WO2018113409A1 - 启动资源加载方法及装置 - Google Patents

启动资源加载方法及装置 Download PDF

Info

Publication number
WO2018113409A1
WO2018113409A1 PCT/CN2017/107457 CN2017107457W WO2018113409A1 WO 2018113409 A1 WO2018113409 A1 WO 2018113409A1 CN 2017107457 W CN2017107457 W CN 2017107457W WO 2018113409 A1 WO2018113409 A1 WO 2018113409A1
Authority
WO
WIPO (PCT)
Prior art keywords
context data
historical
current
similarity
context
Prior art date
Application number
PCT/CN2017/107457
Other languages
English (en)
French (fr)
Inventor
付柳强
Original Assignee
腾讯科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 腾讯科技(深圳)有限公司 filed Critical 腾讯科技(深圳)有限公司
Publication of WO2018113409A1 publication Critical patent/WO2018113409A1/zh
Priority to US16/368,451 priority Critical patent/US11169827B2/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/445Program loading or initiating
    • G06F9/44568Immediately runnable code
    • G06F9/44578Preparing or optimising for loading
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/445Program loading or initiating

Definitions

  • the embodiments of the present invention relate to the field of mobile terminals, and in particular, to a method and an apparatus for loading a startup resource.
  • the startup resource includes: code of the application, user interface (UI) material required by the application at runtime, data that the application needs to process at runtime, and the like.
  • the mobile terminal receives the user's click operation and completely starts the application. It takes a lot of time.
  • the embodiment of the present application provides a startup resource loading method and device.
  • the technical solution is as follows:
  • the application provides a method for loading a startup resource, where the method includes:
  • the current context data group includes at least one current context data for describing a current running scenario
  • each historical situation data set includes at least one historical situation data for describing a historical running situation
  • the boot resource of the predictive application is preloaded into memory, which is the resource required to launch the predictive application.
  • the according to the current context data group and the historical context a data set, calculating a similarity between the current context and each of the historical contexts, including:
  • the current context data set and each of the historical context data sets are input to a similarity model, and a similarity between the current context and each of the historical contexts is calculated.
  • the determining that the similarity between the current situation and the current situation meets the set condition includes:
  • N historical contexts having the greatest similarity to the current context are determined; wherein N is a positive integer.
  • the current context data includes at least one of current user context data, current environment context data, and current terminal context data;
  • the current user context data is used to describe information of a user who uses the mobile terminal when the mobile terminal is currently running, and the current user context data includes the user's name, gender, age, occupation, mood, and At least one of educational backgrounds;
  • the current environment context data is used to describe an environment in which the user of the mobile terminal is located when the mobile terminal is currently running, and the current environment context data includes time, location, weather conditions, temperature, illumination, sound, and At least one of traffic conditions;
  • the current terminal context data is used to describe information of the mobile terminal, and the current terminal context data includes at least one of a terminal identifier, network information, and a device type of the mobile terminal.
  • the current context data group includes at least two of the current context data
  • the historical context data group includes at least two of the historical context data
  • the calculating results in a similarity between the current context and each of the historical contexts including:
  • the sub-similarity is the current context data and the historical context data type of the same context data type
  • the similarity between the historical context data; the context data type includes: at least one of user context data, environmental context data, and terminal context data; and the sub-similarity corresponding to each of the current context data is according to the current
  • the weights corresponding to the context data types to which the context data belongs are added, and the similarity between the current context and the historical context is obtained.
  • the sub-similarity corresponding to each of the current context data is added according to a weight corresponding to the context data type to which the current context data belongs, to obtain the current context. Similarities with the historical context, including:
  • C1 is the current context data set
  • C2 is the historical context data set
  • Sim (C1, C2) is a similarity between the current context and the historical context
  • n is the current context data set The number of current context data in the current context data
  • S i is the i-th current context data in the current context data group
  • p i is the weight corresponding to the context data type to which S i belongs
  • S j is the historical context the historical context data with the data set S i belong to the same context data type
  • Sim (S i, S j ) is the current context data corresponding to the sub-similarity
  • n ⁇ 2 and n is a positive integer.
  • the calculating the sub-similarity corresponding to each of the current context data in the current context data group includes:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data
  • S i is an i-th current context data in the current context data group
  • S j is the historical context data.
  • the calculating the sub-similarity corresponding to each of the current context data in the current context data group includes:
  • the sub-similarity is calculated according to the following formula:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data
  • S i is an i-th current context data in the current context data group
  • S j is the historical context data.
  • the minimum value, r i e is the maximum value of S i
  • r j s is the minimum value of S j
  • r j e is the maximum value of S j .
  • the calculating the sub-similarity corresponding to each of the current context data in the current context data group includes:
  • the sub-similarity is calculated according to the following formula:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data
  • S i is an i-th current context data in the current context data group
  • S j is the historical context data.
  • the determining, by the history application corresponding to the N historical contexts with the highest degree of similarity, the prediction application corresponding to the current context including:
  • association initiation probability Determining an association initiation probability of each of the historical applications relative to the current application, the association initiation probability being used to characterize a probability that the historical application is associatedly initiated during operation of the current application;
  • P history applications having the largest product of the recommendation weight and the corresponding association start probability are determined as the prediction application, and P is an integer greater than or equal to 1.
  • the acquiring the current context data group corresponding to the current context further includes:
  • the original context data is pre-processed to generate the current context data group corresponding to the current context.
  • the method before the inputting the current context data group and each of the historical context data groups into the similarity model, the method further includes:
  • each set of the historical context samples including the historical context data corresponding to each of the two historical contexts, and a context similarity score between the two historical contexts, the context
  • the similarity score is calibrated according to the similarity between the historical applications corresponding to the two historical contexts;
  • the initial similarity model is trained according to the historical situation sample to obtain the trained similarity model, and the initial similarity model includes at least one context data type and an initial weight corresponding to each context data type.
