Application recommendation method, medium and electronic equipment

Abstract

The invention relates to the technical field of intelligent terminals, and discloses an application recommendation method, a medium and electronic equipment, which can accurately recommend a preference APP for a user within a period of time after entering fences of recommendation factors such as a subway scene and a home scene, for example, within the first 5 minutes, and improve the accuracy and personalization of application recommendation. The method is applied to the electronic equipment, and a first application and a second application are installed on the electronic equipment. The method comprises the following steps: determining that the electronic equipment enters a first fence corresponding to a first recommendation factor; within a first time period after the electronic equipment enters the first fence, displaying a first application corresponding to the first fence and recommended for the user; and after a first time period after the electronic equipment enters the first fence, displaying a second application corresponding to the first fence and recommended to the user.

Description

Technical Field

[0001] The present application relates to the field of smart terminal technology, and in particular to an application recommendation method, medium and electronic device. Background Art

[0002] As more and more applications (APPs) are installed in mobile phones, most mobile phones will provide APP suggestions to users to facilitate users to operate recommended APPs. However, most current APP suggestion schemes use the recent popular APPs and recent high-frequency APPs to combine and score to recommend APPs to users. Specifically, the mobile phone displays APP suggestion cards, such as YOYO suggestion TM The mobile phone recommends apps to users through cards. Among them, the recent popular apps are usually obtained based on the data statistics of multiple users, and the recent high-frequency apps can be all the apps with high usage frequency in the user's recent historical period regardless of the scene. Specifically, the mobile phone displays app suggestion cards, such as YOYO recommendation TM Cards to recommend apps to users. Figure 1 The following are YOYO suggestions displayed on the phone: TM An example of a card that YOYO recommends TM The card contains TikTok TM , Taobao TM and Alipay TM Icons of Tik Tok TM For the latest popular APP and Taobao TM and Alipay TM It is the most recent high-frequency APP. At this time, the user can click YOYO suggestion TM However, the current APP recommendation schemes lack personalization in recommending APPs, which results in the APPs recommended by the mobile phone not being the APPs that the user actually needs to use. Summary of the invention

[0003] The embodiments of the present application provide an application recommendation method, medium, and electronic device, which can accurately recommend preferred APPs to users within a period of time after entering the fence of recommendation factors such as subway scenes and home scenes, for example, within the first 5 minutes, thereby improving the accuracy and personalization of application recommendations.

[0004] In the first aspect, the embodiment of the present application provides an application recommendation method, which is applied to an electronic device, and a first application and a second application are installed on the electronic device; and the method includes: determining that the electronic device enters a first fence corresponding to a first recommendation factor; displaying the first application corresponding to the first fence recommended to the user within a first time period after the electronic device enters the first fence; and displaying the second application corresponding to the first fence recommended to the user after the first time period after the electronic device enters the first fence. For example, when the first recommendation factor is a home scene, the electronic device can recommend access card APPs and smart home APPs (i.e., first applications) to the user within a period of time after entering the fence, and recommend entertainment APPs preferred to be used at home to the user after a period of time after entering the fence. That is, the APPs that the user needs to use are usually different within a period of time before entering the fence of a certain recommendation factor and after that period of time. In this way, the present application can recommend different preferred APPs to the user in different time periods when the user enters the fences of different recommendation factors, thereby accurately recommending the user's preferred APPs in different time periods, facilitating the user to quickly use the required APPs, and greatly improving the user experience.

[0005] In a possible implementation, a third application is also installed in the electronic device; the preference probability of the first application is greater than the preference probability of the third application that is not recommended to the user, wherein the higher the preference probability, the higher the probability that the application has been used within the first historical time period within the second time period after entering the first fence, wherein the first time period is less than or equal to the second time period. For example, the duration of the second time period may be the first duration below, such as the second time period being within the first 5 minutes after entering the first fence. Correspondingly, the first time period may be the first period after entering the first fence, such as within the first 4 minutes. In this way, the first application is the APP that the user prefers to use within the first 5 minutes after entering the first fence.

[0006] In a possible implementation, the preference probability is determined based on the use probability of the application in the second time period after entering the first fence in the first historical time period and the promotion degree of the application in the third time period after entering the first fence in the first historical time period, wherein the higher the promotion degree, the higher the probability that the application has been used in the third time period after entering the first fence in the first historical time period, the second time period is less than the third time period, and the promotion degree is determined based on the use probability of the application in the third time period after entering the first fence in the first historical time period and the use probability of the application in the third time period after entering the fences corresponding to multiple recommendation factors in the first historical time period, and the multiple recommendation factors include the first recommendation factor. It can be understood that the third time period can be the entire time period after entering the first fence, that is, the entire time period under the current first recommendation factor, such as the entire time period from the user entering the station to exiting the station in the subway scenario. Among them, the application with a larger promotion degree indicates that the user has a high probability of using the application under the current recommendation factor. In this way, the present application comprehensively considers the use probability of the application in the second time period after entering the first fence, and the use probability of the application in the entire third time period after entering the first fence, and can accurately recommend the user's preferred APP when the user enters the fences of different recommendation factors, so as to facilitate the user to quickly use the required APP, which greatly improves the user experience. In addition, the first recommendation factor may be any one of multiple recommendation factors.

[0007] In some possible implementations, the usage probability of the application in the second time period after entering the first fence in the first historical time period is calculated by the following method: Calculate the topK prop = count(s|app in topN) / M, where topK prop represents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each sequence includes the application used in the second time period after entering the first fence corresponding to the first recommendation factor, count(s|app intopN) represents the number of sequences containing applications in the multiple sequences corresponding to the first recommendation factor, topK prop Indicates the usage probability of the application in the second time period after entering the first fence in the first historical time period. Wherein, s in the above formula here represents the first recommendation factor, and the formula can calculate the usage probability of different applications under different recommendation factors, that is, the parameter s can take different values ​​to represent different recommendation factors, and the parameter app can also take different values ​​to represent different factors.

[0008] In some possible implementations, the preference probability is calculated by the following formula: prop =exp(lift-a)*topK prop , among which, hit proprepresents the preference probability applied in the second time period after entering the first fence in the first historical time period, exp() represents the exponential operation, topK prop represents the usage probability of the application in the second time period after entering the first fence in the first historical time period, lift represents the lift degree of the application in the third time period after entering the first fence in the first historical time period, and a is a constant related to the lift degree.

[0009] In a possible implementation, a third application is also installed in the electronic device; the preference probability of the first application is greater than the preference probability of the third application that is not recommended to the user, and the higher the preference probability, the higher the probability of the application appearing in the first N applications used after entering the first fence in the first historical time period, where N is a positive integer. That is, the period of time after the electronic device enters the first fence is the time for using the first N applications. In this way, the first application is the APP that the user prefers to use during the process of using the first N applications after entering the first fence.

[0010] In a possible implementation, the preference probability is determined based on the probability of the application appearing in the first N applications used after entering the first fence in the first historical time period and the promotion degree of the application in the third time period after entering the first fence in the first time period, wherein the higher the promotion degree, the higher the probability that the application has been used in the third time period after entering the first fence in the first historical time period, the first time period is less than the third time period, and the promotion degree is determined based on the use probability of the application in the third time period after entering the first fence in the first historical time period and the use probability of the application in the third time period after entering the fences corresponding to multiple recommendation factors in the first historical time period, and the multiple recommendation factors include the first recommendation factor. It can be understood that the third time period can be the entire time period after entering the first fence, that is, the entire time period under the current first recommendation factor, such as the entire time period from entering the station to exiting the station in the subway scenario. In this way, the present application comprehensively considers the use probability of the application in the process of using the first N applications after entering the first fence, and the use probability of the application in the entire third time period after entering the first fence, and can accurately recommend the user's preferred APP when the user enters the fences of different recommendation factors, so as to facilitate the user to quickly use the required APP, which greatly improves the user experience. In addition, the above-mentioned first recommendation factor can be any one of the multiple recommendation factors.

[0011] In a possible implementation, the probability of an application appearing in the top N applications used after entering the first fence in the first historical time period is calculated as follows: Calculate the topK prop = count(s|app in topN) / M, where topK proprepresents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each sequence includes the top N applications used after entering the first fence corresponding to the first recommendation factor, count(s|app in topN) represents the number of sequences containing applications in the multiple sequences corresponding to the first recommendation factor, topK prop It represents the probability of occurrence of the application in the first N applications used after entering the first fence in the first historical time period. In the above formula, s here represents the first recommendation factor. The formula can calculate the usage probability of different applications under different recommendation factors, that is, the parameter s can take different values ​​to represent different recommendation factors, and the parameter app can also take different values ​​to represent different factors.

[0012] In one possible implementation, the preference probability is calculated as follows: prop =exp(lift-a)*topK prop (app|s), where hit prop represents the preference probability of the application among the first N applications used after entering the first fence in the first historical time period, exp() represents the exponential operation, topK prop (app|s) represents the probability of an application appearing in the top N applications used after entering the first fence in the first historical time period, lift represents the lift of the application in the third time period after entering the first fence in the first historical time period, and a is a constant related to the lift.

[0013] In one possible implementation, the lift value is a value between 0 and 1, and a is 0.5. For example, lift is a value between 0 and 1. Then, exp(lift-a) is used to calculate the topK prop Make corrections. If lift is greater than 0.5, increase topK prop Get hit prop On the contrary, if lift is less than 0.5, reduce topK prop Get hit prop , lift = 0.5, no need to adjust topK prop .

[0014] In one possible implementation, after the electronic device enters the first fence for the first time period, the second application corresponding to the first fence is displayed as recommended to the user, including: detecting that the recommended first application is used by the user, ending the first time period, and displaying the second application. At this time, even if the first time period is less than the second time period, or the number of APPs used in the first time period does not reach N, if the currently recommended first application is clicked and used, the recommendation of the first application can be ended, that is, the first time period is ended. In this way, it is convenient to accurately recommend applications to users in actual scenarios.

[0015] In a possible implementation, the first confidence of the first application is greater than the first confidence of the third application, and the first confidence (i.e., the trigger confidence hereinafter) is determined based on the preference probability of the application and the total number of applications in the electronic device. It can be understood that an appropriate number of applications are recommended to the user based on the total number of applications, for example, the greater the total number of applications, the greater the number of preferred applications determined at a fence triggering moment.

[0016] In a possible implementation, the first confidence is: the ratio of the first ranking of the application to the total number of applications in the electronic device, wherein the first ranking of the application is the ranking of the preference probability of the application in the ascending order of the preference probability of each application in the electronic device. In this way, it is convenient to compare the sizes of each first confidence, and convenient to set and output several APPs with trigger confidence greater than the trigger confidence threshold as the user's preferred APP under one recommendation factor.

[0017] In a possible implementation, the preference probability of the first application is greater than the preference probability threshold, or the preference probability of the first application is greater than the preference probability threshold and the first confidence of the first application is greater than the first confidence threshold. At this time, the number of the first application can be one or more.

[0018] In a possible implementation, a fourth application is also installed on the electronic device, wherein the promotion degree of the second application is greater than the promotion degree of the fourth application that is not recommended to the user. It is understood that the second application may be an APP that the user prefers to use under the first recommendation factor 5 minutes after entering the first fence, or the second application may be an APP that the user prefers to use under the first recommendation factor after entering the first fence and using the first 5 applications.

[0019] In a possible implementation, the lift is calculated in the following manner: p=Count(app|s) / Count(s), where p is the probability of using the application in the third time period after entering the first fence in the first historical time period, Count(app|s) represents the number of times the application is used in the third time period after entering the first fence in the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the third time period after entering the first fence each time in the first historical time period;

[0020] calculate Among them, willson ctr (app|s) represents the modified usage probability of the application in the first historical time period in the third time period after entering the first fence, n is the number of times entering the first fence in the first historical time period, and z is a constant;

[0021] calculate Among them, lift represents the lift of the application in the third time period after entering the first fence in the first historical time period, and p(app|all) represents the ratio of the number of times the application is used in the third time period after entering the fences corresponding to multiple recommendation factors in the first historical time period to the number of times multiple applications are used in the third time period after entering the fences corresponding to multiple recommendation factors in the first historical time period. ctr (app|s) is the Wilson usage probability. p refers to the uncorrected usage probability of each APP under each recommendation factor.

[0022] In some possible implementations, the weighted probability of the second application is greater than the weighted probability of the fourth application; and the weighted probability of the application in the third time period after entering the first fence is calculated in the following manner: p(app|s)=Count(app|s) / Count(s) is calculated, where p(app|s) is the usage probability of the application in the third time period after entering the first fence in the first historical time period, Count(app|s) represents the number of times the application is used in the third time period after entering the first fence in the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the third time period after entering the first fence each time in the first historical time period; p(app|days)=count(app|s,day) / c is calculated. ount(s,day), where p(app|days) represents the probability of the number of days of use of the application in the third time period after entering the first fence in the first historical time period, count(app|s,day) represents the number of days of use of the application in the third time period after entering the first fence in the first historical time period, and count(s,day) represents the number of days of use of multiple applications in the third time period after entering the first fence in the first historical time period; calculate Weight_P(app|s) = normalize(p(app|s)*p(app|Days)), where Weight_P(app|s) represents the weighted probability of the application in the third time period after entering the first fence, and normalize() represents normalization. In this way, the value of Weight_P(app|s) comprehensively considers the number of days of use and the number of times of use, and obtains a more accurate weighted probability of each application being used, and after normalization, it is convenient to compare the size.

[0023] In some possible implementations, the second confidence of the second application is greater than the second confidence of the fourth application; the second confidence of the application (i.e., the scenario confidence below) is: the ratio of the second ranking of the application to the total number of applications in the electronic device, wherein the second ranking of the application is the ranking of the weighted probability of the application in the order of the weighted probabilities of the various applications in the electronic device from small to large. In this way, it is convenient to compare the sizes of the various second confidences, and to conveniently set and output several APPs whose scenario confidences are greater than the scenario confidence threshold as the user's preferred APPs under one recommendation factor. It can be understood that an appropriate number of APPs are recommended to the user based on the above total number of APPs. For example, the larger the total number of APPs, the greater the number of preferred APPs determined to be under the first recommendation factor after a fence is triggered.

[0024] In some possible implementations, the method further includes: determining that the number of days entering the first fence in the first historical time period is greater than the first day. It is understandable that the electronic device can filter out historical data under the recommendation factor whose number of days of appearance is less than or equal to the first day (e.g., 1 day). However, applications under the recommendation factor whose number of days of appearance in the first historical time period is less than or equal to the first day are usually occasionally used by the user and cannot reflect the user's application usage preferences under various recommendation factors.

[0025] In some possible implementations, the first recommendation factor includes at least one of the following: a time factor, a space factor, and a device usage factor, wherein the device usage factor is used to indicate whether the electronic device is connected to an external device (such as wearing headphones or connecting to a car's Bluetooth); the first fence includes at least one of the following: a time fence, a geographic fence, and a device fence.

[0026] In some possible implementations, the application corresponding to the first fence recommended to the user is displayed in an application suggestion card of the electronic device, such as YOYO TM suggestion.

[0027] In the second aspect, the embodiment of the present application provides an application recommendation method, which is applied to an electronic device, and a first application to be recommended and a second application to be recommended are installed on the electronic device; and the method includes: determining that the electronic device enters a first fence corresponding to a first recommendation factor; displaying a first application to be recommended corresponding to the first fence recommended to the user, wherein the preference probability of the first application to be recommended is greater than the preference probability of the second application to be recommended that is not recommended to the user, and the higher the preference probability, the higher the probability that the application has been used in the first target time period after entering the first fence in the first historical time period. The first application to be recommended can be the first application mentioned above, the second application to be recommended can be the third application mentioned above, and the first application to be recommended and the second application to be recommended can both include multiple applications. It can be understood that the first target time period can be the second time period mentioned above. For example, the duration of the first target time period can be the first duration mentioned below, such as the second time period is within the first 5 minutes after entering the first fence. In this way, the first application to be recommended is the APP that the user prefers to use within the first 5 minutes after entering the first fence.

