CN108596711B

CN108596711B - Application recommendation method and device and electronic equipment

Info

Publication number: CN108596711B
Application number: CN201810266636.9A
Authority: CN
Inventors: 梁徽科
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2018-03-28
Filing date: 2018-03-28
Publication date: 2021-04-23
Anticipated expiration: 2038-03-28
Also published as: CN108596711A

Abstract

The invention discloses an application recommendation method and device and electronic equipment. The method comprises the following steps: acquiring historical association degree between any two candidate applications in the candidate application set; acquiring historical application behavior data of a target user; for each candidate application, acquiring the recommendation novelty of the candidate application to a target user according to the historical association degree and the historical application behavior data between the candidate application and any other candidate application; and selecting the candidate application with the recommendation novelty meeting the preset recommendation condition as a target application and recommending the target application to a target user. According to the invention, the application with higher popularization degree or popularity can be prevented from being recommended, and the probability of recommending the application with higher novelty is improved. Accurate recommendation of the application is achieved, and user experience is improved.

Description

Application recommendation method and device and electronic equipment

Technical Field

The invention relates to the technical field of internet, in particular to an application recommendation method and device and electronic equipment.

Background

With the rapid development of internet technology and the popularization of mobile intelligence of electronic devices, more and more users are used to download applications (APP for short) to be installed through electronic devices such as mobile phones and tablet devices, so as to obtain services provided by the applications. Therefore, an application platform for aggregating a large number of applications for users to download and install is produced accordingly.

With the explosive growth of the scale of internet users, a large number of applications are often collected in the current application platform, and the application updating or adding speed also grows exponentially, so that in order to avoid the problem that the user cannot easily and quickly select application downloading meeting the requirements in a large number of applications to influence the experience, the application platform generally provides application recommendation service to recommend some similar or related applications which may meet the requirements of the user for the user, and a large number of application search works are omitted for the user, so that the application experience is improved.

Therefore, obtaining the similarity (degree of association) between applications is a key to providing the application recommendation service. The traditional application similarity calculation is usually based on cosine similarity, Euclidean distance, a Jacard algorithm, a Pearson correlation coefficient and other methods, but the methods recommend users only based on the application similarity, applications with high application similarity are easily recommended to the users, but the applications are corresponding to the applications with high heat and high popularity, the phenomenon that the applications recommended to different users converge is finally formed, the applications with high novelty cannot be recommended to different users, the application recommendation accuracy is influenced, and the actual experience of the users is poor.

Disclosure of Invention

It is an object of the present invention to provide a new solution for recommending applications.

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

acquiring historical association between any two candidate applications in the candidate application set;

wherein the set of candidate applications comprises a plurality of candidate applications; the historical relevance is a metric value of a relevant application event occurring in a preset statistical period of the corresponding two candidate applications; the related application events are events of application events which occur in the corresponding two candidate application associations; the application event is an event that the corresponding application is subjected to application behavior by a user;

acquiring historical application behavior data of a target user;

wherein the historical application behavior data is historical data of the target user implementing the application behavior on each candidate application within the statistical period;

for each candidate application, acquiring the recommendation novelty of the candidate application to the target user according to the historical association degree between the candidate application and any other candidate application and the historical application behavior data;

and selecting the candidate application with the recommendation novelty meeting the preset recommendation condition as a target application to recommend to the target user.

Optionally, the step of obtaining the historical association degree includes:

counting the event times of the corresponding associated application events in the counting period aiming at the two corresponding candidate applications;

and calculating the historical association degrees of the two corresponding candidate applications according to the event times.

Alternatively,

the two corresponding candidate applications comprise a first candidate application and a second candidate application;

the associated application events comprise a first associated event, a second associated event, a third associated event and a fourth associated event; the first correlation event is an event that the first candidate application and the second candidate application both generate corresponding application events; the second associated event is an event that the first candidate application has an application event and the second candidate application has not a corresponding application event; the third related event is an event that the first candidate application does not have an application event, but the second candidate application has a corresponding application event; the fourth correlation event is an event that no corresponding application event occurs in the first candidate application and the second candidate application;

the event times comprise a first event time of occurrence of the first correlation event, a second event time of occurrence of the second correlation event, a third event time of occurrence of the third correlation event and a fourth event time of occurrence of the fourth correlation event.