  • the method further includes:
  • the application provides a boot resource loading device, the device including a processor and a memory:
  • the memory is configured to store program code and transmit the program code to the processor
  • the processor is configured to perform the boot resource loading method of any one of the first aspects according to the instructions in the program code.
  • the present application provides a storage medium for storing program code, the program code for performing the boot resource loading method according to any one of the first aspects.
  • the present application provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the boot resource loading method of any of the first aspects.
  • the present application provides a method for loading a startup resource, which is applied to a mobile terminal, and the mobile terminal performs the startup resource loading method according to any one of the first aspects.
  • the current situation data group corresponding to the current situation is obtained, and the historical situation data group corresponding to each of the plurality of historical situations is obtained. Then, based on the current context data group and the historical context data group, the similarity between the current context and each of the historical contexts is calculated. Then, determining a historical situation in which the similarity with the current situation meets the set condition, and The prediction application corresponding to the current context is determined according to the historical application corresponding to the historical context, wherein the historical application is an application running in the historical context. Finally, the boot application's startup resources are preloaded into memory.
  • the mobile terminal can predict the historical application corresponding to the historical situation with higher similarity as the predicted application that the mobile terminal is about to start. Program and preload the boot resources required by the predictive application when it starts up to memory. In this way, when the mobile terminal receives the click operation of the user, the mobile terminal can directly use the pre-loaded startup resource to start the prediction application, so that the mobile terminal does not start loading the startup resource after receiving the user's click operation. To start the predictive application, which reduces the time it takes for the mobile terminal to launch the application.
  • FIG. 1 is a schematic structural diagram of an implementation environment involved in various embodiments of the present application.
  • FIG. 2 is a flowchart of a method for loading a startup resource according to an exemplary embodiment of the present application
  • FIG. 3 is a flowchart of a method for loading a startup resource according to another exemplary embodiment of the present application.
  • FIG. 4 is a flowchart of a method for loading a startup resource according to another exemplary embodiment of the present application.
  • FIG. 5 is a flowchart of a method for loading a startup resource according to another exemplary embodiment of the present application.
  • FIG. 6 is a flowchart of a method for loading a startup resource according to another exemplary embodiment of the present application.
  • FIG. 7 is a flowchart of a method for loading a startup resource according to another exemplary embodiment of the present application.
  • FIG. 8 is a block diagram of a boot resource loading apparatus according to an exemplary embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of a mobile terminal according to an exemplary embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a boot resource loading device according to an exemplary embodiment of the present application.
  • the startup resource loading method provided by various embodiments of the present application can be compressed by a Picture Picture Experts Group Audio Layer III (MP3) player and a motion picture, such as a smart phone, a tablet computer, an e-book reader, and a motion picture expert.
  • MP3 Picture Picture Experts Group Audio Layer III
  • the expert implements a mobile terminal such as a Moving Picture Experts Group Audio Layer IV (MP4), and an application is installed in the mobile terminal.
  • MP4 Moving Picture Experts Group Audio Layer IV
  • An implementation environment of a method for loading a load resource may also be as shown in FIG. 1 , where the implementation environment may include: a mobile terminal 120 and a server 140.
  • the mobile terminal 120 can be connected to the server 140 via a wireless network.
  • the server 140 may be a server, a server cluster composed of a plurality of servers, or a cloud computing center, and the server 140 has a model training function.
  • the above wireless network may be using standard communication technologies and/or protocols.
  • the network can usually be the Internet, but can also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile network, and a wired network. Any combination of networks or wireless networks, private networks, or virtual private networks.
  • data exchanged over a network may be represented using techniques and/or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), and the like.
  • SSL Secure Socket Layer
  • TLS Transport Layer Security
  • VPN Virtual Private Network
  • IPsec Internet Protocol Security
  • Regular encryption techniques are used to encrypt all or some of the links.
  • the above described data communication techniques may also be replaced or supplemented using custom and/or dedicated data communication techniques.
  • FIG. 2 is a flow chart of a boot resource loading method shown in an exemplary embodiment of the present disclosure. This embodiment is exemplified by the method used in the above mobile terminal 120. The method includes:
  • Step 201 The mobile terminal 120 acquires a current context data group corresponding to the current context, where the current context data group includes at least one current context data for describing the current running scenario.
  • the current context data may be classified according to the context data type, wherein the context data type may include at least one of user context data, environment context data, and terminal context data. Therefore, the current context data may include at least one of current user context data, current environmental context data, and current terminal context data.
  • the current user context data may be used to describe information of the user who uses the mobile terminal when the mobile terminal is currently running.
  • the current user context data may include the user's name, gender, age, occupation, mood, and educational background.
  • At least one of the current environment context data may be used to describe an environment in which the user who uses the mobile terminal is currently running when the mobile terminal is currently running.
  • the current environment context data may include time, location, weather conditions, temperature, illumination, At least one of a sound and a traffic situation; the current terminal context data may be used to describe information of the mobile terminal during current operation, such as the current terminal context data may include at least the terminal identifier, network information, and device type of the mobile terminal One.
  • a set of illustrative current situation data sets can be expressed in the form of a set: ⁇ terminal identification, time, gender, age, location, temperature ⁇ .
  • the current context data may be represented in the form of quantized values and/or textual form. For example, when the current context data includes gender, the current context data may be “male”; when the current context data includes temperature, the current context data may be “20° C.”.
  • the current context data may be represented by any one of a numerical value, an interval, and a coordinate.
  • the age included in the current context data may be “18”; for example, if the current context data includes temperature, And the temperature is expressed in the form of a section, and the temperature included in the current context data may be “18° C-25° C.”; for example, if the current context data includes a location, and the location is identified by a coordinate form.
  • the position included in the current context data may be "(40°N, 116°E)".