[0028] In some possible implementations, the preference probability is calculated by the following formula: Calculate topK prop = count(s|app in topN) / M; calculate hit prop =exp(lift-a)*topK prop ; Among them, topK prop represents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each sequence includes the application used in the first target time period after entering the first fence corresponding to the first recommendation factor, count(s|app in topN) represents the number of sequences containing applications in the multiple sequences corresponding to the first recommendation factor, topK prop Indicates the probability of using the application in the first target time period after entering the first fence in the first historical time period; prop represents the preference probability applied in the first target time period after entering the first fence in the first historical time period, exp() represents exponential operation, topK propIt represents the usage probability of the application in the first target time period after entering the first fence within the first historical time period, lift represents the lift degree of the application in the second target time period after entering the first fence within the first historical time period, and a is a constant related to the lift degree. In this way, the present application comprehensively considers the usage probability of the application in the first target time period after entering the first fence, and the usage probability of the application in the entire second target time period (i.e., the third time period above) after entering the first fence. When the user enters the fences of different recommendation factors, it can accurately recommend the user's preferred APP, which is convenient for the user to quickly use the required APP, greatly improving the user experience. In addition, the above-mentioned first recommendation factor can be any one of multiple recommendation factors.

[0029] In some possible implementations, a third application to be recommended is also installed in the electronic device, the first confidence of the first application to be recommended is greater than the first confidence of the third application to be recommended, and the first confidence is: the ratio of the first ranking of the application to the total number of applications in the electronic device, wherein the first ranking of the application is the ranking of the preference probability of the application in the ascending order of the preference probabilities of the various applications in the electronic device. In this way, it is convenient to compare the sizes of the various first confidences (i.e., the trigger confidence below), and it is convenient to set and output several APPs whose trigger confidence is greater than the trigger confidence threshold as the user's preferred APPs under one recommendation factor.

[0030] In some possible implementations, the lift is calculated in the following manner: p=Count(app|s) / Count(s) is calculated, where p is the probability of using the application in the second target time period after entering the first fence in the first historical time period, Count(app|s) represents the number of times the application is used in the second target time period after entering the first fence in the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the second target time period after entering the first fence each time in the first historical time period;

[0031] calculate Among them, willson ctr (app|s) represents the corrected usage probability of the application in the first historical time period in the second target time period after entering the first fence, n is the number of times entering the first fence in the first historical time period, and z is a constant;

[0032] calculate Among them, lift represents the lift of the application in the second target time period after entering the first fence in the first historical time period, and p(app|all) represents the ratio of the number of times the application is used in the second target time period after entering the fences corresponding to multiple recommendation factors in the first historical time period to the number of times multiple applications are used in the second target time period after entering the fences corresponding to multiple recommendation factors in the first historical time period. ctr (app|s) is the Wilson usage probability. p refers to the uncorrected usage probability of each APP under each recommendation factor.

[0033] In a third aspect, an embodiment of the present application provides an application recommendation method, which is applied to an electronic device, on which a first application to be recommended and a second application to be recommended are installed; and the method includes: determining that the electronic device enters a first fence corresponding to a first recommendation factor; displaying a first application to be recommended corresponding to the first fence recommended to the user, wherein the preference probability of the first application to be recommended is greater than the preference probability of the second application to be recommended that is not recommended to the user, and the higher the preference probability, the higher the probability of the application appearing in the first N applications used after entering the first fence within the first historical time period, where N is a positive integer. The first application to be recommended may be the first application mentioned above, the second application to be recommended may be the third application mentioned above, and the first application to be recommended and the second application to be recommended may include multiple applications. In this way, the first application to be recommended is the APP that the user prefers to use during the process of using the first N (such as 5) applications after entering the first fence.

[0034] In some possible implementations, the preference probability is calculated by the following formula: Calculate topK prop = count(s|app in topN) / M; calculate hit prop =exp(lift-a)*topK prop ; Among them, topK prop represents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each sequence includes the top N applications used after entering the first fence corresponding to the first recommendation factor, count(s|app in topN) represents the number of sequences containing applications in the multiple sequences corresponding to the first recommendation factor, topK prop Indicates the probability of an application appearing in the first N applications used after entering the first fence in the first historical time period; hit prop represents the preference probability of the application among the first N applications used after entering the first fence in the first historical time period, exp() represents the exponential operation, topK prop(app|s) represents the probability of the application appearing in the top N applications used after entering the first fence in the first historical time period, lift represents the lift of the application in the target time period after entering the first fence in the first historical time period, and a is a constant related to the lift. For example, the lift value is a value between 0 and 1, and a is 0.5. For example, 0.5. For example, the lift value is a value between 0 and 1. Then, exp(lift-a) is used to calculate the topK prop Make corrections. If lift is greater than 0.5, increase topK prop Get hit prop On the contrary, if lift is less than 0.5, reduce topK prop Get hit prop , lift = 0.5, no need to adjust topK prop .

[0035] In some possible implementations, a third application to be recommended is also installed in the electronic device, the first confidence of the first application to be recommended is greater than the first confidence of the third application to be recommended, and the first confidence is: the ratio of the first ranking of the application to the total number of applications in the electronic device, wherein the first ranking of the application is the ranking of the preference probability of the application in the ascending order of the preference probabilities of each application in the electronic device.

[0036] In some possible implementations, the lift is calculated in the following manner: p=Count(app|s) / Count(s) is calculated, where p is the probability of using the application in the target time period after entering the first fence in the first historical time period, Count(app|s) represents the number of times the application is used in the target time period after entering the first fence in the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the target time period after entering the first fence each time in the first historical time period;

[0037] calculate Among them, willson ctr (app|s) represents the corrected usage probability of the application in the first historical time period in the target time period after entering the first fence, n is the number of times the application enters the first fence in the first historical time period, and z is a constant;

[0038] calculate Among them, lift represents the lift of the application in the target time period after entering the first fence in the first historical time period, and p(app|all) represents the ratio of the number of uses of the application in the target time period after entering the fences corresponding to multiple recommendation factors in the first historical time period to the number of uses of multiple applications in the target time period after entering the fences corresponding to multiple recommendation factors in the first historical time period.

[0039] In a fourth aspect, an embodiment of the present application provides a readable medium having instructions stored thereon, which, when executed on an electronic device, causes the electronic device to execute an application recommendation method such as the first aspect and any possible implementation thereof, or an application recommendation method such as the second aspect and any possible implementation thereof, or an application recommendation method such as the third aspect and any possible implementation thereof.

[0040] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a memory for storing instructions executed by one or more processors of the electronic device, and a processor, which is one of the processors of the electronic device, for executing an application recommendation method as in the first aspect and any possible implementation thereof, or an application recommendation method as in the second aspect and any possible implementation thereof, or an application recommendation method as in the third aspect and any possible implementation thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 According to some embodiments of the present application, a YOYO suggestion displayed on a mobile phone is shown. TM Schematic diagram of the card;

[0042] Figure 2 According to some embodiments of the present application, a schematic diagram of an application recommendation process of an electronic device is shown;

[0043] Figure 3 According to some embodiments of the present application, a schematic diagram of the hardware structure of a mobile phone is shown;

[0044] Figure 4 According to some embodiments of the present application, a schematic diagram of a software system of a mobile phone is shown;

[0045] Figure 5 According to some embodiments of the present application, a schematic diagram of a process of executing application recommendation by a module of an electronic device is shown;

[0046] Fig. 6A According to some embodiments of the present application, a schematic diagram of the process of APP preference mining within a period of time after each fence is triggered is shown;

[0047] Figure 6B According to some embodiments of the present application, a schematic diagram of the process of mining APP preferences under various recommendation factors is shown;

[0048] Fig. 7A According to some embodiments of the present application, a flowchart of an application recommendation method is shown;

[0049] Figure 7BAccording to some embodiments of the present application, a YOYO suggestion displayed on a mobile phone is shown. TM Schematic diagram of the card;

[0050] Fig. 8A According to some embodiments of the present application, a flowchart of an application recommendation method is shown;

[0051] Figure 8B According to some embodiments of the present application, a YOYO suggestion displayed on a mobile phone is shown. TM Schematic diagram of the card;

[0052] Fig. 9 According to some embodiments of the present application, a flowchart of an application recommendation method is shown;

[0053] Fig.10 According to some embodiments of the present application, a flowchart of an application recommendation method is shown;

[0054] Fig.11 According to some embodiments of the present application, a flowchart of an application recommendation method is shown. DETAILED DESCRIPTION

[0055] Illustrative embodiments of the present application include, but are not limited to, application recommendation methods, media, and electronic devices.

[0056] As can be seen from the background technology, in the prior art, the APP suggestion cards of electronic devices usually only push high-frequency applications. Specifically, mobile phones and other electronic devices can set different fences. For example, the fence set by the mobile phone can be a geographic fence, such as the geographic fence where the subway station is located, the geographic fence where the user's home is located, the geographic fence where the user's company is located, etc. However, in specific scenarios, users usually have different behavioral preferences when using mobile phones and other electronic devices. For example, when users enter different fences, they usually prefer to click to use different APPs. Specifically, when a user just enters the subway fence, he may turn on Cloud Quick Pass. TM ; When a user first enters the home fence, he or she may open the community access control app, smart home app, etc.; when a user first enters the company fence, he or she may open the sign-in app, etc. However, the current APP recommendation scheme does not consider the different behavioral preferences of users when entering fences in different scenarios. It may only push high-frequency applications, and in special scenarios, the overall frequency is relatively low but the user's preferred APP will not be pushed. Then, every time a user first enters the subway fence, he or she will use Cloud QuickPass. TM Scan the QR code, but in the subway scenario, users may use WeChat multiple times TM ,Tik Tok TM ,quick worker TM And other applications, such as Cloud QuickPass TM The frequency of use is lower than WeChat TM ,Tik TokTM ,quick worker TM Applications such as Cloud QuickPass TM The overall usage frequency is relatively low, resulting in the APP suggestion card recommending APPs mainly WeChat TM ,Tik Tok TM ,quick worker TM Applications such as QQ Pay, excluding the user-preferred Cloud QuickPass TM , the user experience is poor. For the convenience of description, the application program is referred to as APP or application in this application.

[0057] It can be understood that in each embodiment of this application, the frequency of use refers to the number of times a user clicks to use an APP regardless of the scenario within a period of time. For the convenience of description, this application also refers to the number of clicks on the APP as the number of uses or the number of clicks to use. For example, within 1 month, the user clicked on Douyin 60 times. TM , 22 Cloud QuickPass transactions TM , then Tik Tok TM The usage frequency is higher than that of Cloud QuickPass TM frequency of use.

[0058] In each embodiment of the present application, preference refers to the probability of a user using each APP under specific conditions. Accordingly, the specific condition may be that the user uses the APP during the entire time period in a specific scene such as a home scene or a subway scene, or the recommendation factor may be that the user uses the APP during the process of using the first few applications (such as applications used within the first 5 minutes, or the first 5 applications used) after entering a specific scene such as a home scene or a subway scene, or the specific condition may be that the user uses the APP after entering a specific scene such as a home scene or a subway scene and has used several applications (such as the first 5 applications or applications used within the first 5 minutes).

[0059] It can be understood that in each embodiment of the present application, the user's preferred APP can be determined based on different recommendation factors, where the recommendation factor can be a spatial factor, such as the different scenarios in which the user is located, such as home, company, subway, shopping mall, etc., or it can be a usage factor, such as whether the user is wearing headphones, whether the user device is connected to the car Bluetooth, etc., or it can be a time factor, for example, dividing a day into multiple time intervals, such as dividing a day into 24 time intervals in units of hours.

[0060] In some embodiments, the present application may determine the preferred APP within the first few minutes of entering the fence corresponding to a recommendation factor, or determine the preferred APP in the process of using the first N applications after entering the fence corresponding to a recommendation factor, that is, the probability of the user using each APP when the user just enters the fence. Taking the home scene as an example, when the user just enters the home fence, such as when the user arrives at the gate of the community where the home is located, the preference refers to the probability of each APP used when the user enters the home fence. At this time, a larger preference for the APP indicates that the user is very likely to click to use the APP when entering the home fence, otherwise it indicates that the user usually does not click to use the APP when entering the home fence. For example, when a user uses an electronic device to just enter the home fence, such as when the user arrives at the gate of the community where the home is located, the user's preferred APP is the access card APP.

[0061] In some other embodiments, the present application can determine the preferred APP of the user a few minutes (such as 5 minutes) after entering the fence corresponding to a recommendation factor, or determine the preferred APP after using the first N applications after entering the fence corresponding to a recommendation factor, that is, the probability of the user using each APP after the user has entered the fence of the recommendation factor for a certain period of time. Taking the home scene as an example, when the user is at home, the preference refers to the probability of each APP used by the user in the home scene. At this time, a greater preference for an APP indicates that the user is very likely to click to use the APP at home, otherwise it means that the user usually does not click to use the APP at home. For example, when a user uses an electronic device at home, the preferred APP is usually an entertainment APP, such as TikTok. TM and WeChat TM wait.

[0062] It should be emphasized that the application recommendation method in this application is mainly used to identify and recommend the user's preferred APPs when the user just enters a fence corresponding to a recommendation factor, such as the first 5 minutes after entering the fence or the first N applications clicked and used after entering the fence.

[0063] Then, in order to solve the problem that the applications recommended by the above application recommendation method do not meet the actual preferences of users, the embodiment of the present application provides an application recommendation method, which pre-learns the APPs that users prefer to click and use within the first 5 minutes after the electronic device enters each fence, or during the process of using the first N applications after entering the fence. Then, when a fence trigger is detected, an APP suggestion card such as YOYO suggestion TM Recommend the user a preferred APP corresponding to the triggering moment of the fence.

[0064] Specifically, the method can obtain multiple historical application usage information in the most recent historical time period, such as the most recent month, wherein each historical application usage information records the APP that the user clicked to use under a certain recommendation factor, such as recommendation factor s1, in a certain historical time period, such as the historical time period T1. Then, the preference probability of each APP when entering the fence of each recommendation factor in the historical time period T1 is calculated through the above multiple historical application usage information. Among them, the value of the preference probability of a certain APP, such as APP1 under the recommendation factor s1 depends on: the use probability of APP1 when entering the fence corresponding to the recommendation factor s1 in T1 and the promotion degree of APP1 under the recommendation factor s1 in T1. Among them, the value of the promotion degree depends on: the use probability of the application under the recommendation factor s1 in T1 and the use probability of APP1 under multiple recommendation factors in T1. It can be understood that the preference probability of APP1 under the recommendation factor s1 is used to reflect the association between the fence triggering moment of the recommendation factor s1 in the historical time period T1 and the user clicking to use APP1, that is, it reflects the user's preference for clicking to use APP1 within a period of time after entering the fence of the recommendation factor. In addition, the promotion degree of APP1 under the recommendation factor s1 reflects the correlation between the recommendation factor s1 and the user's click to use APP1 in the historical time period T1, that is, it reflects the user's preference for clicking to use APP1 after entering the recommendation factor s1. It can be understood that the higher the promotion degree of APP1 under the recommendation factor s1, the higher the probability of being biased. Furthermore, this application recommends to users the APPs whose preference probability under the recommendation factor s1 is greater than the preference threshold as the APPs preferred by users within a period of time after the fence of the recommendation factor s1 is triggered. These APPs are the APPs that users prefer to click to use when they just enter the period of time of the recommendation factor s1 in T1.