Optionally, the step of calculating the historical association degree according to the number of events includes:

respectively calculating the association uncertainty of the first event, the association uncertainty of the second event and the association uncertainty of the whole event according to the first event frequency, the second event frequency, the third event frequency and the fourth event frequency;

wherein the first event association uncertainty is an uncertainty of occurrence of the associated application event based on a condition that the application event occurs for the first candidate application; the second event association uncertainty is based on the uncertainty of the occurrence of the associated application event under the condition that the second candidate application has the application event; the overall event association uncertainty is the overall uncertainty of the occurrence of the associated application event;

and calculating the historical association degree according to the first event association uncertainty, the second event association uncertainty and the overall event association uncertainty.

Optionally, the step of calculating the first event association uncertainty, the second event association uncertainty, and the overall event association uncertainty includes:

summing an entropy value obtained by calculation according to the first event frequency and the second event frequency with an entropy value obtained by calculation according to the third event frequency and the fourth event frequency to obtain the first event relevancy;

summing an entropy value obtained by calculation according to the first event times and the third event times and an entropy value obtained by calculation according to the second event times and the fourth event times to obtain a second event relevancy;

and calculating an entropy value according to the four of the first event frequency, the second event frequency, the third event frequency and the fourth event frequency to obtain the integral event association uncertainty.

Optionally, the step of calculating the historical relevance includes:

calculating an uncertainty difference according to the overall event association uncertainty, the first event association uncertainty and the second event association uncertainty;

and calculating to obtain the historical association degree according to a preset association factor and the uncertainty difference.

Optionally, the step of calculating the uncertainty difference comprises:

subtracting the first event association uncertainty and the second event association uncertainty from the overall event uncertainty to obtain an uncertainty difference value;

and/or the presence of a gas in the gas,

respectively processing the first event association uncertainty and the second event association uncertainty by using preset optimization factors to obtain the optimized first event association uncertainty and the optimized second event association uncertainty;

and subtracting the optimized first event association uncertainty and the optimized second event association uncertainty from the overall event uncertainty to obtain the uncertainty difference.

Optionally, the step of obtaining the recommended novelty of the candidate application includes:

taking any other candidate application except the candidate application as a comparison application, and acquiring a historical behavior value of the target user for implementing preset target application behaviors on the comparison application from the historical behavior data;

for each comparison application, multiplying the historical association degree of the candidate application and the comparison application, which corresponds to the target application behavior, by the corresponding historical behavior value to obtain a recommended novelty value of the candidate corresponding to the comparison application;

and selecting the maximum value from all the obtained recommended novelty values as the recommended novelty.

Alternatively,

the recommendation condition is that the ranking value of the recommendation novelty after descending ranking is within a preset numerical range;

and/or the presence of a gas in the gas,

the application behavior at least comprises one of clicking an application, downloading the application, installing the application and using the application.

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

the association degree acquiring unit is used for acquiring historical association degrees between any two candidate applications in the candidate application set;

the data acquisition unit is used for acquiring historical application behavior data of a target user;

a novelty obtaining unit, configured to obtain, for each candidate application, a recommended novelty of the candidate application for the target user according to the historical association between the candidate application and any other candidate application and the historical application behavior data;

and the application recommending unit is used for selecting the candidate application with the recommended novelty meeting the preset recommending condition as a target application and recommending the target application to the target user.

According to a third aspect of the present invention, there is provided an electronic apparatus, comprising:

a memory for storing executable instructions;

a processor, configured to execute the electronic device to perform any one of the application recommendation methods according to the first aspect of the present invention, according to the control of the executable instruction.

According to one embodiment of the invention, the recommendation novelty of each candidate application relative to the target user is calculated by acquiring the historical association degree between any two candidate applications in the candidate application set and the historical application behavior data of the target user, and the candidate application with the recommendation novelty conforming to the recommendation condition is selected as the target application and recommended to the target user. The application with higher popularization degree or popularity is prevented from being recommended, and the probability of recommending the application with higher novelty is improved. Accurate recommendation of the application is achieved, and user experience is improved.

Other features of the present invention and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description, serve to explain the principles of the invention.