  • Step 202 The mobile terminal 120 acquires a historical context data group corresponding to each of the plurality of historical contexts, and each historical context data group includes at least one historical context data for describing the historical running scenario.
  • the historical context data may also be classified according to the context data type. Therefore, the historical context data may include at least one of historical user context data, historical environment context data, and historical terminal context data. Specifically, the context data type and the representation of the historical context data may be combined with the current context data in the foregoing step 201, which is not described in this embodiment.
  • a historical context data group corresponding to each historical context may be stored in the database in the mobile terminal. Therefore, the mobile terminal may acquire a historical context data group corresponding to each historical context in the database, or may acquire a historical context data group corresponding to a part of the historical context in the database.
  • the mobile terminal when the mobile terminal acquires a historical context data group corresponding to a part of the historical context in the database, the mobile terminal may acquire a historical context data group corresponding to the latest M historical contexts in the database, or the mobile terminal may acquire the M randomly.
  • step 202 and step 201 have no specific sequence, and may generally be performed simultaneously.
  • Step 203 The mobile terminal 120 calculates the similarity between the current context and each historical context according to the current context data group and the historical context data group.
  • the current context data set and each historical context data set may be input into the similarity model, and the similarity between the current context and each historical context is calculated according to the similarity model.
  • the similarity model may be a model that the mobile terminal and/or the server trains in advance according to at least one set of historical context samples, and each set of historical context samples may include historical context data corresponding to each of the two historical contexts, and the two historical contexts.
  • a context similarity score which may be calibrated based on the similarity between historical applications corresponding to the two historical contexts.
  • Step 204 The mobile terminal 120 determines a historical context in which the similarity with the current context meets the set condition, and determines a predicted application corresponding to the current context according to the historical application corresponding to the historical context.
  • the history application is an application that runs in a historical situation.
  • the historical application corresponding to the historical context may be saved in the database, so that the mobile terminal can obtain each historical context from the database. History app.
  • the mobile terminal may first determine N historical contexts with the highest similarity, and then use the historical application corresponding to the N historical contexts as the prediction application.
  • the value of N can be preset by the system or customized by the user. It should be noted that the value of N is not limited in this embodiment.
  • the mobile terminal may first determine a historical context in which the similarity between the current context and the current context is greater than the similarity threshold, and then use the historical application corresponding to the historical context as the predictive application.
  • the similarity threshold may be a system preset value or a user-defined value. It should be noted that the manner in which the similarity threshold is set in this embodiment is not limited.
  • the prediction application determined by the mobile terminal includes a history application corresponding to the historical context 1 and a history application corresponding to the historical context 4.
  • the prediction application determined by the mobile terminal may include a history application corresponding to the historical context 1, a history application corresponding to the historical context 3, and a history application corresponding to the historical context 4.
  • Step 205 the mobile terminal 120 preloads the startup resource of the prediction application into the memory, and the startup resource is a resource required when starting the prediction application.
  • the method for loading a startup resource obtains a current context data group corresponding to the current context, and a historical context data group corresponding to each of the M historical contexts, and according to the current context data group and the history.
  • the context data set obtains the similarity between the current context and each of the historical contexts.
  • the prediction application corresponding to the current context is determined according to the history application corresponding to the N historical contexts with the largest similarity.
  • the boot application's startup resources are preloaded into memory. Since the mobile terminal has a higher probability of starting the same application when in a similar situation, the mobile terminal can predict the historical application corresponding to the historical situation with higher similarity as the predicted application that the mobile terminal is about to start.
  • Program and preload the boot resources required by the predictive application when it starts up to memory In this way, when the mobile terminal receives the click operation of the user, the mobile terminal can directly use the pre-loaded startup resource to start the prediction application, so that the mobile terminal does not start loading the startup resource after receiving the user's click operation. To start the predictive application, which reduces the time it takes for the mobile terminal to launch the application.
  • the method may further include a process of model training. Specifically, the following steps may be included:
  • Step 301 The mobile terminal 120 acquires at least one set of historical context samples.
  • Each set of historical context samples may include historical context data corresponding to each of the two historical contexts, and a context similarity score between the two historical contexts, and the context similarity score may be corresponding to the two historical contexts.
  • the similarity between the historical applications is calibrated.
  • the relationship between the situation similarity score and the similarity between the historical applications may be preset by the system. For example, the similarity between the situation similarity score and the historical application may be proportional to the relationship. If the situation similarity score is higher, the similarity between the historical applications is higher. Conversely, the lower the situation similarity score, the lower the similarity between the historical applications.
  • the similarity between two historical applications may be determined by calculating the similarity between the tags corresponding to the two applications, for example, obtaining the corresponding tags of the two applications. After that, the same label in the corresponding label of each of the two historical applications may be determined, and then the proportion of the same label in the label corresponding to the two historical applications is calculated, and then, the ratio may be determined according to the ratio. The similarity between these two historical applications.
  • the similarity score has a value range of [0, 1], when two historical contexts correspond to the same historical application.
  • the situational similarity score between the two historical contexts is 1; assuming that the historical context a corresponds to the historical application a, the labels corresponding to the historical application a include food, group purchase, and mobile phone, and the historical context b corresponds to the historical application b.
  • the tag corresponding to the history application b includes a mobile phone, a clothing, and a KTV.
  • the history application a and the history application b have only one same tag, that is, a mobile phone, the history application a and the history application b
  • the similarity between the historical context a and the historical context b may be equal to the similarity between the historical application a and the historical application b of 0.33.
  • Step 302 The mobile terminal 120 trains the initial similarity model according to the historical context sample to obtain a trained similarity model.
  • the initial similarity model includes at least one context data type, and an initial weight corresponding to each context data type, wherein the initial weight is a system preset or a user-defined weight.
  • each context data type in the initial similarity model has equal initial weights. For example, when there are four context data types in the initial similarity model, the initial weights corresponding to each context data type are It is 0.25.