[0065] In addition, the application can set a corresponding fence for each recommendation factor in advance. When an electronic device is identified to enter a fence, the APP suggestion card can be used to recommend the user's preferred APP within a period of time after entering the fence. In this way, the APP recommendation in the application is not limited by the overall usage frequency of the APP. When the user enters the fence of different recommendation factors, the user's preferred APP can be accurately recommended, which is convenient for the user to quickly use the required APP, greatly improving the user experience.

[0066] In some embodiments, the usage probability of APP1 when entering the fence corresponding to the recommendation factor s1 within T1 is taken as an example. For example, the electronic device can find out multiple continuous historical application usage information under the recommendation factor s1 within the historical time period T1, select the first N APPs that the user has clicked and used in each continuous historical application usage information as an APP sequence, and then obtain multiple APP sequences corresponding to the recommendation factor s1. For example, N is a preset number, such as 5. Then, the present application can calculate the probability of appearance of APP1 in multiple APP sequences corresponding to the recommendation factor s1 (also called hit probability). Then, according to the hit probability of APP1 when entering the fence and the lifting degree of APP1 under the recommendation factor s1, the preference probability of APP1 is calculated. For example, if the lifting degree of APP1 under the recommendation factor s1 is large, the hit probability of APP1 can be increased by the lifting degree to increase the corresponding preference probability. Conversely, if the lifting degree of APP1 under the recommendation factor s1 is small, the hit probability of APP1 can be reduced by the lifting degree to reduce the corresponding preference probability.

[0067] In some embodiments, the usage probability of APP1 when entering the fence corresponding to the recommendation factor s1 within T1 is taken as an example. For example, the electronic device can find out multiple continuous historical application usage information under the recommendation factor s1 within the historical time period T1, select the APP used within the first time length (such as 5 minutes) after entering the corresponding fence in each continuous historical application usage information as an APP sequence, and then obtain multiple APP sequences corresponding to the recommendation factor s1. Then, the present application can calculate the occurrence probability (also called hit probability) of APP1 in the multiple APP sequences corresponding to the recommendation factor s1, that is, the hit probability of APP1 within a period of time (such as the first 5 minutes) triggered by entering the fence of the recommendation factor s1. Then, according to the hit probability of APP1 when entering the fence and the lifting degree of APP1 under the recommendation factor s1, the preference probability of APP1 is calculated. For example, if the lifting degree of APP1 under the recommendation factor s1 is large, the hit probability of APP1 can be increased by the lifting degree to increase the corresponding preference probability. On the contrary, if the lifting degree of APP1 under the recommendation factor s1 is small, the hit probability of APP1 can be reduced by the lifting degree to reduce the corresponding preference probability.

[0068] For example, referring to Table 1, T1 is the number of times users click to use each APP in the last week, that is, the usage of these APPs.

[0069] Table 1:

[0070] Subway Scene 1 Subway Scene 2 Subway Scene 3 Other scenes 1 … All scenes <![CDATA[UnionPay QuickPass]]> TM > 2 times 2 times 2 times 0 times … 6 times <![CDATA[Douyin TM > 8 times 4 times 0 times - … 62 times … 0 0 0 0 … - All APPs 10 times 6 times 2 times - … 300 times

[0071] As shown in Table 1, in the past week, the user entered the subway scene 3 times, and clicked to use UnionPay QuickPass once in the subway scene.TM 2 times, click to use TikTok TM 8 times; enter the subway scene and use Cloud QuickPass 2 times TM 2 times, click to use TikTok TM 4 times; enter the subway scene and tap 3 times to use Cloud QuickPass TM 2 times, and no click to use TikTok TM In addition, in the past week, the user also clicked and used various APPs in other scenarios (such as other scenario 1). For example, in the past week, the user clicked and used APP 300 times in total, and clicked and used Douyin 50 times in other scenarios. TM , but did not use Cloud QuickPass TM .

[0072] So, in the past week, in all scenarios, Cloud QuickPass TM The usage frequency is 6 times, Douyin TM The usage frequency is 62 times, and Cloud QuickPass has been used in subway scenarios in the past week. TM The usage frequency is 6 times, Douyin TM The frequency of use is 12 times. Obviously, Douyin has TM The overall usage frequency is higher than that of Cloud QuickPass TM , and its usage frequency in subway scenarios is also higher than that of Cloud QuickPass. TM .

[0073] Furthermore, the promotion degree can consider the usage probability of each APP under a certain recommendation factor and the usage probability of the corresponding APP under all recommendation factors. Assume that in the past week, the user only used Cloud QuickPass in various subway scenarios. TM and TikTok TM , combined with the three subway scenarios shown in Table 1 TM The number of times used is 6, Douyin TM The number of times used is 12. Under all the recommended factors, Cloud QuickPass TM The number of times used is 6, Douyin TM The number of times used is 62, and the number of times used for all apps under all recommended factors is 300. Then, TM The improvement can be (6 / 18) / (6 / 300)=16.66. TM The improvement can be (12 / 18) / ((62) / 300)=3.22. Then, in the subway scenario, TM The improvement is higher than that of TikTok TM Obviously, in this case, although Douyin in the subway scene TM The overall usage frequency is high, but Douyin TM It is not an APP for increasing user base in subway scenarios.

[0074] Furthermore, combined with Table 1, taking the number of the first N apps after entering the subway fence as 5 as an example, assuming that in the three subway scenarios, the user clicked to use UnionPay in the first 5 apps clicked after entering the subway fence. TM 1 time. In addition, in subway scene 1, after entering the subway fence, TikTok was clicked in the first 5 apps used. TM 1 time, while in subway scenes 2 and 3, Douyin was not clicked in the first 5 apps clicked after entering the subway fence. TM So, in the past week, when Cloud QuickPass entered the subway scene, TM The hit probability is 3 / 3=1, Douyin TM The hit probability is 1 / 3 = 0.33. Assuming the correction factor 2 is 8, then Cloud QuickPass TM The improvement (i.e. 16.66) is greater than the correction factor 2, which can increase the Cloud QuickPass TM The hit probability is used to obtain the corresponding preference probability. TM The improvement (i.e. 3.22) is less than the correction factor 2, so the vibrato can be reduced TM The hit probability is used to obtain the corresponding preference probability. Obviously, within a period of time after entering the subway fence in the past week, Cloud QuickPass TM The probability of preference is greater than TikTok TM The preference probability is that users prefer to use UnionPay QuickPass within the first 5 minutes after entering the subway station. TM Scan the QR code to enter the station.

[0075] It can be understood that in actual applications, in order to facilitate the quantification and comparison of various parameters between different APPs, parameters such as the degree of improvement and the probability of preference can also be normalized, which will be described in detail below and will not be repeated here.

[0076] In some embodiments, for multiple apps clicked and used within a period of time after a fence is triggered, such as the first N apps clicked and used after entering the fence, or the apps used within the first 5 minutes after entering the fence, the present application can sort the preference probability of each app from small to large to obtain the ranking of each app, and divide the ranking of each app by the total number of apps in the electronic device to obtain the trigger confidence (also called the first confidence) of each app, so that several apps with a trigger confidence greater than or equal to the trigger confidence threshold within a period of time after a fence is triggered are used as the preferred apps within a period of time after the fence is triggered. It can be understood that an appropriate number of apps are recommended to the user based on the above total number of apps, for example, the larger the total number of apps, the more preferred apps are determined at the moment of fence triggering.

[0077] It can be understood that a fence usually corresponds to a recommendation factor, and the moment of entering the fence also indicates the moment of triggering the recommendation factor, and after the electronic device enters a fence, it may continue to be under the recommendation factor corresponding to the fence for a long time. For example, when the electronic device enters the subway fence (such as the geographic fence where the subway station is located), it means that the electronic device has entered the subway scene (i.e., a recommendation factor), and the user may have been riding the subway for half an hour, so that the electronic device continues to be in the subway scene for a long time.

[0078] In some embodiments, based on the historical application usage information, the present application can calculate the number of days of appearance of each recommendation factor in history, the number of times the APP is used under all recommendation factors, the number of times the APP is used under each recommendation factor, the number of times each APP is used under all recommendation factors, the number of times each APP is used under each recommendation factor, etc. Then, based on this information, the promotion degree of each APP under each recommendation factor and the hit probability and preference probability of each APP within a period of time after each fence is triggered are calculated, and the trigger confidence of each APP within the first time period after each fence is triggered or within the process of using the first N applications after the trigger can be calculated based on the promotion degree, hit probability and preference probability of each APP. Among them, the detailed calculation process of preference probability, promotion degree and trigger confidence will be described in detail below and will not be repeated here.

[0079] In some embodiments, the electronic device can subscribe to different fences for different recommendation factors. Then, the electronic device can determine the recommendation factor in which the electronic device is located by identifying the fence. For example, the recommendation factors of the electronic device include but are not limited to: home, company, hospital, shopping mall, airport, railway station, high-speed service area, subway, wearing headphones, and car Bluetooth.

[0080] Among them, the fence refers to a trigger condition. When the fence meets the trigger condition, the mobile phone or other electronic device starts to execute a specific business or provide a specific service, such as displaying, refreshing or closing messages, warning information or cards. For example, the card provided by the electronic device in this application can be an APP suggestion card such as YOYO suggestion card. TM For the convenience of description, in some of the following embodiments, APP suggestion cards are also referred to as cards or YOYO suggestion cards. TM card.

[0081] In some embodiments, since the APP that a user prefers to click to use when entering a fence is usually different from the APP that a user prefers to click to use after entering the fence, the electronic device in the present application can not only predetermine the preferred APP within a period of time after the fence is triggered, but also predetermine the preferred APP within the scene corresponding to the fence. Thus, when the electronic device recognizes that the user has just entered the fence using the electronic device, the preferred APP within a period of time after the fence is triggered is recommended to the user through the APP suggestion card. Furthermore, after the electronic device recognizes that the user has just entered the fence using the electronic device for a period of time or moved a certain distance, if the electronic device is still under the recommendation factor corresponding to the fence, the preferred APP under the recommendation factor corresponding to the fence is recommended to the user through the APP suggestion card.

[0082] In some embodiments, the present application can calculate the weighted probability of each APP under each recommendation factor in the historical time period T1 through multiple historical application usage information in the historical time period, such as calculating the weighted probability of APP1 under each recommendation factor s1 in T1. The value of the weighted probability of APP1 under the recommendation factor s1 depends on the probability of users clicking to use APP1 under the recommendation factor s1 in T1, the number of days users click to use APP1 under the recommendation factor s1, and the number of days the recommendation factor s1 appears. For example, in the subway scenario in the past week, users use Cloud QuickPass. TM The number of days of use is 3 days, and the number of days of appearance of the subway scene is also 3 days. It can be understood that the weighted probability of APP1 under the recommendation factor s1 is used to reflect whether the user clicks to use APP1 more times and for more days under the recommendation factor s1 within T1. Furthermore, the present application can recommend to the user the APPs whose lift degree under the recommendation factor s1 is greater than the lift degree threshold and whose weighted probability is greater than the weighted probability threshold as the user's preferred APPs under the recommendation factor s1. These APPs are the APPs that the user prefers to click and use under the recommendation factor in the historical time period and have a large number of uses and days of use. Thereby, it is helpful to improve the accuracy of recommending preferred APPs to users under specific recommendation factors through APP suggestion cards.

[0083] In some embodiments, the present application can calculate the scene confidence (also called the second confidence) of each APP under each recommendation factor in the historical time period according to the weighted probability of each APP under each recommendation factor in the above-mentioned historical time period and the total number of APPs in the electronic device. Then, the scene confidence of each APP under each recommendation factor is equivalent to the total number of APPs in the electronic device. The weighted probability of each APP under these recommendation factors is normalized, that is, these weighted probabilities are numerically corrected, so as to facilitate the comparison of the size between each weighted probability. In addition, considering the total number of APPs can facilitate the setting of the size of the scene confidence threshold, and the scene confidence threshold is used to measure the weighted probability after numerical correction, so as to facilitate the setting and output of several APPs with scene confidence greater than the scene confidence threshold as the user's preferred APP under one recommendation factor. At this time, the present application can use an APP with a lift greater than the lift threshold and a scene confidence greater than the scene confidence threshold under a recommendation factor as the user's preferred APP under the recommendation factor.

[0084] It can be understood that the present application mainly describes the process of the APP suggestion card recommending the preferred APPs to the user within a period of time after the fence is triggered.

[0085] In some embodiments, the APP suggestion card generally includes one or more third-party APPs or mini-programs icons. For example, the APP suggestion card may include multiple APP icons such as TikTok. TM , Taobao TM and Alipay TM etc., or may only include an APP or mini-program icon, such as Cloud QuickPass. TM Card or WeChat TM card.

[0086] In some embodiments, the APP suggestion card in the present application can be fixedly displayed in a fixed area of ​​the desktop interface, for example, the APP suggestion card can be displayed in a rectangular area in the upper left corner of the main interface. In other embodiments, the APP suggestion card can be temporarily superimposed and displayed on any interface when entering a fence, and the APP suggestion card can be automatically canceled after the user clicks on the APP suggestion card to enter a recommended APP or mini program.

[0087] In some embodiments, the shape, number of apps, and arrangement format of the app suggestion card in this application are not specifically limited. For example, the app suggestion card can be in any shape such as a rectangle or a circle. TM The card's APP arrangement format can be 4 APPs in 2×2 format, 4 APPs in 2×3 format, or Figure 1 3 APPs are shown with 1 APP in the 1st row and 2 APPs in the 2nd row.

[0088] Generally, fences include multiple types, such as geographic fences, time fences, device fences, and associated event fences, etc. In some embodiments, the fences subscribed by the electronic device for a recommendation factor may include one or more types, such as the fence for a user taking the subway includes a geographic fence and a time fence, which is not specifically limited in this application.

[0089] Geofences are used to characterize trigger conditions related to locations. Specifically, geofences are location-based services (LBS) that use a virtual fence to enclose a geographic area with a virtual boundary, that is, a virtual boundary defines a specific geographic area. In some embodiments, the shape of the geofence can be circular, polygonal, or irregular, which is not specifically limited in this application. In some embodiments, the size of the geofence can be determined based on business needs, which is not specifically limited in this application. For example, the locations of geofences include but are not limited to: home, company, hospital, shopping mall, airport, train station, highway service area, government agency, subway, etc. In some embodiments, the electronic device can monitor the current location of the electronic device in real time to determine whether the electronic device has entered the geofence.

[0090] The time fence is used to represent the trigger conditions related to time. For example, taking the subway scene as an example, the historical time when the user enters and exits the subway station (for example, 7:00 every day) is set as the time fence corresponding to the subway scene.

[0091] Device fences are used to represent trigger conditions related to a specific usage state of a specific electronic device. For example, a specific usage state may be when a user wears headphones or is connected to a car's Bluetooth.

[0092] The associated event fence is used to represent the trigger conditions related to a specific event. Among them, the specific event can be the trigger condition for entering the geo-fence, such as every day before entering the subway, users will use electronic devices to scan the code to unlock the shared bicycle.

[0093] In some embodiments, the electronic device can determine whether the electronic device enters a geo-fence, time fence, or other fence corresponding to a recommendation factor to determine whether the electronic device should switch to the recommendation factor, and then recommend the user's preferred APPs in the time period before or after entering the fence through the displayed APP suggestion card. For example, when the electronic device enters the geo-fence corresponding to the subway, the electronic device can automatically pop up a pop-up containing Cloud QuickPass. TM App suggestion cards for apps.