Fig. 1 is a block diagram showing an example of a hardware configuration of an electronic apparatus 1000 that can be used to implement an embodiment of the present invention.

Fig. 2 shows a flowchart of the application recommendation method of the present embodiment.

Fig. 3 is a flowchart illustrating a step of acquiring a historical association degree between two corresponding candidate applications according to the present embodiment.

Fig. 4 is a flowchart showing the step of calculating the historical association degrees of the corresponding two candidate applications according to the present embodiment.

Fig. 5 is a flowchart illustrating specific steps of calculating the historical association degrees of two corresponding candidate applications according to the present embodiment.

Fig. 6 is a flowchart showing the steps of obtaining the recommended novelty of the candidate application for the target user according to the embodiment.

Fig. 7 shows a block diagram of the application recommendation apparatus 3000 of the present embodiment.

Fig. 8 shows a block diagram of the electronic apparatus 4000 of the present embodiment.

Detailed Description

Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.

In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

< hardware configuration >

Fig. 1 is a block diagram showing a hardware configuration of an electronic apparatus 1000 that can implement an embodiment of the present invention.

The electronic device 1000 may be a laptop, desktop, cell phone, tablet, etc. As shown in fig. 1, the electronic device 1000 may include a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, and the like. The processor 1100 may be a central processing unit CPU, a microprocessor MCU, or the like. The memory 1200 includes, for example, a ROM (read only memory), a RAM (random access memory), a nonvolatile memory such as a hard disk, and the like. The interface device 1300 includes, for example, a USB interface, a headphone interface, and the like. The communication device 1400 is capable of wired or wireless communication, for example, and may specifically include Wifi communication, bluetooth communication, 2G/3G/4G/5G communication, and the like. The display device 1500 is, for example, a liquid crystal display panel, a touch panel, or the like. The input device 1600 may include, for example, a touch screen, a keyboard, a somatosensory input, and the like. A user can input/output voice information through the speaker 1700 and the microphone 1800.

The electronic device shown in fig. 1 is merely illustrative and is in no way meant to limit the invention, its application, or uses. In an embodiment of the present invention, the memory 1200 of the electronic device 1000 is configured to store instructions for controlling the processor 1100 to operate to execute any one of the application recommendation methods provided by the embodiment of the present invention. It will be appreciated by those skilled in the art that although a plurality of means are shown for the electronic device 1000 in fig. 1, the present invention may relate to only some of the means therein, e.g. the electronic device 1000 relates to only the processor 1100 and the storage means 1200. The skilled person can design the instructions according to the disclosed solution. How the instructions control the operation of the processor is well known in the art and will not be described in detail herein.

< example >

< method >

In the embodiment, an application recommendation method is provided for recommending an application to a user.

The application is an application program which provides a corresponding service and can be downloaded, installed and run or directly run through an electronic device such as a mobile phone and a tablet device.

The application recommendation method, as shown in fig. 2, includes: steps S2100-S2400.

In step S2100, a historical association degree between any two candidate applications in the candidate application set is obtained.

The set of candidate applications includes a plurality of candidate applications. The candidate application set is an application set used for selecting a target application recommended to the user, and can be constructed according to an actual application scene or application requirements.

The historical relevance between any two candidate applications is a metric value of the occurrence of relevant application events in a preset statistical period of the corresponding two candidate applications. The historical association degree can be used for measuring the association degree of application events corresponding to any two candidate applications, and based on the historical association degree, the association degree between the applications can be embodied, and the difference of novelty (or popularity index or popularity) between the candidate applications can also be embodied, so that the recommendation novelty of any one candidate application relative to a target user can be obtained based on the historical association degree subsequently.

The statistical period may be set according to an actual application scenario or an application requirement, and may be set to 1 month, for example.

The associated application event is an event that the application event occurs in association with the corresponding two candidate applications.

An application event is an event in which a corresponding application is acted upon by a user. The application behavior is a behavior implemented by the user on the application. In one example, the application behavior includes at least one of clicking on the application, downloading the application, installing the application, and using the application.

In one example, the corresponding two candidate applications include a first candidate application and a second candidate application. The associated application events comprise a first associated event, a second associated event, a third associated event and a fourth associated event. The first correlation event is an event that the first candidate application and the second candidate application both generate the same application event; the second associated event is an event that the application event occurs in the first candidate application and the application event does not occur in the second candidate application; the third associated event is an event that the first candidate application does not have an application event, but the second candidate application has the application event; the fourth associated event is an event in which the same application event has not occurred for both the first candidate application and the second candidate application.