  • the initial similarity model may be any one of an artificial neural network, an Adaboost, a Support Vector Machine (SVM), a genetic algorithm, and a naive Bayesian model.
  • the artificial neural network may be a Region-based Convolutional Neural Networks (R-CNN), a Fast R-CNN learning algorithm or a Faster R-CNN learning algorithm.
  • the mobile terminal may send each historical context sample to a server having model training capability, according to the history of the server.
  • the situation sample training results in a similarity model.
  • step 301 and step 302 are usually performed before step 201 above. However, in actual implementation, it may be performed simultaneously with step 201 and step 202. This is not limited.
  • the running scenario in which the mobile terminal is located usually includes a large amount of context data
  • the current context data group acquired by the mobile terminal may include at least two current context data, and correspondingly, the historical context data acquired by the mobile terminal.
  • the group may also include at least two historical context data. Therefore, in an implementation manner of the embodiment of the present application, as shown in FIG. 4, the foregoing step 203 may include the following steps:
  • Step 401 The mobile terminal 120 calculates a sub-similarity corresponding to each current context data in the current context data group.
  • the sub-similarity corresponding to the current context data may be the similarity between the current context data and the historical context data of the same context data type in the historical context data group.
  • the context data type refer to the related introduction in step 201 above. This embodiment will not be described again.
  • the current situation data is a location
  • the historical context data belonging to the same context data type as the current context data is also a location
  • the sub-similarity corresponding to the current context data is the current context data “location” and the historical context data “location”. Similarity between the two.
  • the method used by the mobile terminal 120 to calculate the sub-similarity corresponding to the current context data is also different, as follows:
  • the mobile terminal 120 may calculate the sub-similarity according to the following formula:
  • Sim(S i , S j ) is the sub-similarity corresponding to the current context data; S i is the i-th current context data in the current context data group; S j is the same as the S i in the historical context data group Historical context data of the context data type; 1 ⁇ i ⁇ n, n is the number of current context data in the current context data group.
  • the mobile terminal 120 calculates the sub-similarity corresponding to the current context data “temperature” as 1-
  • /25. °C 0.8.
  • the mobile terminal 120 may calculate the sub-similarity according to the following formula:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data;
  • S i is the i-th current context data in the current context data group;
  • S j is the historical context data the historical context data set S i and belonging to the same context data type;
  • S i ⁇ [r i s , r i e], S j ⁇ [r j s, r j e], r i s the form S i
  • r i e is the maximum value of the current context data represented by S i ;
  • r j s is the minimum value of the historical context data represented by S j ;
  • r j e is the historical context data represented by S j The maximum value.
  • the mobile terminal 120 calculates the sub-similarity corresponding to the current context data "temperature” as 1-
  • the mobile terminal 120 may calculate the sub-similarity according to the following formula:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data;
  • S i is the i-th current context data in the current context data group;
  • S j is the historical context data the historical context data set S i and belonging to the same context data type;
  • S i (x i, y i),
  • S j (x j, y j).
  • the current context data when the current context data is represented by a text form, the current context data may be quantized to be represented in the form of a quantized value, and then the sub-similarity corresponding to the current context data is calculated.
  • the current situation data when the current situation data is gender, the "male” may be quantized as 1 and the “female” may be quantified as 0.
  • This embodiment does not limit this, but the embodiment of the present application will be expressed in the form of quantized values.
  • the current situation data is explained as an example.
  • Step 402 The mobile terminal 120 adds the sub-similarity corresponding to each current context data according to the weight corresponding to the context data type to which the current context data belongs, to obtain the similarity between the current context and the historical context.
  • the mobile terminal 120 may calculate the similarity according to the following formula:
  • C1 is the current context data set
  • C2 is the historical context data set
  • Sim (C1, C2) is the similarity between the current context and the historical context
  • n is the number of current context data in the current context data set, N ⁇ 2 and n is a positive integer
  • S i is the i th the current context data in the current context data group
  • S j is the historical context in the historical context data group that belongs to the same context data type as S i Data
  • Sim (S i , S j ) is the sub-similarity corresponding to the current context data
  • Sim(S i , S j ) may be calculated by using any one of the above three possible implementation manners.
  • the specific calculation manner may be combined with the example in the foregoing step 401, which is not described in this embodiment.
  • each current context data in the current context data group exists with historical context data belonging to the same context data type.
  • a current context data There may be no historical context data in the historical context data group that belongs to the same context data type.
  • the data types included in the historical context data corresponding to each historical context may be exemplified, but the present invention is not limited thereto.
  • the mobile terminal 120 when the mobile terminal 120 is in the current context startup prediction application, other applications may be started instead due to the interruption of the emergency event. For example, when the mobile terminal 120 is about to start the playback application, it is in the background.
  • the running instant messaging client receives the instant messaging message, the user may open the image class application in the process of replying to the instant messaging message.
  • the mobile terminal 120 may The historical application corresponding to the N historical contexts with the highest degree of similarity is not directly determined as the prediction application. As shown in FIG. 5, the foregoing step 204 may include the following steps:
  • Step 501 The mobile terminal 120 determines the recommended weight of the historical application corresponding to the historical context according to the similarity of each of the N historical contexts.
  • the recommended weight may be in the range of [0, 1], and the recommended weight may be proportional to the similarity between the historical context and the current situation. It can be understood that the greater the similarity between the historical context and the current situation is.
  • the recommendation weight of the historical application corresponding to the historical context is larger, and the correspondence between the similarity and the recommended weight may be system preset or user-defined.
  • the mobile terminal 120 determines the recommended weights of the historical applications corresponding to the N historical contexts with the greatest similarity between the current contexts. In actual implementation, the mobile terminal 120 may also determine each The recommended weight of the historical application corresponding to the historical context.
  • step 502 the mobile terminal 120 acquires the current application that is already in the running state.
  • the current application in the running state may include an application running in the foreground of the mobile terminal and An application running in the background of a mobile terminal.