[0094] In some embodiments, taking home, company, and subway scenes as examples, the card display process of the electronic device when the electronic device enters the fence of these recommended factors is described. Figure 2 As shown, it is assumed that the electronic device has registered geo-fence 01 for the recommendation factor of the user taking the subway, geo-fence 02 for the recommendation factor of the user at home, and geo-fence 03 for the recommendation factor of the user at work. Then, after the user 04 enters the geo-fence 01, the electronic device such as a mobile phone can display the YOYO suggestion card 05, which can include the user's preference to click on the payment APP such as UnionPay when taking the subway. TM , and information apps such as Sina Weibo TM , Toutiao TM Or reading apps such as WeChat Reading TM After the user 04 enters the geo-fence 02, the electronic device such as a mobile phone may display a YOYO suggestion card 06, which may include entertainment apps that the user prefers to click on when at home, such as WeChat. TM ,Tik Tok TM After the user 04 enters the geo-fence 03, the electronic device such as a mobile phone can display a YOYO suggestion card 07, which can include work-related apps such as documents, conference software, etc. that the user prefers to click on when in the company. It can be understood that the fences registered by the electronic device for each recommendation factor in the embodiment of the present application are not limited to Figure 2 The geographic fence shown may also be a device fence, a time fence, etc., which is not specifically limited in this application.

[0095] In some embodiments, the application recommendation method in the present application can be applied to various electronic devices, including but not limited to: mobile phones, tablet computers, wearable devices, vehicle-mounted devices, augmented reality (AR) devices, virtual reality (VR) devices, laptop computers, ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (PDA) or smart screens and other terminal devices with display functions, and can also be other devices or devices that can recommend applications. The embodiments of the present application do not impose any restrictions on the specific types of electronic devices.

[0096] Taking a mobile phone as an example, Figure 3 A schematic diagram of the hardware structure of a mobile phone provided in an embodiment of the present application is shown.

[0097] like Figure 3As shown, the mobile phone 10 may include a processor 110, a power module 140, a memory 180, a mobile communication module 130, a wireless communication module 120, a sensor module 190, an audio module 150, a camera 170, an interface module 160, a button 101 and a display screen 102, etc.

[0098] The processor 110 may include one or more processing units, for example, a processing module or processing circuit including a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor DSP, a microprocessor (MCU), an artificial intelligence (AI) processor, or a programmable logic device (field programmable gate array, FPGA). Different processing units may be independent devices or integrated into one or more processors. A storage unit may be provided in the processor 110 for storing instructions and data. In some embodiments, the storage unit in the processor 110 is a cache memory 180.

[0099] The power module 140 may include a power source, a power management component, etc. The power source may be a battery. The power management component is used to manage the charging of the power source and the supply of power to other modules.

[0100] The mobile communication module 130 may include, but is not limited to, an antenna, a power amplifier, a filter, a low noise amplifier (LNA), etc. The mobile communication module 130 may provide a solution for wireless communications including 2G / 3G / 4G / 5G, etc., applied to the mobile phone 10. The mobile communication module 130 may receive electromagnetic waves by an antenna, and perform filtering, amplification, and other processing on the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation. The mobile communication module 130 may also amplify the signal modulated by the modulation and demodulation processor, and convert it into electromagnetic waves for radiation through the antenna. In some embodiments, at least some of the functional modules of the mobile communication module 130 may be arranged in the processor 110. In some embodiments, at least some of the functional modules of the mobile communication module 130 may be arranged in the same device as at least some of the modules of the processor 110. Wireless communication technologies may include global system for mobile communications (GSM), general packet radio service (GPRS), code division multiple access (CDMA), wide band code division multiple access (WCDMA), time-division code division multiple access (TD-SCDMA), long term evolution (LTE), Bluetooth (BT), global navigation satellite system (GNSS), wireless local area networks (WLAN), near field communication (NFC), frequency modulation (FM) and / or field communication (NFC), infrared technology (IR), etc.The GNSS may include a global positioning system (GPS), a global navigation satellite system (GLONASS), a Beidou navigation satellite system (BDS), a quasi-zenith satellite system (QZSS) and / or a satellite based augmentation system (SBAS).

[0101] The wireless communication module 120 may include an antenna, and transmit and receive electromagnetic waves via the antenna. The wireless communication module 120 may provide wireless communication solutions including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication technology (NFC), infrared technology (IR), etc., which are applied to the mobile phone 10. The mobile phone 10 can communicate with the network and other devices through wireless communication technology.

[0102] In some embodiments, the mobile communication module 130 and the wireless communication module 120 of the mobile phone 10 may also be located in the same module.

[0103] The display screen 102 is used to display a human-computer interaction interface, images, videos, etc. The display screen 102 includes a display panel. For example, the display screen 102 may display a YOYO suggestion TM The main interface of the card.

[0104] The interface module 160 includes an external memory interface, a universal serial bus (USB) interface, and a subscriber identification module (SIM) card interface, etc. The external memory interface can be used to connect an external memory card, such as a MicroSD card, to expand the storage capacity of the mobile phone 10. The external memory card communicates with the processor 110 through the external memory interface to implement a data storage function.

[0105] In some embodiments, the mobile phone 10 further includes a button 101, a motor, and an indicator. The button 101 may include a volume button, a power on / off button, etc. The motor is used to make the mobile phone 10 vibrate, for example, vibrate when the user's mobile phone 10 is called to prompt the user to answer the call. The indicator may include a laser indicator, a radio frequency indicator, an LED indicator, etc.

[0106] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on the mobile phone 10. In other embodiments of the present application, the mobile phone 10 may include more or fewer components than shown in the figure, or combine some components, or separate some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0107] The software system of the electronic device can adopt a layered architecture, an event-driven architecture, a micro-core architecture, a micro-service architecture or a cloud architecture. TM Taking the system as an example, the software structure of the electronic device is exemplarily explained. Figure 4 Schematic diagram of the software structure of the electronic device of the present application embodiment. The layered architecture divides the software into several layers, each layer has a clear role and division of labor. The layers communicate with each other through software interfaces. In some embodiments, Android TM The software layers of the system are divided into the following from top to bottom: application layer, framework layer (FWK), library layer (FWK LIB), and kernel layer (Kernal). The application layer may include a series of application packages, such as desktop applications, payment applications, perception modules 41, business modules 42, presentation modules 43, and learning modules 44. When these application packages are running, they can access the various service modules provided by the application framework layer through the application programming interface (API) and execute corresponding intelligent services.

[0108] The desktop application can be used to realize the display of the desktop, such as the display of the application icon, the display of the desktop background, the display of the status bar, the display of the notification bar, etc. In addition, the desktop application can also be used to display the APP suggestion card according to the instruction of the presentation module 43, or cancel the card, etc. For example, the payment application can be Cloud QuickPass. TM , used to provide a QR code for boarding, so that users can scan the code to enter the subway station.

[0109] The perception module 41 runs in resident mode or in low-power mode, and has the ability to perceive external facts or environments. When the card reminder service is turned on, the perception module 41 monitors fences such as capabilities or matters (such as specific time, specific location, or specific event, etc.) registered with the business module 42. If the user triggers one of the capabilities or matters, the perception module 41 will send a notification to the business module 42. In addition, the perception module 41 can also detect related events and obtain the status of events from other applications in the application layer, or the application framework layer, or the system layer, or the kernel layer through the API, such as detecting Bluetooth connections, network connections, monitoring user text messages, customizing timers, etc.

[0110] Exemplarily, the function of the sensing module 41 includes sensing whether the user has entered a registered fence. For example, the sensing module 41 can detect whether the electronic device has entered the area of ​​the geo-fence of a certain location based on the mapping relationship between the geographical location detected by the electronic device, such as GPS signal, and the location of home, company, subway station, etc.

[0111] The business module 42, which can also be called a computing engine or a brain, has business logic processing capabilities and is used to implement the logic of displaying and disappearing various cards. For example, the business module 42 can query various recommendation factors and the user's preferred APPs under each recommendation factor from the learning module 44, and register the fences corresponding to each recommendation factor in the perception module 41. Furthermore, the business module 42 receives a notification sent by the perception module 41 indicating that something has been detected (such as a fence being triggered), and can send a command to the presentation module 43 (such as YOYO suggestion) according to the logic. TM ) to display the card. The command may carry a link to a third-party APP or mini program. When a user clicks on an APP in the card, the user will be redirected to the corresponding third-party APP or mini program according to the link. In addition, the business logic processing module 42 may also control the cancellation of the display of the card by receiving a notification sent by the context perception module 41 indicating that the user has used the APP in the APP suggestion card superimposed on the current interface.

[0112] For the convenience of description, in this application, the APP that the user prefers to click to use is also referred to as the preferred APP, the user's preferred APP, etc.

[0113] Business presentation module 43, also known as YOYO suggestion TM, used to display or disappear the determined card on the screen of the electronic device. For example, the business presentation module 43 can receive the command to display the reminder card transmitted by the business logic processing module 42, and display the reminder card to the user. When the user clicks on an APP or mini program in the card, it will link to a page of the APP or mini program. In some embodiments, the user clicks on the Cloud QuickPass in the APP suggestion card. TM The APP icon will display the QR code page used by the user last time, so that the user can directly scan the code to enter and exit the subway station. Correspondingly, if the business presentation module 43 receives the card disappearance command sent by the business logic processing module 42, the business presentation module 43 will immediately disappear the displayed card.

[0114] The learning module 44 is used to analyze the user's historical application usage information and pre-process the information. Furthermore, the user's preferred APPs within a period of time after the fence corresponding to each recommendation factor is triggered can be mined and learned. Specifically, the learning module 44 can determine the preferred APPs within a period of time (such as 5 minutes) after the fence corresponding to each recommendation factor is triggered and save them as the first portrait table, thereby providing the business logic processing module 42 with the user's preferred APPs within a period of time after the fence is triggered.

[0115] In addition, the learning module 44 can also mine and learn the APPs that users prefer to click and use under various recommendation factors. Then, the learning module 44 can determine the preferred APPs under various recommendation factors (such as 5 minutes after entering the corresponding fence) and save them as the second portrait table, thereby providing the business module 42 with the preferred APPs under various recommendation factors.

[0116] The framework layer provides APIs and programming frameworks for applications in the application layer. The application framework layer includes some predefined functions. For example, the framework layer may include window managers and perception management services.

[0117] Among them, the window manager can be used to manage window programs. The window manager can obtain the size of the display screen, determine whether there is a status bar, lock the screen, capture the screen, etc. The perception management service can be used to provide the application layer with APIs related to the processing functions of time information and location information, such as obtaining the time information and location information of the electronic device, and provide the context perception module with APIs related to the processing functions of geo-fences, such as determining whether the electronic device enters or leaves the geo-fence. The library layer includes the system library and the Android runtime. The system library can include multiple functional modules, such as the surface manager. The Android runtime is responsible for the scheduling and management of the Android system, including the core library and the virtual machine. The core library contains two parts: one part is the function function that needs to be called by the Java language, and the other part is the core library of Android. The application layer and the application framework layer run in the virtual machine. The virtual machine executes the Java files of the application layer and the application framework layer as binary files. The virtual machine is used to perform object life cycle management, stack management, thread management, security and exception management, and garbage collection.

[0118] The kernel layer is the interface layer between hardware and software. The kernel layer may include mobile communication drivers, Wi-Fi drivers, GNSS drivers, display drivers, and sensor drivers.

[0119] It should be noted that although the embodiment of the present application is based on Android TM The system is used as an example, but the basic principle is also applicable to iOS-based TM or Windows TM Electronic devices with other operating systems.

[0120] The executor of the card display method provided in the embodiment of the present application may be the above-mentioned electronic device, or may be a functional module and / or functional entity (such as a processor) in the electronic device that can implement the card display method, and the present application scheme can be implemented by hardware and / or software, which is not limited in the present application.

[0121] Combination Figure 5 ,by Figure 4 Some functional modules in the electronic device are shown as execution entities, and the application recommendation method provided in the embodiment of the present application is illustrated.

[0122] S501: The presentation module 43 sends a card activation message to the service module 42.

[0123] In some embodiments, the card startup message may be a message sent by the presentation module 43 to the business module 42 when the electronic device is turned on, restarted, the system is updated, the YOYO suggestion is updated, or the learning task is updated, and is used to instruct the business module 42 to subscribe to one or more learning tasks.

[0124] The above learning tasks may be learning tasks added or updated by users or software developers in the setting options of YOYO Suggestions. YOYO Suggestions usually supports one or more learning tasks.

[0125] In some embodiments, the learning task in the present application may be to learn the APPs preferred by the user within a period of time after the fences corresponding to the various recommendation factors are triggered, which is recorded as learning task 1 or the first learning task.

[0126] In some embodiments, the learning task in the present application can be to learn the APPs preferred by the user under various recommendation factors, where the electronic device has entered the fence corresponding to the recommendation factor for a period of time, such as having entered the fence for 5 minutes or having used N applications after entering the fence), such as a geographic fence. Among them, this learning task can be recorded as learning task 2 or the second learning task.

[0127] S502: The business module 42 subscribes to a learning task from the learning module 44. For example, the subscribed learning tasks include learning task 1 and learning task 2.

[0128] Of course, in some other embodiments, the learning task subscribed by the business module 42 in the present application may also be one of the first learning task and the second learning task.

[0129] S503: The learning module 44 pre-processes the historical data when the screen is turned off during charging to obtain a plurality of historical application usage information, and mines and learns the plurality of historical application usage information to obtain various first portrait tables and various second portrait tables.

[0130] Among them, the learning module 44 mines and learns the user's preferred APPs within a period of time after each fence is triggered according to multiple historical application usage information to obtain each first portrait table, and mines and learns the user's preferred APPs under the recommendation factors corresponding to each fence according to multiple historical application usage information to obtain each second portrait table.

[0131] It can be understood that the first portrait table is used to save the user's preferred APPs within a period of time after each fence is triggered, and the second portrait table is used to save the user's preferred APPs under the recommendation factors corresponding to each fence.

[0132] In some embodiments, the recommendation factors may be determined based on a plurality of historical application usage information within the historical time period, such as several recommendation factors that appear most frequently within the historical time period.

[0133] In some embodiments, the state in which the learning module 44 performs the mining learning process is not limited to the above-mentioned charging and screen-off state, but can also be the standby state of other electronic devices. This application does not make specific limitations on this.

[0134] In some embodiments, the above historical data may include user APP click records and scene trigger records pre-acquired by the electronic device. In addition, the above multiple historical application usage information may be obtained by integrating the historical data.

[0135] In some embodiments, each piece of historical application usage information may include an APP information, APP click time (i.e., the time information when the APP was used), and recommendation factor information (i.e., the recommendation factor of the electronic device when the APP was used). That is, a piece of historical application usage information is used to indicate that a user clicked and used an APP under a recommendation factor at a historical time. For example, a piece of historical application usage information (recorded as information 1) is used to indicate that a user clicked and used Cloud QuickPass in a subway scenario at 13:32:20 on May 15, 2023. TM APP. At this time, the APP information in information 1 is "YunQuanFu TM ", the APP click time is "May 15, 2023 13:32:20", and the recommendation factor information is "subway scene".

[0136] The above user APP click record may include multiple APP click records, each APP click record includes APP information and APP click time, wherein the APP information may be the APP name, etc. The above scene trigger record may include multiple scene trigger records, each scene trigger record may include the identification information trigger time of the scene where the electronic device is located, wherein the identification information of the recommendation factor may be the geographic location, GPS information, whether to wear headphones, etc. For example, the identification of the above recommendation factor may indicate: home, company, hospital, shopping mall, airport, railway station, highway service area, government agency, subway, wearing headphones, car Bluetooth, etc.

