is an application environment diagram of an information recommendation method provided in an embodiment of the present application;

fig. 2 is a flow chart of an information recommendation method according to an embodiment of the present application;

fig. 3 is a flow chart of an information recommendation method according to an embodiment of the present application;

FIG. 4 is a schematic diagram of liveness count distribution data provided in one example of the present application;

FIG. 5 is a schematic diagram of a scheme for denoising liveness count distribution data provided in one example of the present application;

FIG. 6 is a schematic diagram of liveness count distribution data provided in one example of the present application;

FIG. 7 is a schematic diagram of determining a similarity fuzzy relation matrix based on liveness count distribution data in one example of the present application;

fig. 8 is a schematic diagram of a scheme for classifying based on liveness distribution data according to an embodiment of the present application;

FIG. 9 is a flow chart of an information recommendation method in examples provided herein;

fig. 10 is a schematic structural diagram of an information recommendation device according to an embodiment of the present application;

fig. 11 is a schematic structural diagram of an electronic device for information recommendation according to an embodiment of the present application.

Detailed Description

Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for the purpose of illustrating the present application and are not to be construed as limiting the present application.

As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. The term "and/or" as used herein includes all or any element and all combination of one or more of the associated listed items.

In the specific embodiments of the present application, any data related to a user, such as a terminal usage record, needs to be licensed or agreed upon by the user when the embodiments of the present application are applied to specific products or technologies, and the collection, usage and processing of the related data needs to comply with relevant laws and regulations and standards of the relevant country and region. That is, in the embodiments of the present application, if any of the above-mentioned user-related data is referred to, the data needs to be obtained through the subject authorization consent, and in compliance with relevant laws and regulations and standards of the country and region.

For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

Machine Learning (ML) is a multi-domain interdisciplinary, involving multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, confidence networks, reinforcement learning, transfer learning, induction learning, teaching learning, and the like.

With research and advancement of artificial intelligence technology, research and application of artificial intelligence technology is being developed in various fields, such as common smart home, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned, automatic driving, unmanned aerial vehicles, robots, smart medical treatment, smart customer service, etc., and it is believed that with the development of technology, artificial intelligence technology will be applied in more fields and with increasing importance value.

Big data (Big data) refers to a data set which cannot be captured, managed and processed by a conventional software tool within a certain time range, and is a massive, high-growth-rate and diversified information asset which needs a new processing mode to have stronger decision-making ability, insight discovery ability and flow optimization ability. With the advent of the cloud age, big data has attracted more and more attention, and special techniques are required for big data to effectively process a large amount of data within a tolerant elapsed time. Technologies applicable to big data include massively parallel processing databases, data mining, distributed file systems, distributed databases, cloud computing platforms, the internet, and scalable storage systems.

The fuzzy relation among a plurality of users can be determined by combining machine learning and big data, and accurate recommendation of information is realized.

First, a brief description will be given of the fuzzy relation correlation technique.

Fuzzy set, which is provided with a domain U and called mapping

Determining fuzzy subsets on UMapping->Called->Membership functions of (a). />Called x pair->Is a membership program of (a). For example: the membership of the 20 years old to "young" was 100%, the membership of the 30 years old to "young" was 80%, and the membership of the 50 years old to "young" was 10%. Fuzzy sets are typically recorded using zade notation:

young setExpressed as:

note that: here "+" does not represent summation nor score. Only in symbolic sense.

Operation of fuzzy set, namely setting the argument U= { x ₁ ,x ₂ ,…x _n There are two fuzzy sets above:

note that: the sign ΣΛabove does not represent a summation, and the previous zade notation "+" is found to be consistent.

Defining intersections of fuzzy sets:

defining a union of fuzzy sets:

where "Λ" means that the two values are at a minimum and "Λ" means that the two values are at a maximum.

Fuzzy relation is a subset of fuzzy relation with argument U, V and U X VIs a fuzzy relationship from U to V. The membership function is a mapping:

And is called asIs (x, y) about fuzzy relation +.>Is a correlation degree of (a). The fuzzy relationship is a generalization of the common relationship, for example, the "father-son" is the common relationship, and the "familiar" relationship between two persons is the fuzzy relationship.