  • step 503 the mobile terminal 120 determines an association start probability of each history application relative to the current application.
  • the mobile terminal 120 Since the mobile terminal 120 is in the process of running an application, other applications may be associated with each other. For example, when the application runs the shopping application, the startup payment application is usually triggered, and the shopping application is launched. Start with a payment class application that has a high probability of being associated.
  • the mobile terminal can collect the data initiated by the correlation between the applications as the association start sample, and train the association start sample according to the predetermined training algorithm to obtain the association startup model.
  • the association start probability of each historical application can be determined based on the association startup model.
  • the association start probability may be used to represent the probability that the history application is associated with the start of the current application.
  • the value of the association start probability may be [0, 1]. It can be understood that the historical application The greater the probability that a program is launched by the current application association, the greater the probability that the history application will be associated with the current application.
  • the mobile terminal 120 may determine the association initiation probability of the historical applications relative to the current application, or determine that the mobile terminal includes For the meaning of the association start of each application relative to the current application, the meaning of the predetermined training algorithm can be referred to the related description of step 302 above, which is not described in this embodiment.
  • Step 504 The mobile terminal 120 determines, in the N historical applications, P historical applications that have the largest product of the recommended weight and the corresponding association start probability as the prediction application, and P may be an integer greater than or equal to 1.
  • the mobile terminal can calculate A prediction application is determined, where Q is a historical application in which N ⁇ i * ⁇ i takes a maximum value among N historical applications, ⁇ i is a recommendation weight, and ⁇ i is an association start probability.
  • the history application corresponding to the three historical contexts in which the mobile terminal determines that the similarity between the current situation and the current situation is the largest is the history application a, the history application b, and the history application c, respectively, and the recommended weight corresponding to the history application a is 0.5, the recommendation weight corresponding to the history application b is 0.3, and the recommendation weight corresponding to the history application c is 0.2.
  • the mobile terminal 120 may calculate that the association start probability of the history application a is 0.1 according to the association startup model, the association startup probability of the history application b is 0.3, and the association startup probability corresponding to the history application c is 0.8.
  • the method for loading a startup resource is that the mobile terminal does not directly determine a historical application corresponding to several historical contexts with the highest similarity in the current situation, and is determined as a prediction application, but a comprehensive analysis. Whether the current application that is already running is associated with launching other applications, and then these other applications that may be associated with the launch are determined to be predictive applications, thereby improving the accuracy of the determined predictive application and avoiding loading. A waste of resources caused by incorrect prediction of the application's resources.
  • the method may further include the following steps, as shown in FIG. 6:
  • step 601 the mobile terminal 120 determines an application that is actually launched in the current context.
  • Step 602 When the actually launched application is different from the predicted application, the mobile terminal 120 updates the weight corresponding to each context data type in the similarity model according to the current context data group and the actually launched application.
  • the mobile terminal can continuously update the weight in the similarity model according to the actual running result when the prediction application is predicted to have an error, thereby improving the determined similarity.
  • the accuracy of the model, as well as the accuracy of the resulting predictive application avoids wasting resources caused by loading errors in predicting the resources of the application.
  • the current context data in the current context data group acquired by the mobile terminal is usually the context data after the data pre-processing, and the foregoing step 201 is implemented as shown in FIG. step:
  • Step 701 The mobile terminal 120 collects an original context data group corresponding to the current context.
  • the original context data set may include at least one original context data, which may be used to describe the current running context. It can be understood that the mobile terminal 120 can determine terminal information such as its own terminal identifier, network information, device type, and running thread in the current running scenario, and use the terminal information as original context data, and/or record the current running situation.
  • the browsing data of the user of the mobile terminal 120, the download record, the personal information filled by the user, and the rating feedback filled by the user are used, and the running data is used as the original context data.
  • the mobile terminal 120 may collect the environment information in the current running context by using the sensor component, and use the environment information as the original context data
  • the sensor component may include any one or more of the following: a gravity sensor, Acceleration sensors, magnetic field sensors, gyroscopes, light sensors, distance sensors, GPS (Global Positioning System), fingerprint sensors, and temperature sensors.
  • the mobile terminal can collect the location through the GPS, and can collect the temperature through the temperature sensor.
  • the mobile terminal 120 may also retrieve raw context data from the server, such as obtaining temperature from the server.
  • Step 702 The mobile terminal 120 preprocesses the original context data of the original context data group by using a predetermined algorithm, and generates a current context data group corresponding to the current context.
  • the predetermined algorithm may be used to perform at least one of data cleaning, data integration, data specification, and data conversion on the original context data. among them:
  • the predetermined algorithm can be used to smooth out the original context data, fill in missing data, and delete data such as erroneous data.
  • the predetermined algorithm may include a Bayesian network algorithm.
  • the predetermined algorithm may be used for data integration of the original context data collected by the mobile terminal, where the mobile terminal 120 may collect multiple original context data belonging to the same context data type by using different methods, and the multiple original context data may be the same or different.
  • the mobile terminal can collect the temperature through the temperature sensor, or can obtain the temperature from the server, and the temperature collected by the mobile terminal through the temperature sensor and the temperature obtained from the server.
  • the difference may be within the error range, however, this may result in data redundancy of the original context data; in addition, the difference between the temperature collected by the mobile terminal through the temperature sensor and the temperature acquired from the server may also exceed the error range. In this way, the data of the original situation data may be inconsistent.
  • the predetermined algorithm includes other correlation analysis methods such as chi-square test, correlation coefficient analysis, and covariance analysis.
  • the predetermined algorithm can be used to perform data specification on the original context data to reduce the amount of original context data.
  • the predetermined algorithm includes a principal component analysis method, a wavelet transform, a sampling algorithm, and the like.