[0137] In some embodiments, the above historical data may be data from a recent period of time, for example, the above historical data may be data from the recent month or multiple months, that is, the above multiple historical application usage information may reflect the APP usage of the user under each recommendation factor in the recent month or multiple months. It is understandable that the above multiple historical application usage information may reflect the APP usage when the user uses the electronic device to switch to each recommendation factor in the recent month or multiple months, that is, the APP usage when the user uses the electronic device to enter the fence corresponding to each recommendation factor.

[0138] In some embodiments, the multiple historical application usage information acquired by the electronic device may be in the form of a data collection table, and the historical application usage information under different types of recommendation factors may be stored in different data collection tables. For example, the historical application usage information under recommendation factors corresponding to scenes in fixed locations such as home and company may be stored in the same data collection table, the historical application usage information under recommendation factors corresponding to mobile scenes such as buses and subways may be stored in the same data collection table, and the historical application usage information under recommendation factors related to the device such as wearing headphones and car Bluetooth may be stored in the same data collection table.

[0139] In some embodiments, the mining learning process of the first learning task can determine the preference probability of the APP within a period of time after the fence corresponding to each recommendation factor is triggered, and save the APP with a preference probability greater than the preference probability threshold to the first portrait table.

[0140] In some embodiments, the mining learning process of the above-mentioned first learning task can determine the trigger confidence of the APP within a period of time after the fence corresponding to each recommendation factor is triggered, and save the APP whose trigger confidence is greater than the preset trigger confidence threshold to the first portrait table.

[0141] In some embodiments, the mining learning process of the second learning task can determine the promotion degree of the APP under each recommendation factor, and save the APPs with a promotion degree greater than or equal to a threshold value under each recommendation factor to the second portrait table.

[0142] In some embodiments, the mining learning process of the above-mentioned second learning task can determine the scene confidence of the APP after the fence corresponding to each recommendation factor is triggered for a period of time, and save the APP whose scene confidence is greater than a preset scene confidence threshold to the second portrait table.

[0143] Reference Fig. 6AAs shown, it is a flow chart of APP preference mining within a period of time after each fence is triggered in the embodiment of the present application. Specifically, the electronic device can obtain the user APP click record and the scene trigger record, and pre-process these data to obtain ① the APP behavior under each recommendation factor, and ② the APP behavior under the current calculation cycle. For example, the current fence can be the subway fence for the electronic device to enter the subway, and the current calculation cycle can be one month before the current system time. The APP behavior refers to the behavior of the user clicking the APP icon to open the APP, that is, the behavior of the user using the APP. Furthermore, ③ the electronic device can calculate the APP promotion degree (denoted as lift) under each recommendation factor, and ④ calculate the hit probability (denoted as topKprop) and preference probability (denoted as hitprop) of each APP within a period of time after the fence corresponding to each recommendation factor is triggered, and calculate the confidence of each APP (also called fence trigger confidence, denoted as conf 1). Furthermore, the electronic device can set the trigger confidence threshold of the APP suggestion card, and save the APP that is greater than the trigger confidence threshold within a period of time after each fence is triggered to the first portrait table. It can be understood that the fence trigger confidence threshold is also called the first confidence threshold, and the value of the threshold can be determined according to actual needs, and this application does not make any specific limitations on this.

[0144] Among them, the process of APP preference mining within a period of time after each fence is triggered in this application will be combined below Fig. 7A A detailed description is given and no further elaboration is given here.

[0145] Reference Figure 6BAs shown, it is a flow chart of APP preference mining under various recommendation factors provided in the embodiment of the present application. Specifically, the electronic device can obtain the user APP click record and the scene trigger record, and pre-process these data to obtain ① APP behavior under each recommendation factor, and ② APP behavior under the current calculation cycle. For example, the current recommendation factor can be within a day of the current system time of the electronic device, and the current calculation cycle can be a month before the current system time. And APP behavior refers to the behavior of the user clicking the APP icon to open the APP, that is, the behavior of the user using the APP. Furthermore, the electronic device can ③ calculate the APP promotion degree (i.e., lift) under each recommendation factor, and ③ calculate the weighted probability of the number of times each APP is used and the number of days of occurrence under each recommendation factor to obtain the confidence (recorded as conf 2). Furthermore, the electronic device can set the card lift threshold and conf 2 threshold of the APP suggestion card, and save the APP greater than these thresholds under each recommendation factor as the second portrait table. It can be understood that the lift threshold is also called the promotion threshold, the conf 2 threshold is also called the scene confidence threshold, and the values ​​of these two thresholds can be determined according to actual needs, and this application does not make specific restrictions on this. For example, the lift threshold can be 0.6 and the threshold for conf 2 can be 0.4.

[0146] Among them, the process of APP preference mining under the recommendation factors corresponding to each fence in this application will be combined below Fig. 8A A detailed description is given and no further elaboration is given here.

[0147] S504 : the business module 42 queries each first portrait table and each second portrait table from the learning module 44 .

[0148] Among them, the first portrait table includes preferred apps within a period of time after a fence is triggered, and a first portrait table includes app information, recommendation factor information, preference probability or trigger confidence of each preferred app. And, a second portrait table includes preferred apps under a recommendation factor, and a second portrait table includes app information, recommendation factor information, promotion degree, weighted probability or scenario confidence of each preferred app.

[0149] S505: The business module 42 registers fences associated with various recommendation factors with the perception module 41, wherein each fence is associated with a first portrait table to which preferred APPs belong within a period of time after the fence is triggered and a second portrait table to which preferred APPs belong under the corresponding recommendation factor.

[0150] S506: The business module 42 monitors the task running status.

[0151] It can be understood that the business module 42 can monitor whether the current desktop interface is running, or monitor whether a certain APP interface is opened, etc.

[0152] S507: The perception module 41 sends a first fence trigger notification to the business module 42.

[0153] Among them, the perception module 41 can detect in real time whether a certain fence is triggered, and notify the business module 42 after a fence is triggered.

[0154] The first fence is a fence registered in the sensing module 41. It can be understood that a fence being triggered indicates that the electronic device enters the recommended factor corresponding to the fence.

[0155] S508: The business module 42 notifies the presentation module 43 to display the APP suggestion card where the user's preferred APP is located within a period of time after the triggering of the first fence according to the first portrait table corresponding to the current fence, that is, to issue a card. TM , APP suggestion card is YOYO suggestion TM card.

[0156] In some embodiments, the service module 42 may notify the presentation module 43 so that the presentation module 43 displays the YOYO suggestions in the target area when switching to the main interface including the target area. TM The card shows the preferred apps within a period of time after the current fence is triggered.

[0157] In other embodiments, the service module 42 may notify the presentation module 43 to overlay the YOYO icon on any interface displayed at the current fence triggering moment. TM Suggestion card to show the YOYO suggestion TM The preferred apps contained in the card.

[0158] S509: The business module 42 determines that the electronic device is under the recommendation factor corresponding to the first fence.

[0159] Specifically, the business module 42 can determine that the triggering time for the electronic device to enter the first fence is greater than the first time (such as 5 minutes), or the triggering time is within the first time and the APP in the APP suggestion card is clicked and used, to determine that the electronic device is in the scene corresponding to the first fence. At this time, the APP in the APP suggestion card is the APP in the first portrait table.

[0160] Alternatively, the business module 42 may determine that the electronic device has used N APPs after entering the first fence, or the number of APPs used after entering the first fence is less than N and the APP in the APP suggestion card is clicked and used, to determine that the electronic device is under the recommendation factor corresponding to the first fence. At this time, the APP in the APP suggestion card is the APP in the first portrait table.

[0161] S510: The business module 42 notifies the presentation module 43 to display the APP suggestion card where the user's preferred APP is located under the recommendation factor corresponding to the first fence according to the second portrait table corresponding to the first fence.

[0162] In this way, when the electronic device just enters the fence corresponding to a certain recommendation factor, the electronic device can display the user's preferred APPs within a period of time (such as the first 5 minutes) after the fence is triggered in the YOYO suggestion card, which is convenient for users to quickly use these APPs and improve the user experience. In addition, as the user moves and time passes, after the electronic device enters the fence for a period of time (such as 5 minutes), the user uses the electronic device under the recommendation factor, and the electronic device can display the user's preferred APPs under the recommendation factor in the YOYO suggestion card, which is convenient for users to quickly use these APPs and improve the user experience. Thus, the accuracy of application recommendations is improved.

[0163] Next, combine Fig. 7A , with electronic equipment as the execution subject, Figure 5 Based on the method shown, the process of learning the first task in the application recommendation method provided by the application embodiment and recommending the application in the first portrait table by the APP suggestion card is described in detail. Specifically, the method includes the following steps:

[0164] S701: The electronic device obtains a plurality of historical application usage information within a historical time period. For example, each piece of historical application usage information may include APP information, APP click time, and recommendation factor information.

[0165] For example, the multiple historical application usage information can be obtained by integrating the historical data in the last month, so the multiple historical application usage information represents the situation of users clicking on the APP under various recommendation factors in the last month. Figure 5 The relevant description in S503 is not repeated here.

[0166] S702: The electronic device selects historical application usage information corresponding to multiple recommendation factors from multiple historical application usage information.

[0167] In some embodiments, the multiple recommendation factors in the present application can be pre-set in the electronic device by the user or relevant staff. For example, the above recommendation factors can be home, company, hospital, shopping mall, airport, railway station, high-speed service area, subway, wearing headphones, car Bluetooth, etc.

[0168] S703: The electronic device calculates the number of days that each recommendation factor appears based on the historical application usage information corresponding to the multiple recommendation factors.

[0169] The number of days that each recommendation factor appears can be recorded as Count(s, day), the parameter s represents the recommendation factor and different values ​​represent different recommendation factors, and day represents the number of days that the recommendation factor indicated by the parameter s appears. For example, multiple recommendation factors are represented by s1, s2, ..., sn-1 and sn respectively, that is, the values ​​of the parameter s in Count(s, day) are s1, s2, ..., sn-1 and sn respectively, and n is a positive integer.

[0170] S704: The electronic device filters out historical application usage information corresponding to recommendation factors whose appearance days are less than a minimum number of days, and obtains valid historical application usage information corresponding to multiple recommendation factors.

[0171] Among them, the value of the above-mentioned minimum number of days can be selected according to actual needs. For example, the minimum number of days can be 2 days, but it is not limited to this. It can be understood that since the historical application usage information under the recommendation factor with a number of days less than the minimum number of days may be a recommendation factor that the user is in by chance, it cannot reflect the user's APP usage preference under the recommendation factor. Therefore, the relevant data of the recommendation factor is not considered, and there is no need to recommend the user's preferred APP for the recommendation factor in the future. Among them, in the embodiment of the present application, the user's APP usage preference can be an APP click preference, that is, the user's preference to click the APP icon to open the APP.

[0172] S705: The electronic device calculates the usage count of the APP under all the recommendation factors according to the plurality of historical application usage information.

[0173] The number of times the APP is used under all recommendation factors can be recorded as Count(all). It can be understood that the number of times the APP is used in the embodiment of the present application can be the number of times the APP is used.

[0174] S706: The electronic device calculates the number of times the APP is used under each recommendation factor according to the valid historical application usage information corresponding to the multiple recommendation factors.

[0175] The number of times the APP is used under each recommendation factor can be recorded as Count(s). For example, the values ​​of the parameter s in Count(s) are s1, s2, ..., sn-1 and sn in sequence.

[0176] S707: The electronic device calculates the usage times of each APP under all recommendation factors according to the plurality of historical application usage information.

[0177] The number of times each APP is used under all recommendation factors can be recorded as Count(app), and the parameter app represents an application program and different values ​​represent different applications. For example, the APPs in an electronic device can be recorded as app1, app2, ..., app m-1, and app m, that is, the values ​​of the parameter app in Count(app) are app1, app2, ..., app m-1, and app m, respectively, and m is a positive integer.

[0178] S708: The electronic device calculates the number of times each APP is used under each recommendation factor based on the valid historical application usage information corresponding to the multiple recommendation factors.

[0179] The number of times each APP is used under each recommendation factor is recorded as Count(app|s). For example, the values ​​of the parameter s in Count(app|s) are s1, s2, ..., sn-1 and sn, and the values ​​of the parameter app are app1, app2, ..., app m-1 and app m. It can be understood that the number of times each APP is used under each recommendation factor, Count(app|s), is the frequency of use of each APP under each recommendation factor.

[0180] S709: The electronic device calculates the promotion degree of each APP under each recommendation factor based on the number of times the APP is used under all recommendation factors in the historical time period, the number of times the APP is used under each recommendation factor, the number of times each APP is used under all recommendation factors Count(app), and the number of times each APP is used under each recommendation factor Count(app|s), and calculates the first evaluation parameter of each APP within the first time period after each fence is triggered or during the use of the first N applications based on the promotion degree of each APP.

[0181] The first duration mentioned above is the first 5 minutes after each fence is triggered.

[0182] For example, the first evaluation parameter may be a preference probability or a trigger confidence.

[0183] The process of calculating the promotion degree, preference probability and trigger confidence of each APP in S709 will be combined below. Fig. 9 The specific description will not be repeated here.

[0184] S710: The electronic device saves, as a first portrait table, APPs whose first evaluation parameter is greater than the first evaluation parameter threshold within a first time period after each fence is triggered or during the use of the first N applications.

[0185] For example, the first evaluation parameter is the preference probability, and the first evaluation parameter threshold is a set preference probability threshold.

[0186] For example, the first evaluation parameter is the scene confidence, and the first evaluation parameter threshold is a set scene confidence threshold.

[0187] It can be understood that the first evaluation parameter of an APP under a recommendation factor is used to reflect the association between the recommendation factor and the user clicking to use the APP, that is, to reflect the user's preference for clicking to use the APP under the recommendation factor.

[0188] In some embodiments, each first portrait in the above-mentioned first portrait table includes APP information, recommendation factor information, preference probability, and trigger confidence, that is, the above-mentioned first portrait table can reflect the APP that the user prefers to click and use within a period of time after the triggering of the fence corresponding to each recommendation factor.

[0189] S711: The electronic device detects that the first fence is triggered, and displays an APP suggestion card corresponding to the first fence according to the corresponding first portrait table. The APP suggestion card includes the user's preferred APPs within the first time period after the first fence is triggered or during the use of the first N applications.

[0190] In some embodiments, in the present application, the electronic device may pre-register fences corresponding to various recommendation factors, and the first fence corresponds to one recommendation factor.

[0191] It can be understood that the above-mentioned first fence can be a geographical fence, a time fence, etc., and this application will not go into details.

[0192] In some embodiments, when the electronic device enters the first fence, the electronic device can recommend to the user through an APP suggestion card the user's preferred APPs within a period of time after the first fence is triggered.

[0193] In some embodiments, the number of APPs in the above-mentioned APP suggestion card is less than or equal to a preset number, for example, 4, but not limited thereto, and the value may also be 3 or 5.

[0194] It can be understood that when the number of APPs whose first evaluation parameter is greater than the first evaluation parameter threshold is greater than or equal to the preset number within a period of time after a fence is triggered, the electronic device will recommend the preset number of APPs among these APPs in the APP suggestion card. However, when the number of APPs whose first evaluation parameter is greater than the first evaluation parameter threshold is less than the preset number within a period of time after a fence is triggered, the electronic device will recommend these APPs in the APP suggestion card.

[0195] For example, combined with Figure 2 , the first fence may be geo-fence 01, then the electronic device triggers geo-fence 01, indicating that the electronic device enters the subway scene, refer to Figure 7B As shown, the YOYO suggestion card 05a displayed by the electronic device may include the user's preference to click to use UnionPay when entering the subway station. TM , Alipay TM ,WeChat TM , Sina Weibo TM Understandably, when users enter a subway station, they may use Alipay TM or WeChat TM When entering the subway fence, they also prefer to use these Alipay TM ,WeChat TM .