For example, the first candidate application is application a, the second candidate application is application B, and the application event is an event that the corresponding application is clicked by the user. The first associated event is an event where both application a and application B are clicked on by the user. The second associated event is an event that the application a is clicked by the user and the application B is not clicked by the user. The third associated event is an event that the application a is not clicked by the user and the application B is clicked by the user. The fourth associated event is an event in which neither application a nor application B has been clicked on by the user.

In this embodiment, by obtaining the historical association degree between any two applications in the candidate application set, which is used for measuring the occurrence of the associated application event between the applications, the recommendation novelty of any one candidate application with respect to the target user can be obtained based on the historical association degree in combination with the subsequent steps, and the application is recommended for the target user based on the recommendation novelty degree, so that the applications with higher popularity or popularity are prevented from being recommended, the probability of recommending the applications with higher novelty to the user is improved, the accurate recommendation of the applications is realized, and the user experience is improved.

In one example, the step S2100 of obtaining the historical association degree between two corresponding candidate applications may be as shown in fig. 3, and includes steps S2110 to S2120.

Step S2110, for the two corresponding candidate applications, counting the event times of the corresponding associated application event occurring within the counting period.

In this embodiment, the number of events occurring in the associated application event may be statistically obtained by the application platform providing the candidate application according to history data of clicking, installing, downloading, using, and the like of each application recorded in the statistical period.

In one example, the two corresponding candidate applications include a first candidate application and a second candidate application; the associated application events comprise a first associated event, a second associated event, a third associated event and a fourth associated event.

The first correlation event is an event in which the first candidate application and the second candidate application both have a corresponding application event. The second associated event is an event in which the first candidate application has an application event and the second candidate application has not a corresponding application event. The third associated event is an event in which the first candidate application does not have an application event, but the second candidate application has a corresponding application event. The fourth correlation event is an event in which no corresponding application event occurs in both the first candidate application and the second candidate application.

Correspondingly, the event times include a first event time of occurrence of a first associated event, a second event time of occurrence of a second associated event, a third event time of occurrence of a third associated event, and a fourth event time of occurrence of a fourth associated event.

For example, the first candidate application is application a and the second candidate application is application B, assuming that the application event EventA occurred by application a corresponds to the event of application B occurring EventB. The first associated event is an event which occurs in both EventA and EventB, and the corresponding first event frequency is frequency K11 which occurs in both EventA and EventB; the second associated event is an event that EventA occurs but EventB does not occur, and the corresponding second event frequency is the frequency K12 of EventA occurrence but EventB non-occurrence; the third associated event is the event that EventA does not occur but EventB occurs, and the corresponding third event frequency is the frequency K21 that EventA does not occur but EventB occurs; the fourth associated event is an event in which both EventA and EventB have not occurred, and the corresponding fourth event count is the count K22 in which both EventA and EventB have not occurred.

Step S2120, according to the event times, calculating historical association degrees of the two corresponding candidate applications.

In one example, step S2120 may be as shown in FIG. 4, including steps S2121-S2122.

Step S2121, respectively calculating the first event association uncertainty, the second event association uncertainty and the overall event association uncertainty according to the first event frequency, the second event frequency, the third event frequency and the fourth event frequency.

The first event association uncertainty is an uncertainty of occurrence of a corresponding associated application event based on the first candidate application for the occurrence of the application event.

The step of calculating the first event association uncertainty may comprise:

and summing the entropy values calculated according to the first event times and the second event times and the entropy values calculated according to the third event times and the fourth event times to obtain the first event relevancy.

In this example, the first event count K11, the second event count K12, the third event count K21, and the fourth event count K22 are assumed as an example in which the application a and the application B are corresponding two candidate applications.