Fuzzy relation matrix: let u1, u2, u3 be three familiarity levels as follows, familiarity level 1 indicating complete familiarity and familiarity level 0 indicating complete unfamiliarity as shown in table 1 below:

TABLE 1

Familiarity with	u1	u2	u3
				u1	1	0.8	0.5
u2	0.8	1	0
				u3	0.5	0	1

We use the fuzzy relation matrix R to represent, namely:

fuzzy equivalence relation: fuzzy relationFor the fuzzy equivalence, if the above 3 points are satisfied:

1) Self-reflexibility

2) Symmetry of

3) Transmissibility of

Fuzzy equivalence matrix: fuzzy equivalence relationIs a fuzzy equivalent matrix.

The user's regular work and rest and activity are valuable features in advertisement recommendation, such as recommending afternoon tea to office workers who are 9 late 5, and recommending night for people who are frequently staying up night. However, the work and rest laws and the active states of the user are very complex, and each classification state is difficult to demarcate. Such as people who stay up frequently: is a calculated stay up several times a week night? Should the night sleep for 5 days stay up and the night sleep for 4 days do not stay? The fuzzy set and fuzzy clustering method in fuzzy data are introduced, and dynamic clustering results under different intercept are output. The result can be used for scenes such as advertisement recommendation. The method skips definition of the user behaviors, directly uses the fuzzy relation on the user behaviors to cluster, and is more accurate and flexible in expression.

The information recommendation method, device, electronic equipment and computer readable storage medium provided by the application aim to solve the technical problems in the prior art.

The following describes the technical solutions of the present application and how the technical solutions of the present application solve the above technical problems in detail with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.

The information recommendation method provided by the application can be applied to an application environment shown in fig. 1. The server 101 acquires terminal usage records of a plurality of users from a plurality of terminals 102, respectively; the server 101 determines a fuzzy relation among the plurality of users based on the terminal usage records of the plurality of users, classifies the plurality of users based on the fuzzy relation to obtain classification results, and then transmits recommendation information to the user terminals 102 of the plurality of users based on the classification results, respectively.

As will be appreciated by those skilled in the art, a "terminal" as used herein may be a cell phone, tablet computer, PDA (Personal Digital Assistant ), MID (Mobile Internet Device, mobile internet device), etc.; the "server" may be implemented as a stand-alone server or as a server cluster composed of a plurality of servers.

In an embodiment of the present application, as shown in fig. 2, a possible implementation manner is provided, and an information recommendation method is provided, which is illustrated by using the method applied to the server in fig. 1 as an example, and may include the following steps:

step S201, terminal use records of a plurality of users are respectively obtained; the terminal usage record includes a usage time of each user for a plurality of applications of the user terminal and a type of the application to be used accordingly.

The terminal usage record may further include an operation of the user terminal by each user, such as clicking, touching, sliding frequency, etc. of each time period.

Specifically, according to the terminal use record of the user, whether the user is in a state of frequently using the terminal or not using the terminal in different time periods can be judged, so that whether the user is at rest, sleeping or working is judged.

Step S202, based on the terminal usage records of the plurality of users, determining fuzzy relations among the plurality of users.

Specifically, a common relationship can only describe that a binary element is either related or unrelated. One of them must be present, and only one of them. However, in addition to absolute relationships or irrelevant relationships, there may be fuzzy concepts such as "some relationships", "close relationships", etc., for example, relationships between body weight and body length may be "relatively normal in development" in addition to absolute normal in development (satisfying the formula) and abnormal, and the fuzzy relationships may be introduced to express such relationships.

Specifically, the fuzzy set may be determined according to the terminal usage record of the user, then the similarity fuzzy relation matrix is obtained according to the fuzzy set, and finally the fuzzy relation among the plurality of users is determined according to the similarity fuzzy relation matrix, and the specific process of determining the fuzzy relation will be described in detail below.

Step S203, classifying the plurality of users based on the fuzzy relation to obtain a classification result.

Specifically, the fuzzy relationship may be described in terms of a similarity fuzzy relationship matrix, a truncated matrix of the similarity fuzzy relationship matrix may be determined, and the classification result may be determined based on the truncated matrix of the similarity fuzzy relationship matrix, and the specific classification process based on the fuzzy relationship will be described in detail below.