[0196] In some embodiments, the order of the multiple preferred APPs displayed in the YOYO suggestion card in this application may not be limited. TM , Alipay TM ,WeChat TM , Sina Weibo TM The display order is not limited to Figure 7B The display order shown may also be any other achievable order.

[0197] In this way, the application recommendation method provided by this application not only considers the overall usage frequency of the APP, but also considers the user's preference for using the APP when entering different recommendation factors. Therefore, the user's preferred APP can be accurately recommended to the user when the user just enters different recommendation factors, which is convenient for the user to quickly use the required APP, greatly improving the user experience.

[0198] Next, combine Fig. 8A , with electronic equipment as the execution subject, Figure 5 Based on the method shown, the second task learning in the application recommendation method provided by the application embodiment and the process of the APP suggestion card recommending the application in the second portrait table are described in detail. Specifically, the method includes the following steps:

[0199] S801-S807. Among them, the S801-S807 and Fig. 7A S701-S707 shown are the same and will not be described again here.

[0200] S808: The electronic device calculates the number of times each APP is used under each recommendation factor and the number of days each APP is used under each recommendation factor based on the valid historical application usage information corresponding to the multiple recommendation factors. For example, the electronic device calculates the number of times and days of use of APP1 under recommendation factor s1 within T1.

[0201] The number of times each APP is used under each recommendation factor is recorded as Count(app|s), and the number of days each APP is used under each recommendation factor is recorded as Count(app|s,day). For example, the values ​​of the parameter s in Count(app|s) and Count(app|s,day) are s1, s2, ..., sn-1, and sn, respectively, and the values ​​of the parameter app are app1, app2, ..., app m-1, and app m, respectively. It can be understood that compared to Fig. 7A In S708 and S808, the number of days each APP is used under each recommendation factor is calculated.

[0202] S809: The electronic device calculates the improvement degree and the second evaluation parameter of each APP under each recommendation factor based on the number of days each recommendation factor is used Count (s, day), the number of times the APP is used under all recommendation factors, the number of times the APP is used under each recommendation factor, the number of times each APP is used under all recommendation factors Count (app), the number of times each APP is used under each recommendation factor Count (app|s), and the number of days each APP is used under each recommendation factor Count (app|s, day).

[0203] In one embodiment, the second evaluation parameter of each APP under the above-mentioned recommendation factors is a weighted probability.

[0204] In another embodiment, the second evaluation parameter of each APP under the above-mentioned recommendation factors is a scene confidence, wherein the scene confidence is calculated based on a weighted probability of the APP and the total number of APPs in the electronic device.

[0205] The process of calculating the promotion degree, weighted probability and scene confidence of each APP in S809 will be combined below. Fig.10 The specific description will not be repeated here.

[0206] S810: The electronic device saves the APPs whose promotion degree under each recommendation factor is greater than the promotion degree threshold and whose second evaluation parameter is greater than the second evaluation parameter threshold to the second portrait table.

[0207] In some embodiments, the second evaluation parameter is a weighted probability, and the above evaluation threshold is a threshold corresponding to the weighted probability of each APP under each recommendation factor, for example, the threshold is 0.5.

[0208] In some embodiments, the second evaluation parameter is the scene confidence, and the evaluation threshold is the scene confidence threshold corresponding to the scene confidence of each APP under each recommendation factor, for example, the threshold is 0.4, and the threshold can be recorded as the second confidence threshold. In addition, the lift threshold can be 0.6.

[0209] It can be understood that the scenario confidence of an APP is specifically determined by the number of times and days of use clicked by users under a recommendation factor and the total number of APPs in the electronic device, and the greater the number of times and days of use, the higher the confidence.

[0210] In some embodiments, each second portrait in the above-mentioned second portrait table includes APP information, recommendation factor information, improvement degree, and scene confidence degree, that is, the above-mentioned second portrait table can reflect the APP that the user prefers to use under each recommendation factor.

[0211] S811: The electronic device detects that the duration of triggering the first fence is greater than the first duration, or the triggering duration is within the first duration and the APP in the current APP suggestion card is clicked and used, and displays the APP suggestion card corresponding to the first fence according to the second portrait table. The APP suggestion card includes the user's preferred APPs under the recommendation factors corresponding to the first fence.

[0212] In some embodiments, after the electronic device enters the first fence and is under the corresponding recommendation factor, the electronic device can continue to recommend to the user the user's preferred APPs under the recommendation factor corresponding to the first fence through APP suggestion cards.

[0213] For example, combined with Figure 2 , the first fence may be geo-fence 01, then the electronic device triggers and enters geo-fence 01, indicating that the electronic device has entered the subway scene, refer to Figure 8B As shown, the YOYO suggestion card 05b displayed by the electronic device may include the user's preference to use UnionPay when taking the subway. TM , Sina Weibo TM , Toutiao TM Reading with WeChat TM .in, Figure 7B The YOYO suggestion card 05a shown shows the Cloud QuickPass TM, Alipay TM ,WeChat TM and Sina Weibo TM ,and Figure 8B The YOYO suggestion card 05b shown shows the Cloud QuickPass TM , Sina Weibo TM , Toutiao TM Reading with WeChat TM Not exactly the same. It is understandable that the APP that a user prefers to use when entering a recommendation factor using an electronic device may be different from the APP that the user prefers to use under the recommendation factor. For example, a user prefers to use entertainment information APPs in the subway scene, such as Today's Headlines. TM Reading with WeChat TM Etc. Similarly, the display order of the APPs in the YOYO suggestion card 05 can also be any other achievable order.

[0214] In this way, the application recommendation method provided by this application, for different recommendation factors, APP recommendations not only consider the overall usage frequency of the APP, but also consider the user's preference for using the APP under different recommendation factors, so that the user's preferred APP can be accurately recommended under different recommendation factors, which is convenient for users to quickly use the required APP, greatly improving the user experience.

[0215] Combination Fig. 9 , the process of calculating the promotion degree, preference probability and trigger confidence of each APP by the electronic device in the embodiment of the present application is described in detail. Fig. 9 As shown, Fig. 7A The illustrated S709 may include S901 to S907:

[0216] S901: The electronic device calculates the usage probability of each APP under each recommendation factor based on the usage count Count(app|s) of each APP under each recommendation factor and the usage count Count(s) of each APP under each recommendation factor in the historical time period.

[0217] For example, the usage probability of each APP under each recommendation factor can be recorded as p(app|s).

[0218] In some embodiments, the electronic device may use the number of times Count(app|s) of use of each APP under each recommendation factor and the number of times Count(s) of use of the APP under each recommendation factor as the usage probability of each APP under each recommendation factor, that is, the usage probability of each APP under each recommendation factor may be implemented by the following formula (1-1):

[0219] p(app|s)=Count(app|s) / Count(s)(1-1)

[0220] It can be understood that different values ​​of parameter s in formula (1-1) represent different recommendation factors, and different values ​​of parameter app represent different applications. Then, when parameter s is fixed, the value of parameter app can be modified to calculate the usage probability of each APP under the current recommendation factor. Similarly, in formula (1-1), parameter s is set to another value to determine the usage probability of each APP under other recommendation factors. For example, when parameter s represents a subway scene, the value of parameter app can be modified to calculate the usage probability of each APP in the subway scene.

[0221] It is understandable that if the amount of historical application usage information under a certain recommendation factor in the historical time period is too small, then the number of APP usage times Count(s) under each recommendation factor is too small, resulting in inaccurate calculated usage probabilities of each APP under each recommendation factor. In some embodiments, in order to improve the occurrence of this situation, the present application can calculate the Wilson probability of each APP under each recommendation factor, that is, correct the usage probability of each APP under each recommendation factor, such as by normalizing.

[0222] In some implementations, the electronic device may modify the ratio of the number of uses Count(app|s) under each recommendation factor to the number of uses Count(s) of the APP under each recommendation factor to calculate the usage probability of each APP under each recommendation factor. At this time, the usage probability of each APP under each recommendation factor is the Wilson usage probability.

[0223] In some embodiments, the probability of Wilson's use can be recorded as willson ctr (app|s), that is, willson ctr (app|s) = willson-ctr(count(s), count(app|s)). Wilson probability willson ctr (app|s) can be achieved by combining formula (1-1) and the following formula (1-2).

[0224]

[0225] p=Count(app|s) / Count(s) (1-3)

[0226] Where n in formula (1-2) is the number of times the APP is used under each recommendation factor, that is, the number of samples, and z is a constant, for example, z = 1.96. In addition, the calculation of formula (1-3) is the same as that of formula (1-1), that is, p in formula (1-3) refers to the uncorrected usage probability of each APP under each recommendation factor.

[0227] It can be understood that as the values ​​of parameter s and parameter app change, the Wilson usage probability of each APP under each recommendation factor can be calculated by formulas (1-1) and (1-2).

[0228] S902: The electronic device calculates the usage probability of the APP under all recommendation factors according to the usage count Count(app) of each APP under all recommendation factors and the usage count Count(all) of the APP under all recommendation factors.

[0229] Among them, the probability of using APP under all recommendation factors can be recorded as P(app|all).

[0230] In some embodiments, the electronic device may use the ratio of the number of times each APP is used under all recommendation factors Count(app) to the number of times each APP is used under all recommendation factors Count(all) as the usage probability of each APP under each recommendation factor. At this time, the usage probability of APP under all recommendation factors P(app|all) can be realized by the following formula (2-1):

[0231] p(app|all)=count(app) / count(all) (2-1)

[0232] It is understandable that in order to avoid an excessive count(app) value leading to an inaccurate result of p(app|all), count(app) can be corrected to appropriately increase the value of count(app). In some embodiments, the present application can use the following formula (2-2) to calculate the usage probability P(app|all) of the APP under all recommendation factors.

[0233]

[0234] Among them, in formula (2-2) is the penalty factor, For example, to prevent the value of count(app) from being too small, which will result in the calculation result of p(app|all) being too small, you can set The value is 5, but not limited to this.

[0235] S903: The electronic device calculates the promotion degree of each APP under each recommendation factor according to the usage probability p(app|s) of each APP under each recommendation factor and the usage probability p(app|all) of the APP under all recommendation factors.

[0236] Among them, the improvement degree of each APP under each recommendation factor can be recorded as

[0237] In some embodiments, the improvement degree of each APP under each recommendation factor can be the ratio of the usage probability p(app|s) of each APP under each recommendation factor to the usage probability p(app|all) of the APP under all recommendation factors, which can be achieved through the following formula (3-1).

[0238]

[0239] In some embodiments, in order to facilitate comparison of the promotion degrees of different apps in actual applications, the promotion degrees may be normalized. For example, the calculation formula for the promotion degree of each app under each recommendation factor may be replaced by the following formula (3-2) instead of the above formula (3-1).

[0240]

[0241] Among them, sigmoid() in formula (4-2) refers to The value is normalized to between 0 and 1.

[0242] In some embodiments, the usage probability corresponding to each APP under each recommendation factor is the Wilson probability, and the calculation formula for the promotion degree of each APP under each recommendation factor can be replaced by the above formula (3-2) with the following formula (3-3).

[0243]

[0244] S904: The electronic device selects the first N APPs clicked and used by the user or the APPs used in the first time period in each continuous historical application usage information according to the multiple continuous historical application usage information under each recommendation factor in the historical time period as an APP sequence to obtain multiple APP sequences corresponding to each recommendation factor, where N is a positive integer.

[0245] It can be understood that each APP sequence is a sequence consisting of the first N APPs in multiple consecutive historical application usage information under one recommendation factor.

[0246] In some embodiments, each continuous segment of historical application usage information is composed of a plurality of pieces of historical application usage information with continuous APP click times contained within a day.

[0247] In some embodiments, N is a preset value such as 5, indicating that the user clicks on the first 5 apps used when entering a fence.

[0248] In some embodiments, the first duration is 5 minutes, and a sequence includes applications that have been clicked and used within the first duration after the user enters a recommendation factor.

[0249] S905: The electronic device determines the hit probability of each APP in the multiple APP sequences corresponding to each recommendation factor according to the number of occurrences of each APP in the multiple APP sequences corresponding to each recommendation factor and the number of the multiple APP sequences corresponding to each recommendation factor.

[0250] For example, the hit probability of each APP in multiple APP sequences corresponding to each recommendation factor can be recorded as topK prop (app|s), the number of occurrences of each recommendation factor in multiple APP sequences can be recorded as count(s|appin topN), and the number of multiple APP sequences corresponding to each recommendation factor can be recorded as M(s). prop The values ​​of the parameter s in (app|s), count(s|app in topN), and M(s) are s1, s2, ..., sn-1, and sn, respectively, indicating different recommendation factors. In addition, the values ​​of the parameter app in count(s|app in topN) are app1, app2, ..., app m-1, and app m, respectively, indicating different applications.

[0251] In some embodiments, the hit probability of each APP in the multiple APP sequences corresponding to each recommendation factor can be the ratio of the number of appearances of each APP in the multiple APP sequences corresponding to each recommendation factor to the number of multiple APP sequences corresponding to the corresponding recommendation factor. For example, it can be implemented using the following formula (4).

[0252] topK prop (app|s)=count(s|app in topN) / M(s) (4)

[0253] S906: The electronic device calculates the preference probability of each APP under each recommendation factor according to the hit probability of each APP in the multiple APP sequences corresponding to each recommendation factor and the promotion degree of each APP under each recommendation factor.

[0254] In some embodiments, the electronic device first uses the correction coefficient 2 to correct the improvement degree of each APP under each recommendation factor to obtain the corresponding correction coefficient 1, and then uses the correction coefficient 1 to multiply the hit probability of the corresponding APP in multiple APP sequences corresponding to the corresponding recommendation factor to obtain the preference probability of the APP under the recommendation factor.

[0255] For example, the preference probability of each APP under each recommendation factor can be recorded as hit prop It can be understood that the preference probability of APP1 under the recommendation factor s1 refers to the probability of APP1 being clicked and used when the electronic device just enters the recommendation factor s1, that is, enters the fence of the recommendation factor s1.

[0256] In some embodiments, the preference probability of each APP in a plurality of APP sequences corresponding to each recommendation factor is implemented using the following formula (5).

[0257] hit prop (app|s)=exp(lift-0.5)*topK prop (app|s) (5)

[0258] Among them, 0.5 in formula (5) can be used as the correction coefficient 2. At this time, lift can be the lift calculated according to formula (3-2) or (3-3), that is, the value range of the lift is a value between 0 and 1. At this time, exp(lift-0.5) is the correction coefficient 1. Then, when lift is greater than 0.5, the correction coefficient 1 is a value greater than 1, and the correction coefficient 1 can increase the corresponding hit probability topK prop (app|s) gets a larger preference probability hit prop (app|s), on the contrary, when lift is greater than 0.5, the correction factor 1 can reduce the corresponding hit probability topK prop (app|s) gets a smaller preference probability hit prop (app|s).

[0259] S907: The electronic device sorts the APPs according to the preference probability of each recommendation factor from small to large, and uses the ratio of the ranking of each APP under each recommendation factor to the total number of APPs in the electronic device as the trigger confidence of each APP under the recommendation factor.