Correspondingly, the Entropy value Encopy (K11, K12) calculated from the first event number and the second event number is:

Entropy(K 11,K 12)＝lg(K 11+K 12)-lg(K 11)-lg(K 12)；

the Entropy value Encopy (K21, K22) calculated according to the third event times and the fourth event times is as follows:

Entropy(K21,K22)＝lg(K21+K22)-lg(K21)-lg(K22)；

the calculated first event relevancy rowEncopy is:

rowEntropy＝Entropy(K 11,K 12)+Entropy(K21,K22)。

and the second event association uncertainty is the uncertainty of the occurrence of the associated application event based on the condition that the application event occurs in the second candidate application.

The step of calculating the second event correlation uncertainty may comprise:

and summing the entropy values obtained by calculation according to the first event times and the third event times and the entropy values obtained by calculation according to the second event times and the fourth event times to obtain a second event relevancy.

Continuing to obtain the result based on the first event frequency K11, the second event frequency K12,

The third event frequency K21 and the fourth event frequency K22 are examples.

Correspondingly, the Entropy value Encopy (K11, K21) calculated from the first event number and the third event number is:

Entropy(K 11,K21)＝lg(K 11+K21)-lg(K 11)-lg(K21)；

the Entropy value Encopy (K12, K22) calculated according to the second event times and the fourth event times is as follows:

Entropy(K 12,K22)＝lg(K 12+K22)-lg(K 12)-lg(K22)；

the calculated second event relevancy columnentry is:

columnEntropy＝Entropy(K 11,K21)+Entropy(K 12,K22)。

the overall event association uncertainty is the overall uncertainty of the occurrence of the corresponding associated application event.

The step of calculating the associated uncertainty of the sort event comprises:

and calculating an entropy value according to the four times of the first event, the second event, the third event and the fourth event, and obtaining the uncertainty of the association of the whole event.

The above mentioned number of events including the first number of events K11, the second number of events K12, the third number of events K21, and the fourth number of events K22 is taken as an example.

Correspondingly, the overall event association uncertainty matrixencopy is:

matrixEntropy＝Entropy(K 11,K 12,K21,K22)

＝lg(K 11+K 12+K21+K22)-lg(K 11)-lg(K 12)-lg(K21)-lg(K22)。

step S2122, calculating historical association degree according to the first event association uncertainty, the second event association uncertainty and the overall event association uncertainty.

The first event uncertainty is the uncertainty of the occurrence of the corresponding associated application event on the condition that the application event occurs on the basis of the first candidate application, and can embody the association degree of the first candidate application and the second candidate application; the second event association uncertainty is the uncertainty of the occurrence of the corresponding associated application event based on the application event of the second candidate application, and can embody the novelty of the second candidate application relative to the first candidate application; the overall event association uncertainty is the overall uncertainty of the occurrence of the corresponding associated application event, and can reflect the overall association of the two candidate applications. Based on the uncertainty of the three, the historical association degree is calculated, so that the calculated historical association degree can not only reflect the association degree between the two corresponding candidate applications, but also reflect the novelty difference between the two corresponding candidate applications.

In one example, the step of calculating the historical association degree may be as shown in fig. 5, including: steps S21221-S21222.

Step S21221, calculating an uncertainty difference according to the overall event association uncertainty, the first event association uncertainty and the second event association uncertainty.

In one example, the step of calculating the uncertainty difference comprises:

and subtracting the first event association uncertainty and the second event association uncertainty from the overall event uncertainty to obtain an uncertainty difference value.

In the above example, the first event relevance rowEntrol, the second event relevance columnEntrol, and the overall event relevance uncertainty matrixEntrol are taken as examples.

The uncertainty difference Delta is:

Delta＝matrixEntropy-rowEntropy-columnEntropy。

the first event uncertainty is the uncertainty of the occurrence of the corresponding associated application event on the condition that the application event occurs on the basis of the first candidate application, and can embody the association degree of the first candidate application and the second candidate application; the second event association uncertainty is the uncertainty of the occurrence of the corresponding associated application event based on the application event of the second candidate application, and can embody the novelty of the second candidate application relative to the first candidate application; the overall event association uncertainty is the overall uncertainty of the occurrence of the corresponding associated application event, and can reflect the overall association of the two candidate applications.

And subtracting the uncertainty of the first event and the uncertainty of the association of the second event from the overall uncertainty, and evaluating whether the recommended uncertainty of the second candidate application is greater than the average value of the recommended applications under any condition without setting through an uncertainty difference.