  • the predetermined algorithm can be used for data conversion of the original context data, thereby converting the original context data into a unified form of data, for example, different magnitudes of original context data can be converted into context data of the same order of magnitude, for example, each original context data is Converted to context data between [0, 1]; for example, the original situation data of different units can be converted into context data of the same unit, for example, the temperature represented by the Fahrenheit temperature value is converted to the temperature expressed by the Celsius temperature value.
  • Step 703 The mobile terminal 120 preprocesses the original context data in the original context data group by using a semantic analysis method to generate a current context data group corresponding to the current context.
  • the mobile terminal 120 may generate a current context data group after pre-processing each original context data in the original context data group by using a semantic analysis method, where the generating the current context data group may include: The original context data and/or the logically generated context data, for example, the user's mood is generated based on the type of music played by the user.
  • the mobile terminal 120 may perform at least one of the foregoing steps 702 and 703.
  • the foregoing steps 702 and 703 have no specific sequence, and Usually it is executed at the same time.
  • the historical context data group and the historical application corresponding to the historical context stored in the database of the mobile terminal 120 may be acquired by the mobile terminal 120 in the historical running context and stored in the database, and moved.
  • the method for the terminal 120 to obtain the historical context data group corresponding to the historical context may be combined with the embodiment shown in FIG. 7 , which is not described in this embodiment.
  • the mobile terminal 120 obtains a historical context data set of a large number of historical contexts.
  • the mobile The terminal 120 may first cluster the historical situation data groups of each historical situation, and then store the historical situation data groups corresponding to each historical situation class in a database, and then use the historical situation data group corresponding to each historical situation class to train.
  • a similarity model is obtained, and the similarity between the current situation and each historical situation class is calculated according to the similarity model.
  • the mobile terminal 120 may calculate a similarity between the respective historical contexts, and determine a historical context whose similarity is greater than a predetermined threshold, where the historical context belongs to the same historical context category, wherein the predetermined threshold may be a system preset. Value or user-defined value.
  • the method for calculating the similarity between the respective historical contexts by the mobile terminal 120 may be combined with the method for calculating the similarity between the current context and the historical context, which is not described in this embodiment.
  • the mobile terminal acquires 100 historical context data groups corresponding to the historical context. If no clustering is performed, the mobile terminal needs to separately calculate the similarity between the current context and the 100 historical contexts; if the mobile terminal pairs the 100 When the historical context is clustered, 10 historical situation classes can be clustered, so that the mobile terminal only needs to calculate the similarity between the current situation and the 10 historical situation classes, thereby reducing the processing capacity of the mobile terminal.
  • FIG. 8 is a schematic structural diagram of a boot resource loading apparatus provided in an embodiment of the present application.
  • the device can be implemented as a mobile terminal in the above method embodiment by software, hardware or a combination of both.
  • the device includes:
  • the first obtaining module 810 is configured to perform the foregoing step 201.
  • the second obtaining module 820 is configured to perform step 202 above.
  • the calculating module 830 is configured to perform step 203 above.
  • the determining module 840 is configured to perform the above step 204.
  • the loading module 850 is configured to perform the above step 205.
  • the calculation module 830 further includes the following units:
  • the calculating unit is configured to perform the above step 401.
  • a weighting unit is configured to perform the above step 402.
  • the determining module 840 further includes the following units:
  • the first determining unit is configured to perform the above step 501.
  • the obtaining unit is configured to perform step 502 above.
  • the second determining unit is configured to perform step 503 above.
  • the third determining unit is configured to perform the above step 504.
  • the first obtaining module 810 is further configured to perform the foregoing steps 701-step 703.
  • the device further comprises:
  • the third obtaining module is configured to perform step 301 above.
  • the model training module is configured to perform step 302 above.
  • the device further comprises:
  • An application determining module is configured to perform step 601 above.
  • An update module is configured to perform step 602 above.
  • the mobile terminal calculates the similarity between the current context and each historical context according to the similarity model, because the mobile terminal has a greater probability when in a similar situation.
  • the same application is launched, so the mobile terminal can predict the historical application corresponding to the historical situation with higher similarity as the predicted application that the mobile terminal is about to start, and preload the startup required when the predicted application starts. Resources to memory.
  • the mobile terminal receives the click operation of the user, the mobile terminal can directly use the pre-loaded startup resource to start the prediction application, so that the mobile terminal does not start loading the startup resource after receiving the user's click operation.
  • the time it takes for the mobile terminal to launch the application is reduced.
  • FIG. 9 is a block diagram of a mobile terminal 900 provided by an embodiment of the present application.
  • the mobile terminal may include a radio frequency (RF) circuit 901, a memory 902 including one or more computer readable storage media, and an input.
  • RF radio frequency
  • WiFi Wireless Fidelity
  • the terminal structure shown in FIG. 9 does not constitute a limitation to the terminal, and may include more or less components than those illustrated, or combine some components, or different component arrangements. among them:
  • the RF circuit 901 can be used for receiving and transmitting signals during and after receiving or transmitting information, in particular, after receiving downlink information of the base station, and processing it by one or more processors 908; in addition, transmitting data related to the uplink to the base station.
  • the RF circuit 901 includes, but is not limited to, an antenna, at least one amplifier, a tuner, one or more oscillators, a Subscriber Identity Module (SIM) card, a transceiver, a coupler, and a low noise amplifier (LNA, Low Noise Amplifier), duplexer, etc.
  • SIM Subscriber Identity Module
  • LNA Low Noise Amplifier
  • the wireless communication may use any communication standard or protocol, including but not limited to Global System of Mobile communication (GSM), General Packet Radio Service (GPRS), and Code Division Multiple Access (CDMA). , Code Division Multiple Access), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), e-mail, Short Messaging Service (SMS), and the like.
  • GSM Global System of Mobile communication
  • GPRS General Packet Radio Service
  • CDMA Code Division Multiple Access
  • WCDMA Wideband Code Division Multiple Access
  • LTE Long Term Evolution
  • SMS Short Messaging Service
  • the memory 902 can be used to store software programs and modules, and the processor 908 executes various functional applications and data processing by running software programs and modules stored in the memory 902.