[0260] In some embodiments, the electronic device can sort the preferences of various APPs under a recommendation factor from small to large. For example, for the recommendation factor of subway scene, the APPs clicked by the user under this recommendation factor are WeChat Reading, TM , Cloud QuickPass TM , Sina Weibo TM , ToutiaoTM , Taobao TM ,WeChat TM , Meitu Xiuxiu TM , Alipay TM , Gallery, Music, a total of 10 APPs. And the order of preference probability of these 10 APPs from small to large is Taobao TM , Meitu Xiuxiu TM , Gallery, Music, WeChat Reading TM , Toutiao TM , Sina Weibo TM ,WeChat TM , Alipay TM , Cloud QuickPass TM , that is, the rankings of these apps are 1 to 10. Then, Taobao TM , Meitu Xiuxiu TM , Gallery, Music, WeChat Reading TM , Toutiao TM , Sina Weibo TM ,WeChat TM , Alipay TM , Cloud QuickPass TM The ratios of the ranking of APPs to the total number of APPs (i.e. 10) are 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9 and 1 respectively, that is, the trigger confidence of these APPs increases in sequence. For example, when the trigger confidence threshold is 0.6, Toutiao TM , Sina Weibo TM ,WeChat TM , Alipay TM , Cloud QuickPass TM will be used as the preferred APP under the current recommendation factor, and when the preset number is 4, Sina Weibo can be TM ,WeChat TM , Alipay TM , Cloud QuickPass TM These four apps are the apps preferred by users in the subway scenario, for example Figure 8B The YOY0 suggestion card shown shows these 4 apps.

[0261] Understandably, Fig. 9 In the case where the method flow in executes S907, when entering a recommendation factor, that is, within a period of time after the corresponding fence is triggered, the user's preferred APP is an APP whose scene confidence under the recommendation factor is greater than the scene confidence threshold.

[0262] In some embodiments, Fig. 9In the method flow shown, S907 may not be performed, that is, the electronic device selects the APP preferred by the user when entering each corresponding fence according to the preference probability of each APP under each recommendation factor. At this time, the APP preferred by the user when entering the fence corresponding to a recommendation factor is the APP whose preference probability under the recommendation factor is greater than the preference probability threshold.

[0263] In this way, the present application can estimate the preferred APPs when the electronic device enters the fence corresponding to each recommendation factor through data within the historical time period, so that when the user subsequently uses the electronic device to enter the fence corresponding to each recommendation factor, the APPs that the user needs to use can be accurately recommended to the user, so as to facilitate the user to quickly use these APPs.

[0264] Combination Fig.10 , the process of calculating the promotion degree, weighted probability and scene confidence of each APP by the electronic device in the embodiment of the present application is described in detail. Fig.10 As shown, Fig. 8A The illustrated S709 may include S1001 to S1006:

[0265] S1001-S1003, and Fig. 9 S901-S903 shown are the same and will not be described again here.

[0266] S1004: The electronic device calculates the probability of each APP being used for each recommendation factor based on the number of days of use Count (s, day) of each recommendation factor and the number of days of use Count (app|s, day) of each APP under each recommendation factor.

[0267] The probability of using each APP for each recommendation factor can be recorded as p(app|days), which can be achieved by the following formula (6):

[0268] p(app|days)=count(app|s,day) / count(s,day) (6)

[0269] S1005: The electronic device calculates the weighted probability of each APP under each recommendation factor according to the number of times each APP is used under each recommendation factor Count (app|s) and the number of days each APP is used under each recommendation factor Count (app|s, day).

[0270] Among them, the weighted probability of each APP under each recommendation factor can be recorded as Weight_P(app s),

[0271] In some embodiments, the electronic device may use the product of the number of times each APP is used under each recommendation factor Count(app|s) and the number of days each APP is used under each recommendation factor Count(app|s, day) as the weighted probability of each APP under each recommendation factor.

[0272] For example, the weighted probability Weight_P(app|s) of each APP under each recommendation factor can be achieved by the following formula (7-1).

[0273] Weight_P(app|s)=p(app|s)*p(app|days) (7-1)

[0274] Among them, p(app|s) in formula (7-1) can be calculated by the above formula (1-1), that is, by Count(app|s) / Count(s).

[0275] S1006: The electronic device sorts the APPs according to the weighted probability of each recommendation factor from small to large, and uses the ratio of the ranking of each APP under each recommendation factor to the total number of APPs in the electronic device as the scene confidence of each APP under the recommendation factor.

[0276] In some embodiments, in actual applications, in order to facilitate the comparison of the weighted probabilities of each APP under each recommendation factor, the calculation results can be normalized. For example, the implementation formula of the weighted probability Weight_P(app|s) of each APP under each recommendation factor can be replaced by the above formula (7-1) with the following formula (7-2).

[0277] Prob(app|s)=normalize(Weight_P(app|s))=normalize(p(app|s)*p(app|days)) (7-2)

[0278] The word "normalize" in formula (7-2) means normalization, that is, formula (7-2) can first calculate the results of p(app|s)*p(app|days) for each APP under each recommendation factor, and then normalize these results, for example, normalize them to a value between 0 and 1. And Prob(app|s) represents the normalized weighted probability of each APP under each recommendation factor.

[0279] In some embodiments, the above “the electronic device sorts the APPs from small to large according to the weighted probability of each APP under each recommendation factor” means that the electronic device can sort the normalized weighted probabilities.

[0280] It can be understood that the weighted probability Weight_P(app|s) of the number of times each APP is used and the number of days it is used under each recommendation factor not only considers the overall number of times the APP is used, but also the number of days the APP is used. Therefore, if an APP is used a large number of times but a small number of days, the corresponding Weight_P(app|s) value will become smaller.

[0281] For example, for the subway scene recommendation factor, the apps clicked by users under this recommendation factor are WeChat Reading and TM , Cloud QuickPass TM , Sina Weibo TM , Toutiao TM , Taobao TM ,WeChat TM , Meitu Xiuxiu TM , Alipay TM , Gallery, Music, a total of 10 APPs. And the Weight_P(app|s) of these 10 APPs are in the order of Taobao, TM , Meitu Xiuxiu TM , Alipay TM , Gallery, Music, WeChat TM , WeChat Reading TM , Sina Weibo TM , Toutiao TM , Cloud QuickPass TM , that is, the rankings of these apps are 1 to 10. Then, Taobao TM , Meitu Xiuxiu TM , Alipay TM , Gallery, Music, WeChat TM , WeChat Reading TM , Sina Weibo TM , Toutiao TM , Cloud QuickPass TM The ratios of the rankings to the total number of apps (i.e., 10) are 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1, respectively, which means that the scene confidence of these apps increases in sequence. For example, when the scene confidence threshold is 0.5, music, WeChat, TM , WeChat Reading TM , Sina Weibo TM , Toutiao TM , Cloud QuickPass TM The preferred APP under the current recommendation factor will be selected, and when the preset number is 4, the WeChat reading app with higher scene confidence can be selected. TM , Sina Weibo TM , Toutiao TM , Cloud QuickPass TMThese four apps are the users' preferred apps in subway scenarios.

[0282] It can be understood that the weighted probability Weight_P(app|s) of each APP under each recommendation factor in this application not only considers the click frequency of the APP, but also the number of days the APP is used. Therefore, if the click frequency of an APP is large but the number of days of use is small, the corresponding Weight_P(app|s) value will become smaller, which will make the weighted probability and scene confidence of the APP smaller. For example, some APPs are clicked multiple times in a row on a certain day, but are not clicked on other days within a month, indicating that the relevant data of the APP is extreme data and can be used as a basis for judging the user's preferred APP. Then, the weighted probability and scene confidence calculated from the extreme data of this type of APP are small, so that this type of APP will not be judged as the user's preferred APP, which is equivalent to eliminating the extreme data of this type of APP.

[0283] Understandably, Fig.10 When the method flow in S1006 is executed, the APP preferred by the user under a recommendation factor is an APP whose lift under the recommendation factor is greater than the lift threshold and whose scenario confidence (ie, the second evaluation parameter) is greater than the corresponding scenario confidence threshold.

[0284] In some embodiments, Fig.10 In the method flow shown, S1006 may not be executed, that is, the electronic device selects the user's preferred APP under each recommendation factor according to the promotion degree and weighted probability of each APP under each recommendation factor. At this time, the user's preferred APP under a recommendation factor is the APP whose promotion degree under the recommendation factor is greater than the promotion degree threshold and whose weighted probability (i.e., the second evaluation parameter) is greater than the corresponding threshold.

[0285] In this way, the application recommendation method provided by this application not only considers the promotion degree of the user's APP under each recommendation factor, that is, the user's preference for the APP under each recommendation factor, but also considers the user's click frequency and click-day number of these APPs. Therefore, it can not only ensure that the card-issuing APP is the APP that the user prefers more under the current recommendation factor, but also ensure that the card-issuing APP has a certain operation frequency, so that the card-issuing APP is more in line with the actual operation requirements of the recommendation factor.

[0286] Next, combine Fig.11 The processing flow of historical data in this application is described in detail. The execution subject of the process can still be an electronic device. The process includes the following steps:

[0287] S1101: For user APP click records and scene trigger records in historical data, query the recommendation factor configuration corresponding to the historical data according to itemTypeId and key to obtain the storage fields sceneX and primaryKey corresponding to the recommendation factors.

[0288] Among them, each user APP click record in this application may have an itemTypeId and a time information. ItemTypeId can also be called an item category identifier. Among them, an itemTypeId is used to uniquely identify an APP click record. That is, each user APP click record is used to record the next APP clicked and used by the user at a historical moment. In addition, each scene trigger record in this application may have a key (that is, the identifier of the recommendation factor), a recommendation factor description information sceneX (such as geographic location Geo), and a time information. Among them, a key is used to uniquely identify a recommendation factor. That is, each scene trigger record is used to record the next recommendation factor being triggered at a historical moment. Specifically, each user APP click record has a storage field corresponding to itemTypeId and a storage field corresponding to the time information. In addition, each scene trigger record has a storage field corresponding to the key, a storage field corresponding to sceneX, and a storage field corresponding to the time information.

[0289] It can be understood that the present application can associate the scene trigger records with the same time information through the time information in the user APP click record to reflect the recommendation factors of the APP and electronic device clicked by the user in the same time period.

[0290] The primaryKey is used to indicate multiple recommendation factors, such as home, company, hospital, shopping mall, airport, railway station, highway service area, government agency, subway, wearing headphones, car Bluetooth, etc. For example, the primaryKey value of the home scene is "home", the primaryKey value of the subway scene is "subway", the primaryKey value of the wearing headphones scene is "earphones", and the primaryKey value of the car Bluetooth scene is "car-bluetooth-phone".

[0291] Then, the recommendation factor configuration in the above step S1101 refers to the recommendation factor configuration indicated by the user APP click record and the scene trigger record, and the electronic device can query the sceneX field in the scene trigger record, and determine the primaryKey indicating the recommendation factor based on each key, and the sceneX field corresponding to each primaryKey. That is, the above 1001 refers to the electronic device determining the primaryKey field and sceneX field corresponding to each recommendation factor from the historical data.

[0292] S1102: query data from the ofSubjectEventSet and SceneAnnotation tables of the historical data according to itemTypeId, longestDays, and sceneX, and obtain data including the fields eventTimestamp, primaryKey (packageName), and sceneX.

[0293] The longestDays mentioned above refers to the time selected by the user APP click records and scene trigger records in the historical data. For example, if longestDays is 30 days, it refers to the historical data of the most recent month.

[0294] The above-mentioned ofSubjectEventSet and SceneAnnotation table joins correspond to the user APP click records and scene trigger records in the historical data. For example, the ofSubjectEventSet table is used to represent the user APP click records, and the SceneAnnotation table is used to represent the scene trigger records in the historical data.

[0295] eventTimestamp is used to indicate time information, for example, the time information is the time information in a user APP click record, the time information in a scene trigger record, or the time information in a user APP click record and the associated time information, or the time information in one or more scene trigger records and the associated time information in a user APP click record integrated together.

[0296] In some embodiments, the primaryKey field may also include packageName, where packageName may be the name of the APP.

[0297] Thus, the data obtained in the above step S1102 including the fields eventTimestamp, primaryKey (packageName), and sceneX are the multiple historical application usage information mentioned above. Among them, each piece of historical usage information includes a field eventTimestamp for carrying information, a field primaryKey for carrying APP information and a recommendation factor identifier (such as "home" corresponding to the home scene), and a field sceneX for carrying scene information (such as the geographical location of the scene).

[0298] S1103: Filter out the data where sceneX = recommended factor value, and calculate the number of days Count(s, day) of use of each recommended factor. If the number of days Count(s, day) of use is less than the minimum number of days validDays, the calculation ends.

[0299] In some embodiments, the above-mentioned scene value corresponds to a recommendation factor, such as a geographical location corresponding to the recommendation factor. For example, the multiple recommendation factors in the present application are recommendation factors s1, s2, ..., sn-1 and sn, and the recommendation factor values ​​are the geographical locations of scenes s1, s2, ..., sn-1 and sn. Then, the data filtered out with sceneX = scene value is the historical application usage information corresponding to the multiple recommendation factors. The data after filtering out the number of days of use Count(s, day) less than the minimum number of days validDays is the valid historical application usage information corresponding to the multiple recommendation factors.

[0300] S1104: Calculate the number of data in S1102 to obtain the number of times the APP is used in all scenarios Count(all), and calculate the number of data in S1103 to obtain the number of times the APP is used under each recommendation factor Count(s). It can be understood that all recommendation factors are the recommendation factors corresponding to the data in S1102.

[0301] The number of data items in S1102 is the number of items of historical application usage information, thereby obtaining the number of times the APP is used in all scenarios Count (all).

[0302] Calculating the number of data items in S1103 is to calculate the number of valid historical application usage information corresponding to multiple recommendation factors, so as to obtain the number of times the APP is used under each recommendation factor.

[0303] S1105: Group the data in S1102 by primaryKey, calculate the number of data items in each group, and obtain the usage count Count(app) of each APP in all scenarios.

[0304] Grouping the data in S1102 by primaryKey means grouping multiple historical application usage information by recommendation factors.

[0305] S1106: Group the data in S1103 by primaryKey, calculate the number of data items in each group to obtain the number of times each APP is used under each recommendation factor Count (app|s), and calculate the number of days each APP is used under each recommendation factor Count (app|s, day).

[0306] Grouping the data in S1102 by primaryKey means grouping the valid historical application usage information corresponding to multiple recommendation factors by the recommendation factors.

[0307] S1107: Calculate the promotion degree and scene confidence of each APP under each recommendation factor, filter according to the promotion degree threshold and scene confidence threshold, and save the filtering results to the second portrait table.

[0308] The process of calculating the lift and confidence of each APP and filtering according to the lift threshold and scenario confidence threshold can be referred to Figure 7B and Fig.10 The relevant descriptions in the illustrated embodiments are not repeated here.

[0309] S1108: Calculate the preference probability and trigger confidence of each APP under each recommendation factor, trigger the confidence threshold for filtering, and save the filtering results to the first portrait table.

[0310] The process of calculating the preference probability and trigger confidence of each APP and filtering according to the trigger confidence threshold can be referred to Fig. 7A and Fig. 9 The relevant descriptions in the illustrated embodiments are not repeated here.

[0311] In some embodiments, the present application provides a readable medium having instructions stored thereon, which, when executed on an electronic device, causes the electronic device to execute the application recommendation method of the present application.

[0312] The various embodiments of the mechanism disclosed in the present application can be implemented in hardware, software, firmware or a combination of these implementation methods. The embodiments of the present application can be implemented as a computer program or program code executed on a programmable system, which includes at least one processor, a storage system (including volatile and non-volatile memory and / or storage elements), at least one input device and at least one output device.

[0313] Program code can be applied to input instructions to perform the functions described in this application and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, a processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application specific integrated circuit (ASIC), or a microprocessor.

[0314] Program code can be implemented with high-level programming language or object-oriented programming language to communicate with the processing system. When necessary, program code can also be implemented with assembly language or machine language. In fact, the mechanism described in this application is not limited to the scope of any specific programming language. In either case, the language can be a compiled language or an interpreted language.