And step S21222, calculating to obtain historical association degree according to a preset association factor and an uncertainty difference value.

The association factor apha may be preset according to a specific application scenario or application requirements, and may be set to 2, for example.

In the above example, taking Delta as an example, the historical association degree hilr calculated according to the association factor apha and the uncertainty difference Delta:

HILLR＝apha×Delta

＝apha×(matrixEntropy-rowEntropy-columnEntropy)

in practical application, when application recommendation is performed based on the historical association degree calculated by the uncertainty difference Delta, the probability that a second candidate application is recommended through a first candidate application is the same as the probability that a first candidate application is recommended through a second candidate application, applications with high popularity can be recommended through some applications with high novelty, the probability of recommending applications with high novelty is reduced, and user experience is affected.

Thus, in one example, the step of calculating the uncertainty difference may comprise:

respectively processing the first event association uncertainty and the second event association uncertainty by using preset optimization factors to obtain the optimized first event association uncertainty and second event association uncertainty;

and subtracting the optimized first event association uncertainty and second event association uncertainty from the overall event uncertainty to obtain an uncertainty difference value.

The optimization factor Beta may be preset according to a specific application scenario or an application requirement, for example, may be set to a value greater than 0 and smaller than the association factor apha.

Taking the first event association degree rowEntrol, the second event association degree columnEntrol, and the overall event association uncertainty matrixEntrol in the above example as examples, the first event association uncertainty is processed by using an optimization factor Beta, and the optimized first event association uncertainty is (apha-Beta) xrowEntrol; and processing the second event association uncertainty by using the optimization factor Beta to obtain the optimized second event association uncertainty which is Beta multiplied by columnEncopy.

Correspondingly, the uncertainty difference Delta is:

Delta＝matrixEntropy-(apha-beta)×rowEntropy-Beta×columnEntropy

and (3) calculating the historical relevance HILLR according to the relevance factor apha and the uncertain difference Delta:

HILLR＝apha×Delta

＝apha×(matrixEntropy-(apha-beta)×rowEntropy-Beta×columnEntropy)

the first event association uncertainty and the second event association uncertainty are processed through the optimization factor Beta, so that the second event association uncertainty reflecting the novelty of the second candidate application relative to the first candidate application occupies more weight when the historical association is calculated, the subsequent recommendation novelty based on the historical association is more accurate, application recommendation is performed based on the recommendation novelty, and the application with high novelty is more easily recommended.

Step S2200 is to acquire historical application behavior data of the target user.

The historical application behavior data is the historical data of the application behavior executed by the target user for each candidate application in the statistical period. In one example, the application behavior includes at least one of clicking on an application, downloading an application, installing an application, and using an application.

The historical application behavior data of the target user can be obtained through an application platform providing candidate applications according to the recorded user log statistics, and is not limited herein.

Step S2300, for each candidate application, obtaining a recommendation novelty of the candidate application to the target user according to the historical association degree between the candidate application and any other candidate application and the historical application behavior data.

In one example, step S2300 may be as shown in FIG. 6, including steps S2310-S2330.

Step S2310, taking any one of the other candidate applications except the candidate application as a comparison application, and obtaining a historical behavior value of the target user performing a preset target application behavior on the comparison application from the historical behavior data.

Assuming that the candidate application set I ═ { item (N) } (N ═ 1, … …, N) in this embodiment includes N candidate applications, for the candidate applications item (I), the candidate applications item (j) (j ≠ I) are selected as comparison applications. Assuming a target application behavior k, identifying whether a target user u has a k behavior on a candidate application item (j) by a corresponding historical behavior value Y (j, k), wherein if the k behavior occurs, the value of Y (j, k) is 1, otherwise, the value is 0.

In the above method, the historical behavior value Y (j, k) of each candidate application item (j) (j ≠ i) as a comparison application can be calculated.

Step S2320, aiming at each comparison application, the historical association degree of the candidate application and the comparison application, which corresponds to the target application behavior, is multiplied by the corresponding historical behavior value, so as to obtain the recommended novelty value of the candidate correspondence relative to the comparison application.

Based on the candidate application as item (i), compare application item (j) (j ≠ i) as an example. HiLLR corresponding to target application behavior k of candidate application and comparison application_k(i, j) may be calculated based on the method of calculating historical relevance described above with reference to FIG. 3.