  • the memory 902 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may be stored according to Data created by the use of the terminal (such as audio data, phone book, etc.).
  • memory 902 can include high speed random access memory, and can also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, memory 902 may also include a memory controller to provide access to memory 902 by processor 908 and input unit 903.
  • the input unit 903 can be configured to receive input numeric or character information and to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.
  • input unit 903 can include a touch-sensitive surface as well as other input devices.
  • Touch-sensitive surfaces also known as touch screens or trackpads, collect touch operations on or near the user (such as the user using a finger, stylus, etc., any suitable object or accessory on a touch-sensitive surface or touch-sensitive Operation near the surface), and drive the corresponding connecting device according to a preset program.
  • the touch sensitive surface can include two portions of a touch detection device and a touch controller.
  • the touch detection device detects the touch orientation of the user, and detects a signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts the touch information into contact coordinates, and sends the touch information.
  • the processor 908 is provided and can receive commands from the processor 908 and execute them.
  • touch-sensitive surfaces can be implemented in a variety of types, including resistive, capacitive, infrared, and surface acoustic waves.
  • the input unit 903 can also include other input devices. Specifically, other input devices may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, joysticks, and the like.
  • the display unit 904 can be used to display information input by the user or information provided to the user and various graphics of the terminal. User interface, these graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof.
  • the display unit 904 can include a display panel.
  • the display panel can be configured in the form of a Liquid Crystal Display (LCD), an Organic Light-Emitting Diode (OLED), or the like.
  • the touch-sensitive surface can cover the display panel, and when the touch-sensitive surface detects a touch operation thereon or nearby, it is transmitted to the processor 908 to determine the type of the touch event, and then the processor 908 displays the type according to the type of the touch event. A corresponding visual output is provided on the panel.
  • the touch-sensitive surface and display panel are implemented as two separate components to implement input and input functions, in some embodiments, the touch-sensitive surface can be integrated with the display panel to implement input and output functions.
  • the mobile terminal may also include at least one type of sensor 905, such as a light sensor, motion sensor, and other sensors.
  • the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel according to the brightness of the ambient light, and the proximity sensor may close the display panel and/or the backlight when the terminal moves to the ear.
  • the gravity acceleration sensor can detect the magnitude of acceleration in all directions (usually three axes). When it is stationary, it can detect the magnitude and direction of gravity.
  • the terminal can also be configured with gyroscopes, barometers, hygrometers, thermometers, infrared sensors and other sensors, no longer Narration.
  • the audio circuit 906, the speaker, and the microphone provide an audio interface between the user and the terminal.
  • the audio circuit 906 can transmit the converted electrical signal of the audio data to the speaker, and convert it into a sound signal output by the speaker; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 906 and then converted.
  • the audio data output processor 908 After the audio data is processed by the audio data output processor 908, it is sent to, for example, another terminal via the RF circuit 901, or the audio data is output to the memory 902 for further processing.
  • the audio circuit 906 may also include an earbud jack to provide communication between the peripheral earphone and the terminal.
  • WiFi is a short-range wireless transmission technology.
  • the terminal can help users to send and receive emails, browse web pages, and access streaming media through the WiFi module 907. It provides wireless broadband Internet access for users.
  • FIG. 9 shows the WiFi module 907, it can be understood that it does not belong to the necessary configuration of the terminal, and may be omitted as needed within the scope of not changing the essence of the invention.
  • the processor 908 is the control center of the terminal, which connects various portions of the entire handset using various interfaces and lines, by executing or executing software programs and/or modules stored in the memory 902, and invoking data stored in the memory 902, executing The various functions of the terminal and processing data to monitor the mobile phone as a whole.
  • the processor 908 can include one or more processing cores; in one embodiment, the processor 908 can integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system The user interface, application, etc., the modem processor primarily handles wireless communications. It will be appreciated that the above described modem processor may also not be integrated into the processor 908.
  • the terminal also includes a power supply 909 (such as a battery) that powers the various components.
  • the power supply can be logically coupled to the processor 908 through a power management system to manage charging, discharging, and power management through the power management system. And other functions.
  • the power supply 909 may also include any one or more of a DC or AC power source, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.
  • the terminal may further include a camera, a Bluetooth module, and the like, and details are not described herein again.
  • the processor 908 in the terminal runs one or more program instructions stored in the memory 902, thereby implementing the startup resource loading method provided in the foregoing various method embodiments.
  • the embodiment of the present application further provides a startup resource loading device.
  • the device may include:
  • the number of processors 1001 in the browser server may be one or more, and one processor is taken as an example in FIG.
  • the processor 1001, the memory 1002, the input device 1003, and the output device 1004 may be connected by a bus or other means, and the input device 1003 and the output device 1004 may be interfaces of the communication module, such as an interface of the GSM module. .
  • the bus connection is taken as an example.
  • the memory 1002 can be used to store software programs and modules, and the processor 1001 executes various functional applications and data processing of the browser server by running software programs and modules stored in the memory 1002.
  • the memory 1002 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function, and the like.
  • memory 1002 can include high speed random access memory, and can also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
  • the input device 1003 can be configured to receive input numeric or character information and to generate key signal inputs related to user settings and function control of the browser server.
  • the processor 1001 loads the executable file corresponding to the process of one or more application programs into the memory 1002 according to the following instructions, and is executed by the processor 1001 to be stored in the memory 1002.
  • the application to implement various functions:
  • the current context data group includes at least one current context data for describing a current running scenario
  • each historical situation data set includes at least one historical situation data for describing a historical running situation
  • the boot resource of the predictive application is preloaded into memory, which is the resource required to launch the predictive application.