Step S204, based on the classification result, recommendation information is sent to the user terminals of the plurality of users.

The classification result may include which users of the plurality of users are in the same class, and what features the users of the class have.

In the implementation process, a plurality of classification results can be set first, and recommendation information corresponding to each type of classification result is set, for example, a certain type of user can be a office worker facing 9 late 5 hours, and afternoon tea can be recommended to the certain type of user; a certain class of users belongs to the crowd who stay up frequently, and the night can be recommended for the class of users.

In the above embodiment, the fuzzy relation among the plurality of users is determined through the terminal usage records of the plurality of users, the users are classified based on the fuzzy relation to obtain the classification result, and the recommendation information corresponding to the classification result is sent.

The process of determining fuzzy relationships between multiple users will be described below in conjunction with the drawings and embodiments.

In an embodiment of the present application, as shown in fig. 3, the determining, based on the terminal usage records of the plurality of users, the fuzzy relationship between the plurality of users in step S202 may include:

step S210, for any user, determining the active frequency distribution data of the user in a preset period based on the terminal use record of the user;

the activity number distribution data may include a cumulative activity number of the user terminal for each preset period of time in a preset period, for example, the cumulative activity number of the user terminal for each hour every day in one month.

The number of activations may be the number of times of various instructions sent by the user through clicking, touching or sliding, which are received by the user terminal, and may be determined, and then the cumulative number of activations in one month, or in one week, is determined based on the number of activations in each hour in each day.

Specifically, for any user in step S210, determining the active frequency distribution data of the user in the preset period based on the terminal usage record of the user may include:

(1) Determining, for any user, the number of activations of the user for each preset time period of each sub-period in a preset period based on a terminal usage record of the user;

(2) Based on the number of activities of the user in each preset time period of each sub-period in the preset period, the accumulated number of activities of the user in each preset time period in the preset period is determined.

Specifically, a user activity timing may be generated. Generating a sequence of 0-24 hours of active state of the user, i.eRepresenting the number of active times of the ith user in the kth hour, as shown in fig. 4, a state sequence of two users (user 1, user 2) in one day can be generated, namely the number of active times of each user in each hour in one day can be obtained, for example, the number of active times of the user1 in 0:00-8:00am is 0; the number of the activations of the user2 in the range of 0:00-5:00am is 0; wherein i and k are positive integers.

Specifically, the activity number distribution of each user in 24 hours in one day may be added together, for example, the cumulative activity number of 0:00-1:00 in one month is obtained by adding the activity number of 0:00-1:00 in one day.

Specifically, clustering using a single day of user activity status is susceptible to noise data and outliers. Because the data superposition technology is adopted to improve the signal-to-noise ratio and the resolution. The specific method comprises the following steps: the data for one period (week, month) of the user is aggregated. The specific period selection is based on the service requirement. If the advertisement scene is sensitive to the behavior change of the user, the advertisement scene is not beneficial to the user for a long time. The polymerization formula is as follows:

wherein: t is a period, such as week, month; c _k ⁽ⁱ⁾ Indicating the aggregation activity times of the kth hour of the ith user on the period T; i and k are positive integers.

In the specific implementation process, when abnormal data interference possibly occurs in the sub-period, the active times of the user in each preset time period of each sub-period in the preset period are respectively and correspondingly overlapped to obtain the accumulated active times of the user in each preset time period in the preset period, so that the influence of the abnormal interference data in the sub-period can be reduced, the classification accuracy is improved, and the accuracy of information recommendation is improved.

In one embodiment, for any user in step S210, determining the active frequency distribution data of the user in the preset period based on the terminal usage record of the user may include:

(1) Determining initial activity frequency distribution data of the user in a preset period according to a terminal use record of the user for any user;

(2) And carrying out signal denoising on the initial activity frequency distribution data of the user to obtain the activity frequency distribution data of the user.

Specifically, in order to further remove the influence of noise data and outliers of the user's behavior, a smoothing denoising step may be added.

The signal denoising method may include: filtering-median filtering based on a spatial domain and denoising-threshold denoising based on a wavelet domain; the median filter is a commonly used nonlinear smoothing filter, and the basic principle is that the value of one point in a signal sequence is replaced by the median of each point in the neighborhood of the point; denoising-threshold denoising based on wavelet domain: the signal is transformed into the frequency domain using wavelet transformation, then the high frequency signal is thresholded, and then the signal is inverse wavelet transformed to recover the signal.