Wherein, the comparison application item (j) (j ≠ i) is used as the first candidate application, and the item (i) is used as the second candidate application. The associated application events comprise a first associated event, a second associated event, a third associated event and a fourth associated event. The first associated event is an event where both item (j) and item (i) occur corresponding target application behavior k. The second associated event is that item (j) occurred with application line k, and item (i) did not occur with the corresponding application lineAn event of k. The third associated event is an event for which item (j) does not have application behavior k, but item (i) has a corresponding application behavior k. The fourth associated event is an event that neither item (j) nor item (i) has the corresponding application behavior k. Correspondingly, the corresponding first, second, third and fourth correlation event times K11, K12, K21 and K22 can be counted. Further, the above HILLR can be calculated based on the above method_k(i, j), which will not be described in detail herein.

Further, a recommended novelty value RN for item (i) versus item (j) may be obtained_k(i,j)：

RN_k(i,j)＝Y(j,k)×HILLR_k(i,j)。

Based on the above example, candidate applications item (i) can be calculated for each item (j) (j ≠ i) as comparison application (i) based on target application behavior k, the corresponding recommended novelty value RN_k(i,j)。

In step S2330, the maximum value is selected from all the obtained recommended novelty values as the recommended novelty.

In this example, the recommended novelty of the candidate application with respect to the target user u, H (u, i):

H(u,i)＝max(RN_k(i,j))(j≠i,j∈I,k∈K)；

wherein, I is a candidate application set comprising a plurality of candidate applications, and item (j) (j ≠ I) traverses any one candidate application in the candidate application set; and K is an application behavior set and comprises a plurality of application behaviors, and the behavior K traverses any one application behavior in the application behavior set, such as clicking an application, downloading the application, installing the application, using the application and the like.

And step S2400, selecting candidate applications with the recommendation novelty meeting the preset recommendation conditions as target applications, and recommending the target applications to the target user.

The recommendation condition may be preset according to a specific application scenario or application requirements.

In one example, the recommendation condition may be that the sorted values of the recommended novelty sorted in descending order are within a preset numerical range. Correspondingly, the candidate applications are sorted in a descending order according to the recommended novelty of each candidate application, and the candidate applications with the sorted order in a preset numerical range are selected as the target applications.

For example, if the value range is 1-3, the candidate applications ranked in the top 3 bits when the recommended novelty is ranked from high to low are selected as the target applications.

< apparatus >

In the present embodiment, there is also provided an application recommendation apparatus 3000, as shown in fig. 7, including: the association degree obtaining unit 3100, the data obtaining unit 3200, the novelty obtaining unit 3300, and the application recommending unit 3400 are configured to implement any one of the application recommending methods provided in this embodiment, and are not described herein again.

An association degree obtaining unit 3100, configured to obtain a historical association degree between any two candidate applications in the candidate application set;

a data obtaining unit 3200, configured to obtain historical application behavior data of a target user;

a novelty obtaining unit 3300, configured to, for each candidate application, obtain a recommended novelty of the candidate application for the target user according to the historical association between the candidate application and any other candidate application and the historical application behavior data;

and the application recommending unit 3400 is configured to select the candidate application with the recommended novelty meeting the preset recommendation condition as a target application and recommend the target application to a target user.

Optionally, the association degree obtaining unit 3100 is configured to:

Alternatively,

Optionally, the association degree obtaining unit 3100 is configured to:

Optionally, calculating the historical association degree according to the first event association uncertainty, the second event association uncertainty, and the overall event association uncertainty includes:

Optionally, the calculating the historical relevance comprises:

Optionally, calculating the uncertainty difference comprises:

and/or the presence of a gas in the gas,

Novelty retrieval unit 3300 is also used to:

Alternatively,

and/or the presence of a gas in the gas,

In this embodiment, the application recommendation device 3000 may be application platform software providing an application recommendation service, and after running, the application recommendation method provided in this embodiment is implemented to recommend an application to a user.