[0315] In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored on one or more temporary or non-temporary machine-readable (e.g., computer-readable) storage media, which may be read and executed by one or more processors. For example, instructions may be distributed over a network or through other computer-readable media. Therefore, machine-readable media may include any mechanism for storing or transmitting information in a machine (e.g., computer) readable form, including, but not limited to, floppy disks, optical disks, optical disks, read-only memories (CD-ROMs), magneto-optical disks, read-only memories (ROMs), random access memories (RAMs), erasable programmable read-only memories (EPROMs), electrically erasable programmable read-only memories (EEPROMs), magnetic or optical cards, flash memory, or a tangible machine-readable memory for transmitting information (e.g., carrier waves, infrared signals, digital signals, etc.) using the Internet in electrical, optical, acoustic, or other forms of propagation signals. Therefore, machine-readable media include any type of machine-readable media suitable for storing or transmitting electronic instructions or information in a machine (e.g., computer) readable form.

[0316] In the accompanying drawings, some structural or method features may be shown in a specific arrangement and / or order. However, it should be understood that such a specific arrangement and / or order may not be required. Instead, in some embodiments, these features may be arranged in a manner and / or order different from that shown in the illustrative drawings. In addition, the inclusion of structural or method features in a particular figure does not mean that such features are required in all embodiments, and in some embodiments, these features may not be included or may be combined with other features.

[0317] It should be noted that the units / modules mentioned in the various device embodiments of the present application are all logical units / modules. Physically, a logical unit / module can be a physical unit / module, or a part of a physical unit / module, or can be implemented as a combination of multiple physical units / modules. The physical implementation method of these logical units / modules themselves is not the most important. The combination of functions implemented by these logical units / modules is the key to solving the technical problems proposed by the present application. In addition, in order to highlight the innovative part of the present application, the above-mentioned device embodiments of the present application do not introduce units / modules that are not closely related to solving the technical problems proposed by the present application, which does not mean that there are no other units / modules in the above-mentioned device embodiments.

[0318] It should be noted that in the examples and description of this patent, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including one" do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0319] Although the present application has been illustrated and described with reference to certain preferred embodiments thereof, it will be apparent to those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present application.

Claims

1. An application recommendation method, characterized in that: Applied to an electronic device, wherein a first application and a second application are installed on the electronic device; and The method comprises: Determining that the electronic device enters a first fence corresponding to a first recommendation factor; Displaying a first application corresponding to the first fence and recommended to the user within a first time period after the electronic device enters the first fence; After a first time period after the electronic device enters the first fence, a second application corresponding to the first fence recommended to the user is displayed.

2. The method according to claim 1, characterized in that: The electronic device also has a third application installed therein; The preference probability of the first application is greater than the preference probability of a third application that is not recommended to the user, wherein: The higher the preference probability, the higher the probability that the application has been used within a second time period after entering the first fence within a first historical time period, wherein the first time period is less than or equal to the second time period.

3. The method according to claim 2, characterized in that The preference probability is determined based on the usage probability of the application in the second time period after entering the first fence in the first historical time period and the promotion degree of the application in the third time period after entering the first fence in the first historical time period, The higher the promotion degree is, the higher the probability that the application has been used in the third time period after entering the first fence in the first historical time period, the second time period is shorter than the third time period, and The degree of promotion is determined based on a probability of using the application in a third time period after entering the first fence within the first historical time period and a probability of using the application in a third time period after entering fences corresponding to multiple recommendation factors within the first historical time period, wherein the multiple recommendation factors include the first recommendation factor.

4. The method according to claim 3, characterized in that The usage probability of the application in the second time period after entering the first fence in the first historical time period is calculated in the following manner: Calculate topK prop = count(s|app in topN) / M, Among them, topK prop represents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each of the sequences including applications used in the second time period after entering the first fence corresponding to the first recommendation factor, count(s|app in topN) represents the number of sequences containing the application in the multiple sequences corresponding to the first recommendation factor, topK prop Indicates the usage probability of the application in the second time period after entering the first fence in the first historical time period.

5. The method according to claim 4, characterized in that The preference probability is calculated by the following formula: Calculating hits prop =exp(lift-a)*topK prop , Among them, hit prop represents the preference probability applied in the second time period after entering the first fence in the first historical time period, exp() represents exponential operation, topK prop represents the usage probability of the application in the second time period after entering the first fence in the first historical time period, lift represents the lift degree of the application in the third time period after entering the first fence in the first historical time period, and a is a constant related to the lift degree.

6. The method according to claim 1, characterized in that The electronic device also has a third application installed therein; The preference probability of the first application is greater than the preference probability of a third application that is not recommended to the user, and The higher the preference probability, the higher the probability of the application appearing in the top N applications used after entering the first fence within the first historical time period, where N is a positive integer.

7. The method according to claim 6, characterized in that The preference probability is determined based on the probability of the application appearing in the top N applications used after entering the first fence in the first historical time period and the promotion degree of the application in a third time period after entering the first fence in the first time period, wherein: The higher the promotion degree, the higher the probability that the application has been used in the third time period after entering the first fence within the first historical time period, the first time period is smaller than the third time period, and The degree of promotion is determined based on a probability of using the application in a third time period after entering the first fence within the first historical time period and a probability of using the application in a third time period after entering fences corresponding to multiple recommendation factors within the first historical time period, wherein the multiple recommendation factors include the first recommendation factor.

8. The method according to claim 7, characterized in that The probability of an application appearing in the first N applications used after entering the first fence during the first historical time period is calculated as follows: Calculate topK prop = count(s|app in topN) / M, where topK prop represents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each of the sequences including the first N applications used after entering the first fence corresponding to the first recommendation factor, count(s|app in topN) represents the number of sequences containing the application in the multiple sequences corresponding to the first recommendation factor, topK prop Indicates the occurrence probability of the application in the first N applications used after entering the first fence within the first historical time period.

9. The method according to claim 7 or 8, characterized in that: The preference probability is calculated as follows: Calculating hits prop =exp(lift-a)*topK prop (app|s), where hit prop represents the preference probability of the application among the first N applications used after entering the first fence during the first historical time period, exp() represents exponential operation, topK prop (app|s) represents the probability of the application appearing in the top N applications used after entering the first fence within the first historical time period, lift represents the lift degree of the application in the third time period after entering the first fence within the first historical time period, and a is a constant related to the lift degree.

10. The method according to claim 5 or 9, characterized in that: The lifting degree is a value between 0 and 1, and a is 0.

5.

11. The method according to any one of claims 1 to 10, characterized in that After a first time period after the electronic device enters the first fence, displaying a second application corresponding to the first fence recommended to the user includes: It is detected that the recommended first application is used by the user, the first time period ends, and the second application is displayed.

12. The method according to claim 2 or 6, characterized in that: The first confidence level of the first application is greater than the first confidence level of the third application, and The first confidence level is determined based on the preference probability of the application and the total number of applications in the electronic device.

13. The method according to claim 12, characterized in that The first confidence level is: The ratio of the first ranking of applications to the total number of applications in the electronic device, wherein: The first ranking of the application is the ranking of the preference probability of the application in the ascending order of the preference probabilities of the applications in the electronic device.

14. The method according to claim 13, characterized in that The preference probability of the first application is greater than a preference probability threshold, or the preference probability of the first application is greater than the preference probability threshold and the first confidence of the first application is greater than a first confidence threshold.

15. The method according to any one of claims 3 or 7, characterized in that: The electronic device also has a fourth application installed, wherein The promotion degree of the second application is greater than the promotion degree of the fourth application that is not recommended to the user.

16. The method according to claim 15, characterized in that The lift is calculated as follows: Calculate p=Count(app|s) / Count(s), where p is the probability of using the application in the third time period after entering the first fence in the first historical time period, Count(app|s) represents the number of times the application is used in the third time period after entering the first fence in the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the third time period after entering the first fence each time in the first historical time period; calculate Among them, willson ctr (app|s) represents the modified usage probability of the application in the first historical time period in the third time period after entering the first fence, n is the number of times the application enters the first fence in the first historical time period, and z is a constant; calculate Among them, lift represents the lift of the application in the third time period after entering the first fence within the first historical time period, and p(app|all) represents the ratio of the number of uses of the application in the third time period after entering the fences corresponding to multiple recommendation factors within the first historical time period to the number of uses of multiple applications in the third time period after entering the fences corresponding to multiple recommendation factors within the first historical time period.

17. The method according to claim 16, characterized in that The weighted probability of the second application is greater than the weighted probability of the fourth application; And, the weighted probability applied in the third time period after entering the first fence is calculated in the following manner: Calculate p(app|s)=Count(app|s) / Count(s), where p(app|s) is the usage probability of the application in the third time period after entering the first fence within the first historical time period, Count(app|s) represents the number of times the application is used in the third time period after entering the first fence within the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the third time period after entering the first fence each time within the first historical time period; Calculate p(app|days)=count(app|s,day) / count(s,day), where p(app|days) represents the probability of the number of days the application is used in the third time period after entering the first fence within the first historical time period, count(app|s,day) represents the number of days the application is used in the third time period after entering the first fence within the first historical time period, and count(s,day) represents the number of days a plurality of applications are used in the third time period after entering the first fence within the first historical time period; Calculate Weight_P(app|s)=normalize(p(app|s)*p(app|days)), where Weight_P(app|s) represents the weighted probability of the application in the third time period after entering the first fence, and normalize() represents normalization processing.

18. The method according to claim 17, characterized in that The second confidence of the second application is greater than the second confidence of the fourth application; The second confidence level applied is: The ratio of the second ranking of applications to the total number of applications in the electronic device, wherein: The second ranking of the application is the ranking of the weighted probability of the application in the ascending order of the weighted probabilities of the applications in the electronic device.

19. The method according to claim 2, characterized in that The method further comprises: It is determined that the number of days entering the first fence during the first historical time period is greater than the first day.

20. The method according to claim 1, characterized in that The first recommendation factor includes at least one of the following: Time factor, space factor, and device usage factor, wherein the device usage factor is used to indicate whether the electronic device is connected to an external device; The first fence includes at least one of the following: a time fence, a geographic fence, and a device fence.

21. The method according to claim 1, characterized in that The application corresponding to the first fence recommended to the user is displayed in an application suggestion card of the electronic device.

22. An application recommendation method, characterized in that: Applied to an electronic device, the electronic device having a first application to be recommended and a second application to be recommended installed thereon; and The method comprises: Determining that the electronic device enters a first fence corresponding to a first recommendation factor; displaying a first application to be recommended corresponding to the first fence that is recommended to the user, wherein the preference probability of the first application to be recommended is greater than the preference probability of the second application to be recommended that is not recommended to the user, and The higher the preference probability is, the higher the probability that the application has been used within the first target time period after entering the first fence within the first historical time period.

23. The method according to claim 22, characterized in that The preference probability is calculated by the following formula: Calculate topK prop = count(s|app in topN) / M; Calculating hits prop =exp(lift-a)*topK prop ; Among them, topK prop represents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each of which includes an application used in the first target time period after entering the first fence corresponding to the first recommendation factor, count(s|app in topN) represents the number of sequences containing the application in the multiple sequences corresponding to the first recommendation factor, topK prop represents the usage probability of the application in the first target time period after entering the first fence in the first historical time period; hit prop represents the preference probability applied in the first target time period after entering the first fence in the first historical time period, exp() represents exponential operation, topK prop represents the usage probability of the application in the first target time period after entering the first fence in the first historical time period, lift represents the lift degree of the application in the second target time period after entering the first fence in the first historical time period, and a is a constant related to the lift degree.

24. The method according to claim 23, characterized in that The electronic device further has a third application to be recommended installed therein, a first confidence level of the first application to be recommended is greater than a first confidence level of the third application to be recommended, and, The first confidence level is: The ratio of the first ranking of applications to the total number of applications in the electronic device, wherein: The first ranking of the application is the ranking of the preference probability of the application in the ascending order of the preference probabilities of the applications in the electronic device.

25. The method according to claim 23 or 24, characterized in that The lift is calculated as follows: Calculate p=Count(app|s) / Count(s), where p is the usage probability of the application in the second target time period after entering the first fence in the first historical time period, Count(app|s) represents the number of times the application is used in the second target time period after entering the first fence in the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the second target time period after entering the first fence each time in the first historical time period; calculate Among them, willson ctr (app|s) represents the corrected usage probability of the application in the first historical time period in the second target time period after entering the first fence, n is the number of times the application enters the first fence in the first historical time period, and z is a constant; calculate Among them, lift represents the lift of the application in the second target time period after entering the first fence within the first historical time period, and p(app|all) represents the ratio of the number of uses of the application in the second target time period after entering the fences corresponding to multiple recommendation factors within the first historical time period to the number of uses of multiple applications in the second target time period after entering the fences corresponding to multiple recommendation factors within the first historical time period.

26. An application recommendation method, characterized in that: Applied to an electronic device, the electronic device having a first application to be recommended and a second application to be recommended installed thereon; and The method comprises: Determining that the electronic device enters a first fence corresponding to a first recommendation factor; displaying a first application to be recommended corresponding to the first fence that is recommended to the user, wherein the preference probability of the first application to be recommended is greater than the preference probability of the second application to be recommended that is not recommended to the user, and The higher the preference probability, the higher the probability of the application appearing in the top N applications used after entering the first fence within the first historical time period, where N is a positive integer.

27. The method according to claim 26, characterized in that The preference probability is calculated by the following formula: Calculate topK prop = count(s|app in topN) / M; Calculating hits prop =exp(lift-a)*topK prop ; Among them, topK prop represents the number of multiple sequences corresponding to the first recommendation factor in the first historical time period, each of the sequences including the first N applications used after entering the first fence corresponding to the first recommendation factor, count(s|app in topN) represents the number of sequences containing the application in the multiple sequences corresponding to the first recommendation factor, topK prop represents the occurrence probability of the application in the first N applications used after entering the first fence within the first historical time period; hit prop represents the preference probability of the application among the first N applications used after entering the first fence during the first historical time period, exp() represents exponential operation, topK prop (app|s) represents the probability of the application appearing in the top N applications used after entering the first fence within the first historical time period, lift represents the lift degree of the application in the target time period after entering the first fence within the first historical time period, and a is a constant related to the lift degree.

28. The method according to claim 27, characterized in that The electronic device further has a third application to be recommended installed therein, a first confidence level of the first application to be recommended is greater than a first confidence level of the third application to be recommended, and, The first confidence level is: The ratio of the first ranking of applications to the total number of applications in the electronic device, wherein: The first ranking of the application is the ranking of the preference probability of the application in the ascending order of the preference probabilities of the applications in the electronic device.

29. The method according to claim 27 or 28, characterized in that The lift is calculated as follows: Calculate p=Count(app|s) / Count(s), where p is the probability of using the application in the target time period after entering the first fence in the first historical time period, Count(app|s) represents the number of times the application is used in the target time period after entering the first fence in the first historical time period, and Count(s) represents the sum of the number of times multiple applications are used in the target time period after entering the first fence each time in the first historical time period; calculate Among them, willson ctr (app|s) represents the corrected usage probability of the application in the first historical time period in the target time period after entering the first fence, n is the number of times the application enters the first fence in the first historical time period, and z is a constant; calculate Among them, lift represents the lift degree of the application in the target time period after entering the first fence within the first historical time period, and p(app|all) represents the ratio of the number of uses of the application in the target time period after entering the fences corresponding to multiple recommendation factors within the first historical time period to the number of uses of multiple applications in the target time period after entering the fences corresponding to multiple recommendation factors within the first historical time period.

30. A readable medium, characterized in that The readable medium stores instructions, which, when executed on an electronic device, enable the electronic device to execute the application recommendation method according to any one of claims 1 to 29.

31. An electronic device, characterized in that: include: A memory for storing instructions executed by one or more processors of an electronic device, and a processor, which is one of the processors of the electronic device, for executing the application recommendation method described in any one of claims 1 to 29.

Images