The following will describe the median filtering in detail as an example.

Specifically, signal denoising is performed on the initial activity number distribution data of the user to obtain activity number distribution data of the user, which may include:

a. determining the median value of each active time value in a preset neighborhood of the active time value according to any active time value in the initial active time distribution data of the user;

b. replacing the active number value with the determined median value;

c. and acquiring the activity number distribution data based on each replaced activity number value.

As shown in fig. 5, the median filter window size may be set to 3, and the calculation formula for each signal point is as follows:

wherein c _k ⁽ⁱ⁾ Representing the accumulated active times of the ith user after denoising in the period T in the kth hour; c _j ⁽ⁱ⁾ Representing the cumulative number of active times before denoising in period T for the jth hour of the ith user; j= (k-1, k, k+1); wherein k is a positive integer greater than or equal to 0 and less than or equal to 23.

In the implementation process, for the accumulated active times of a user in each preset time period in a preset period, the value of one point in the accumulated active times distribution is replaced by the median of each point in the neighborhood of the point through median filtering, so that the influence of noise data and abnormal values of the behavior of the user can be further removed, and the accuracy of information recommendation is improved.

Step S220, determining a fuzzy relationship between the plurality of users based on the activity number distribution data of the plurality of users.

Wherein the fuzzy relationship may comprise a similarity fuzzy relationship matrix.

Specifically, determining the fuzzy relationship between the plurality of users based on the activity number distribution data of the plurality of users may include:

(1) Generating an liveness fuzzy set based on the liveness times distribution data of a plurality of user terminals;

(2) A similarity fuzzy relation matrix is determined based on a fuzzy similarity relation between any two predefined activity times distribution data and an activity fuzzy set.

Specifically, the steps of generating the liveness fuzzy set and determining the similarity fuzzy relation matrix may specifically include the following steps:

a. an liveness fuzzy set is generated, and the formula is as follows:

wherein,representing an liveness fuzzy set; c _k ⁽ⁱ⁾ Representing the accumulated active times of the ith user after denoising in the period T in the kth hour; c _j ⁽ⁱ⁾ Representing the cumulative number of active times before denoising in period T for the jth hour of the ith user; j= (k-1, k, k+1); where k is a positive integer greater than 0 and less than 23.

In one example, as shown in FIG. 6, the generated liveness ambiguity set may be as shown in FIG. 6, where h_0 may represent a time period between 0:00-1:00, and so on, smoothing the cumulative number of liveness in each hour, such as using spatial domain based filtering-median filtering or wavelet domain based denoising-threshold denoising, and determining the liveness ambiguity set based on the denoised cumulative number of liveness.

b. The liveness similarity of the users is calculated, and the fuzzy similarity relationship between the users i and j is defined as follows:

wherein,representing user i andfuzzy similarity relationship between users j;

here "||" denotes the modulus of the fuzzy set, as the modulus of the young fuzzy set above is:

as shown in FIG. 7, based on the user1 user's liveness fuzzy set and the user2 liveness fuzzy set, an intersection and a union between the user1 and the user2 user's liveness fuzzy sets may be calculated, and a modulus of the intersection and a modulus of the union may be determined based on the intersection and the union, and finally a fuzzy relationship between the user1 user's liveness fuzzy set and the user2 liveness fuzzy set may be determined.

c. Generating a similarity fuzzy relation matrix

R＝[*] _n×n (14)

Wherein R represents a similarity fuzzy relation matrix, R _i,j Representing elements in the similarity fuzzy relation matrix; i and j are both positive integers.

Examples: assuming that 5 users are provided, calculating similarity for the users in pairs by using a formula (12) to generate a fuzzy similarity relation matrix:

in the above embodiment, as shown in fig. 8, the number of activations of the user in each preset time period of each sub-period in the preset period may be respectively and correspondingly superimposed to obtain the accumulated number of activations of the user in each preset time period in the preset period, then the accumulated number of activations is smoothed and denoised, a similarity fuzzy relation matrix is generated according to the denoised accumulated number of activations, and finally a classification result is obtained based on the similarity fuzzy relation matrix.