It will be appreciated by those skilled in the art that the application recommendation device 3000 can be implemented in various ways. For example, the application recommendation device 3000 may be implemented by an instruction configuration processor. For example, the application recommendation apparatus 3000 may be implemented by storing instructions in ROM and reading the instructions from ROM into a programmable device when the device is started. For example, the application recommendation device 3000 may be cured into a dedicated device (e.g., ASIC). The application recommendation device 3000 may be divided into units independent of each other, or may be implemented by combining them together. The application recommendation device 3000 may be implemented by one of the various implementations described above, or may be implemented by a combination of two or more of the various implementations described above.

< electronic apparatus >

In this embodiment, an electronic apparatus 4000 is further provided, as shown in fig. 8, including:

a memory 4100 for storing executable instructions;

a processor 4200, configured to run the electronic device to perform any one of the application recommendation methods provided in this embodiment according to the control of the executable instructions.

In this embodiment, the electronic device 4000 may be a mobile phone, a tablet computer, a notebook computer, a palm computer, a notebook computer, or the like. The electronic device 4000 may also include other hardware modules, for example, in one example, the electronic device 4000 may be the electronic device 1000 shown in fig. 1.

The embodiments of the present invention have been described above with reference to the drawings, and according to the embodiments, an application recommendation method, an application recommendation apparatus, and an electronic device are provided, where a recommendation novelty of each candidate application with respect to a target user is calculated by obtaining a historical association degree between any two candidate applications in a candidate application set and historical application behavior data of the target user, and the candidate application whose recommendation novelty meets a recommendation condition is selected as the target application and recommended to the target user. The application with higher popularization degree or popularity is prevented from being recommended, and the probability of recommending the application with higher novelty is improved. Accurate recommendation of the application is achieved, and user experience is improved.

It is well known to those skilled in the art that with the development of electronic information technology such as large scale integrated circuit technology and the trend of software hardware, it has been difficult to clearly divide the software and hardware boundaries of a computer system. As any of the operations may be implemented in software or hardware. Execution of any of the instructions may be performed by hardware, as well as by software. Whether a hardware implementation or a software implementation is employed for a certain machine function depends on non-technical factors such as price, speed, reliability, storage capacity, change period, and the like. Accordingly, it will be apparent to those skilled in the art of electronic information technology that a more direct and clear description of one embodiment is provided by describing the various operations within the embodiment. Knowing the operations to be performed, the skilled person can directly design the desired product based on considerations of said non-technical factors.

The present invention may be a system, method and/or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therewith for causing a processor to implement various aspects of the present invention.

The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: 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), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.

The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.

The computer program instructions for carrying out operations of the present invention may be assembler instructions, Instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions 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 type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing an electronic circuit, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA), with state information of computer-readable program instructions, which can execute the computer-readable program instructions.

Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.

These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowchart 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 invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). 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. It is well known to those skilled in the art that implementation by hardware, by software, and by a combination of software and hardware are equivalent.

Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the invention is defined by the appended claims.

Claims

1. An application recommendation method, comprising:

acquiring historical application behavior data of a target user;

selecting the candidate application with the recommendation novelty meeting the preset recommendation condition as a target application to be recommended to the target user,

wherein the step of obtaining the recommended novelty of the candidate application comprises:

and obtaining the recommended novelty according to all the obtained recommended novelty values.

2. The method of claim 1, wherein the step of obtaining the historical association comprises:

3. The method of claim 2, wherein,

4. The method of claim 3, wherein the step of calculating the historical relevance based on the number of events comprises:

5. The method of claim 4, wherein the step of calculating a first event correlation uncertainty, a second event correlation uncertainty, and an overall event correlation uncertainty comprises:

6. The method of claim 4, wherein the step of calculating the historical association comprises:

7. The method of claim 6, wherein the step of calculating an uncertainty difference comprises:

and/or the presence of a gas in the gas,

8. The method of claim 1, wherein the step of obtaining the recommended novelty of the candidate application further comprises:

9. The method of claim 1, wherein,

and/or the presence of a gas in the gas,

10. An application recommendation apparatus, comprising:

an application recommending unit, configured to select the candidate application with the recommended novelty meeting a preset recommending condition as a target application, and recommend the target application to the target user,

wherein, the novelty obtaining unit is further configured to:

11. An electronic device, comprising:

a memory for storing executable instructions;

a processor, configured to execute the electronic device to perform any one of the application recommendation methods of claims 1-9 according to the control of the executable instructions.