The above embodiments illustrate a specific process of generating a similarity-fuzzy relation matrix, and a specific process of classifying based on fuzzy relations will be described below with reference to the drawings and embodiments.

In the embodiment of the present application, a possible implementation manner is provided, and the classifying, based on the fuzzy relationship, of the multiple users in step S203 to obtain a classification result may include:

(1) And obtaining a truncated matrix of the similarity fuzzy relation matrix.

Wherein R is set as E] _n×n ，R＝(r _i,j ) For a pair ofR is recorded _λ ＝(λr _i,j ) Wherein

R is then _λ Called lambda-cut matrix of R.

Specifically, obtaining the truncated matrix of the similarity fuzzy relation matrix may include:

a. determining a range of category numbers corresponding to the classification result;

b. determining a threshold value of the truncated matrix based on the range of the category number;

c. and obtaining a truncated matrix of the similarity fuzzy relation matrix corresponding to the threshold value.

Specifically, the range of the category number is related to the threshold lambda, and the larger lambda is, the finer the classification is, namely the category number is relatively larger; the smaller λ, the coarser the classification, i.e. the relatively smaller the number of categories, i.e. the inverse correlation between the number of categories and the threshold.

(2) A classification result of the plurality of users is determined based on the truncated matrix.

For example, 5 users (x 1, x2, x3, x4, x 5) are classified:

Taking λ=1 to obtainThen U is divided into 5 classes { x1}, { x2}, { x3}, { x4}, { x5}

Taking λ=0.8 to obtainThen U is divided into 4 classes { x1, x3}, { x2}, { x4}, { x5}

Taking λ=0.6 to obtainThen U is divided into 3 classes { x1, x3}, { x2}, { x4, x5}.

Lambda takes on values from large to small, and the fuzzy classification granularity is coarser and coarser. The classification of different granularities can be used in advertisement recommendation scenes, fine granularity scenes are suitable for accurate recommendation, and coarse granularity scenes are suitable for discharge.

In the above embodiment, different thresholds are set for different classification numbers, so that a plurality of users are classified based on the truncated matrix of the similarity fuzzy relation matrix, and the different classification numbers can be applied to recommendation scenes of different recommendation information.

Specifically, determining the classification result of the plurality of users based on the truncated matrix may include:

a. if the truncated matrix is an equivalent matrix, determining a classification result based on the truncated matrix;

b. if the truncated matrix is not the equivalent matrix, the truncated matrix is adjusted until the adjusted truncated matrix is the equivalent matrix, and a classification result is determined based on the adjusted truncated matrix.

If R is _λ Is equivalent, R can be obtained _λ Classification at the lambda level. If R is _λ Not equivalent, R will _λ The modification is equivalent. The transformation method comprises the following steps:

r is R _λ Middle e.gForm of submatrix is changed to +.>Repeating the transformation until the transformation is stable, and recording the new matrix as +.>Is an equivalent matrix.

For example, when classifying 5 users (x 1, x2, x3, x4, x 5), R _0.5 Not an equivalent matrix, it is required to be modified.

Taking λ=0.5 to obtainR _0.5 Rather than an equivalent matrix, it is modified to:

then U is divided into 2 classes { x2}, { x1, x3, x4, x5}

The above-described embodiments illustrate the process of classifying based on the similarity fuzzy relation matrix, and how to transmit the recommendation information according to the classification result will be further described below.

In the embodiment of the present application, a possible implementation manner is provided, and based on the classification result in step S204, sending recommendation information to user terminals of multiple users may include:

(1) Determining at least one sample user corresponding to each category based on the classification result;

(2) Aiming at any sample user, acquiring an information pushing time period of the sample user;

(3) Determining recommendation information corresponding to the information push time period;

(4) And in the information pushing time period, sending recommendation information to the users with the same category as the sample user.

Specifically, a plurality of sample users can be preset, and each sample user sets a corresponding information pushing time period and recommendation information.

For example, one sample user may be set towards the office executive at 9-night 5, and the setting may recommend afternoon tea to this sample user between 3:00-4:30 pm; setting a sample user to be a frequent night stay, the sample user can be recommended for overnight between 10:00 and 12:00.

In one embodiment, a preset sample user may be mixed into a plurality of users to be classified, the sample user and the plurality of users are classified together, and after the classification, the category of the sample user is confirmed, so that the push time period and the recommendation information corresponding to the category can be determined.

In another embodiment, the characteristics of at least one user in each type of user can be determined directly according to the classification results of the plurality of users, that is, the characteristics of all users in the type of the user can be determined, so that the push time period and the recommendation information corresponding to the characteristics of the user are determined.

In order to better understand the above information recommendation method, an example of the information recommendation method of the present invention is described in detail below:

in one example, as shown in fig. 9, the information recommendation method provided in the present application may include the following steps:

step S901, acquiring terminal usage records of a plurality of users;

Step S902, for any user, determining initial activity frequency distribution data of the user in a preset period based on a terminal use record of the user;

step S903, carrying out signal denoising on the initial activity frequency distribution data of the user to obtain the activity frequency distribution data of the user;

step S904, generating an liveness fuzzy set based on the liveness times distribution data of a plurality of user terminals;

step S905, determining a similarity fuzzy relation matrix based on fuzzy similarity relation between any two pieces of activity number distribution data and an activity fuzzy set;

step S906, obtaining a truncated matrix of the similarity fuzzy relation matrix;

step S907, judging whether the truncated matrix is an equivalent matrix; if not, executing step S908; if yes, go to step S909;

step S908, adjusting the truncated matrix;

step S909, determining a classification result based on the truncated matrix;

step S910, based on the classification result, sends recommendation information to the user terminals of the plurality of users.

According to the information recommendation method, the fuzzy relation among the plurality of users is determined through the terminal use records of the plurality of users, the users are classified based on the fuzzy relation to obtain the classification result, recommendation information corresponding to the classification result is sent, the users with fine differences can be still classified into corresponding categories based on the fuzzy relation, the classification accuracy can be improved, and therefore appropriate information can be recommended to the corresponding users in an appropriate time period.

In an embodiment of the present application, as shown in fig. 10, a possible implementation manner is provided, and an information recommendation device 100 is provided, where the information recommendation device 100 may include: an acquisition module 1001, a determination module 1002, a classification module 1003, and a transmission module 1004, wherein,

An acquiring module 1001, configured to acquire terminal usage records of a plurality of users respectively; the terminal use record comprises the use time of each user for a plurality of application programs of the user terminal and the type of the application programs correspondingly used;

a determining module 1002, configured to determine a fuzzy relationship between a plurality of users based on terminal usage records of the plurality of users;

a classification module 1003, configured to classify a plurality of users based on the fuzzy relationship, to obtain a classification result;

and a sending module 1004, configured to send recommendation information to user terminals of a plurality of users based on the classification result.

In this embodiment of the present application, a possible implementation manner is provided, where the determining module 1002 is specifically configured to, when determining a fuzzy relationship between a plurality of users based on terminal usage records of the plurality of users:

The embodiment of the application provides a possible implementation manner, and the active number distribution data comprises accumulated active numbers of the user terminal in each preset time period in a preset period;

The determining module 1002 is specifically configured to, when determining, for any user, the activity number distribution data of the user in a preset period based on the terminal usage record of the user:

In this embodiment of the present application, a possible implementation manner is provided, where the determining module 1002 is configured to, when, for any user, the determining module 1002 determines, for any user, active frequency distribution data of the user in a preset period based on a terminal usage record of the user, specifically:

In this embodiment of the present application, a possible implementation manner is provided, where the determining module 1002 is specifically configured to, when performing signal denoising on initial activity number distribution data of the user to obtain activity number distribution data of the user:

replacing the active number value with the determined median value;

The embodiment of the application provides a possible implementation manner, and the fuzzy relation comprises a similarity fuzzy relation matrix;

the determining module 1002 is specifically configured to, when determining the fuzzy relationship between the plurality of users based on the activity number distribution data of the plurality of users:

In this embodiment of the present application, a possible implementation manner is provided, where the classification module 1003 is specifically configured to, when classifying a plurality of users based on a fuzzy relationship to obtain a classification result:

obtaining a truncated matrix of the similarity fuzzy relation matrix;

In this embodiment of the present application, a possible implementation manner is provided, where the classification module 1003 is specifically configured to:

In this embodiment, a possible implementation manner is provided, where the classification module 1003 is specifically configured to, when determining classification results of a plurality of users based on a truncated matrix:

In this embodiment of the present application, a possible implementation manner is provided, where the sending module 1004 is specifically configured to, when sending recommendation information to user terminals of multiple users based on the classification result:

According to the information recommending device, the fuzzy relation among the plurality of users is determined through the terminal use records of the plurality of users, the users are classified based on the fuzzy relation to obtain the classification result, the recommended information corresponding to the classification result is sent, and for users with fine differences, the users can be still classified into the corresponding categories based on the fuzzy relation, so that the classifying accuracy can be improved, and therefore appropriate information can be recommended to the corresponding users in an appropriate time period.

The information recommending apparatus for a picture according to the embodiments of the present disclosure may execute a method for recommending information for a picture provided by the embodiments of the present disclosure, and the implementation principle is similar, and actions executed by each module in the information recommending apparatus for a picture in each embodiment of the present disclosure correspond to steps in the method for recommending information for a picture in each embodiment of the present disclosure, and detailed functional descriptions of each module in the information recommending apparatus for a picture may be specifically referred to descriptions in the corresponding method for recommending information for a picture shown in the foregoing, which are not repeated herein.

Based on the same principles as the methods shown in the embodiments of the present disclosure, there is also provided in the embodiments of the present disclosure an electronic device that may include, but is not limited to: a processor and a memory; a memory for storing computer operating instructions; and the processor is used for executing the information recommending method shown in the embodiment by calling the computer operation instruction. Compared with the prior art, the information recommendation method can improve the accuracy of classification, so that proper information can be recommended to a corresponding user in a proper time period.

In an alternative embodiment, there is provided an electronic device, as shown in fig. 11, the electronic device 4000 shown in fig. 11 includes: a processor 4001 and a memory 4003. Wherein the processor 4001 is coupled to the memory 4003, such as via a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004. It should be noted that, in practical applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 is not limited to the embodiment of the present application.

The processor 4001 may be a CPU (Central Processing Unit ), general purpose processor, DSP (Digital Signal Processor, data signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic device, transistor logic device, hardware components, or any combination thereof. Which may implement or perform the various exemplary logic blocks, modules, and circuits described in connection with this disclosure. The processor 4001 may also be a combination that implements computing functionality, e.g., comprising one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

Bus 4002 may include a path to transfer information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus or an EISA (Extended Industry Standard Architecture ) bus, or the like. The bus 4002 can be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 11, but not only one bus or one type of bus.

Memory 4003 may be, but is not limited to, ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, RAM (Random Access Memory ) or other type of dynamic storage device that can store information and instructions, EEPROM (Electrically Erasable Programmable Read Only Memory ), CD-ROM (Compact Disc Read Only Memory, compact disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

The memory 4003 is used for storing application program codes for executing the present application, and execution is controlled by the processor 4001. The processor 4001 is configured to execute application program codes stored in the memory 4003 to realize what is shown in the foregoing method embodiment.

Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 11 is merely an example, and should not impose any limitations on the functionality and scope of use of embodiments of the present disclosure.

The present application provides a computer readable storage medium having a computer program stored thereon, which when run on a computer, causes the computer to perform the corresponding method embodiments described above. Compared with the prior art, the information recommendation method can improve the accuracy of classification, so that proper information can be recommended to a corresponding user in a proper time period.

Embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium; the processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the information recommendation method described above. Compared with the prior art, the information recommendation method can improve the accuracy of classification, so that proper information can be recommended to a corresponding user in a proper time period.

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

It should be noted that the computer readable medium described in the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.

The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.

The computer-readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to perform the methods shown in the above-described embodiments.

Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present disclosure may be implemented in software or hardware. The name of a module is not limited to the module itself in some cases, and for example, the classification module may be also described as a "module that classifies based on a similarity fuzzy relation matrix".

The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in this disclosure is not limited to the specific combinations of features described above, but also covers other embodiments which may be formed by any combination of features described above or equivalents thereof without departing from the spirit of the disclosure. Such as those described above, are mutually substituted with the technical features having similar functions disclosed in the present disclosure (but not limited thereto).

Claims

1. An information recommendation method, comprising:

acquiring terminal use records of a plurality of users respectively; the terminal use record comprises the use time of each user for a plurality of application programs of the user terminal and the type of the application programs which are correspondingly used;

determining the active frequency distribution data of the user in a preset period according to the terminal use record of the user for any user; determining fuzzy relations among the plurality of users based on the active times distribution data of the plurality of users; the active frequency distribution data comprise accumulated active frequencies of the user terminal in each preset time period in a preset period;

based on the classification result, sending recommendation information to user terminals of a plurality of users;

wherein the fuzzy relation comprises a similarity fuzzy relation matrix; the determining the fuzzy relation among the plurality of users based on the active times distribution data of the plurality of users comprises the following steps:

generating an liveness fuzzy set based on the following formula based on the liveness frequency distribution data of a plurality of user terminals;

representing an liveness fuzzy set; c _k ⁽ⁱ⁾ Representing the accumulated active times of the ith user after denoising in the period T in the kth hour; c _j ⁽ⁱ⁾ Representing the cumulative number of active times before denoising in period T for the jth hour of the ith user; j= (k-1, k, k+1); wherein k is a positive integer greater than 0 and less than 23;

and determining the similarity fuzzy relation matrix based on the predefined fuzzy similarity relation between any two pieces of activity frequency distribution data and the activity fuzzy set.

2. The information recommendation method according to claim 1, wherein the activity number distribution data includes an accumulated activity number of the user terminal for each preset period of time within the preset period;

The determining, for any user, the active frequency distribution data of the user in a preset period based on the terminal usage record of the user includes:

determining, for any user, the number of activations of the user for each preset time period of each sub-period in the preset period based on a terminal usage record of the user;

and determining the accumulated active times of the user in each preset time period in the preset period based on the active times of the user in each preset time period in each sub-period in the preset period.

3. The information recommendation method according to claim 1, wherein the determining, for any user, the distribution data of the number of activations of the user in a preset period based on the terminal usage record of the user includes:

4. The information recommendation method according to claim 3, wherein the step of performing signal denoising on the initial activity number distribution data of the user to obtain the activity number distribution data of the user includes:

replacing the active number value with the determined median value;

5. The information recommendation method according to claim 1, wherein the classifying the plurality of users based on the fuzzy relation to obtain a classification result comprises:

obtaining a truncated matrix of the similarity fuzzy relation matrix;

the classification results of the plurality of users are determined based on the truncated matrix.

6. The information recommendation method according to claim 5, wherein the obtaining the truncated matrix of the similarity fuzzy relation matrix comprises:

7. The information recommendation method according to claim 6, wherein said determining the classification result of the plurality of users based on the cutoff matrix comprises:

If the truncated matrix is an equivalent matrix, determining the classification result based on the truncated matrix;

and if the truncated matrix is not the equivalent matrix, adjusting the truncated matrix until the adjusted truncated matrix is the equivalent matrix, and determining the classification result based on the adjusted truncated matrix.

8. The information recommendation method according to claim 1, wherein the transmitting recommendation information to the user terminals of the plurality of users based on the classification result comprises:

and in the information pushing time period, sending the recommendation information to the users with the same category as the sample user.

9. An information recommendation device, characterized by comprising:

the acquisition module is used for respectively acquiring terminal use records of a plurality of users; the terminal use record comprises the use time of each user for a plurality of application programs of the user terminal and the type of the application programs which are correspondingly used;

the determining module is used for determining the active frequency distribution data of the user in a preset period according to the terminal use record of the user for any user; determining fuzzy relations among the plurality of users based on the active times distribution data of the plurality of users; the active frequency distribution data comprise accumulated active frequencies of the user terminal in each preset time period in a preset period;

the sending module is used for sending recommendation information to the user terminals of a plurality of users based on the classification result;

wherein the fuzzy relation comprises a similarity fuzzy relation matrix; the determining module is specifically configured to, when determining the fuzzy relationship between the plurality of users based on the activity number distribution data of the plurality of users:

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the information recommendation method according to any of claims 1-8 when executing the program.

11. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the information recommendation method according to any of claims 1-8.