  • the calculating, according to the current context data group and the historical context data group, the similarity between the current context and each of the historical contexts including:
  • the current context data set and each of the historical context data sets are input to a similarity model, and a similarity between the current context and each of the historical contexts is calculated.
  • the determining the historical situation that the similarity with the current situation meets the set condition includes:
  • N historical contexts having the greatest similarity to the current context are determined; wherein N is a positive integer.
  • the current context data includes at least one of current user context data, current environment context data, and current terminal context data;
  • the current user context data is used to describe information of a user who uses the mobile terminal when the mobile terminal is currently running, and the current user context data includes the user's name, gender, age, occupation, mood, and At least one of educational backgrounds;
  • the current environment context data is used to describe an environment in which the user of the mobile terminal is located when the mobile terminal is currently running, and the current environment context data includes time, location, weather conditions, temperature, illumination, sound, and At least one of traffic conditions;
  • the current terminal context data is used to describe information of the mobile terminal, and the current terminal context data includes at least one of a terminal identifier, network information, and a device type of the mobile terminal.
  • the current context data group includes at least two of the current context data
  • the historical context data group includes at least two of the historical context data
  • the calculating results in a similarity between the current context and each of the historical contexts including:
  • the sub-similarity is the current context data and the historical context data type of the same context data type
  • the similarity between the historical context data; the context data type includes: at least one of user context data, environmental context data, and terminal context data; and the sub-similarity corresponding to each of the current context data is according to the current
  • the weights corresponding to the context data types to which the context data belongs are added, and the similarity between the current context and the historical context is obtained.
  • the sub-similarity corresponding to each of the current context data is added according to the weight corresponding to the context data type to which the current context data belongs, to obtain the current context and the historical context. Similarities between, including:
  • C1 is the current context data set
  • C2 is the historical context data set
  • Sim (C1, C2) is a similarity between the current context and the historical context
  • n is the current context data set The number of current context data in the current context data
  • S i is the i-th current context data in the current context data group
  • p i is the weight corresponding to the context data type to which S i belongs
  • S j is the historical context the historical context data with the data set S i belong to the same context data type
  • Sim (S i, S j ) is the current context data corresponding to the sub-similarity
  • n ⁇ 2 and n is a positive integer.
  • the calculating the sub-similarity corresponding to each of the current context data in the current context data group includes:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data
  • S i is an i-th current context data in the current context data group
  • S j is the historical context data.
  • the calculating the sub-similarity corresponding to each of the current context data in the current context data group includes:
  • the sub-similarity is calculated according to the following formula:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data
  • S i is an i-th current context data in the current context data group
  • S j is the historical context data.
  • the minimum value, r i e is the maximum value of S i
  • r j s is the minimum value of S j
  • r j e is the maximum value of S j .
  • the calculating the sub-similarity corresponding to each of the current context data in the current context data group includes:
  • the sub-similarity is calculated according to the following formula:
  • Sim(S i , S j ) is a sub-similarity corresponding to the current context data
  • S i is an i-th current context data in the current context data group
  • S j is the historical context data.
  • the determining, by the historical application corresponding to the N historical contexts with the highest degree of similarity, the prediction application corresponding to the current context including:
  • association initiation probability Determining an association initiation probability of each of the historical applications relative to the current application, the association initiation probability being used to characterize a probability that the historical application is associatedly initiated during operation of the current application;
  • P history applications having the largest product of the recommendation weight and the corresponding association start probability are determined as the prediction application, and P is an integer greater than or equal to 1.
  • the acquiring the current context data group corresponding to the current context further includes:
  • the original context data is pre-processed to generate the current context data group corresponding to the current context.
  • the method before the inputting the current context data group and each of the historical context data groups into the similarity model, the method further includes:
  • each set of the historical context samples including the historical context data corresponding to each of the two historical contexts, and a context similarity score between the two historical contexts, the context
  • the similarity score is calibrated according to the similarity between the historical applications corresponding to the two historical contexts;
  • the initial similarity model is trained according to the historical situation sample to obtain the trained similarity model, and the initial similarity model includes at least one context data type and an initial weight corresponding to each context data type.
  • the method further includes:
  • the embodiment of the present application further provides a storage medium for storing program code, which is used to execute any one of the startup resource loading methods described in the foregoing embodiments.
  • the embodiment of the present application further provides a computer program product comprising instructions, when executed on a computer, causing the computer to execute any one of the startup resource loading methods described in the foregoing embodiments.
  • the startup resource loading device provided by the foregoing embodiment is only illustrated by the division of the foregoing functional modules when loading the startup resource.
  • the foregoing function may be allocated by different functional modules according to requirements.
  • the internal structure of the mobile terminal is divided into different functional modules to complete all or part of the functions described above.
  • the embodiment of the startup resource loading method and the startup resource loading device provided by the foregoing embodiments are in the same concept, and the specific implementation process is described in detail in the method embodiment, and details are not described herein again.
  • a person skilled in the art may understand that all or part of the steps of implementing the above embodiments may be completed by hardware, or may be instructed by a program to execute related hardware, and the program may be stored in a computer readable storage medium.
  • the storage medium mentioned may be a read only memory, a magnetic disk or an optical disk or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Stored Programmes (AREA)
  • Telephone Function (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种启动资源加载方法及装置,该方法包括:移动终端获取当前情境对应的当前情境数据组,该当前情境数据组包括用于描述该当前运行情境下的至少一个当前情境数据(201);移动终端获取多个历史情境各自对应的历史情境数据组,每个历史情境数据组包括用于描述该历史运行情境的至少一个历史情境数据(202);移动终端根据当前情境数据组和历史情境数据组,计算得到当前情境和每个历史情境之间的相似度(203);移动终端确定与当前情境的相似度符合设定条件的历史情境,根据历史情境对应的历史应用程序确定当前情境对应的预测应用程序(204);移动终端将预测应用程序的启动资源预加载至内存中,启动资源是启动预测应用程序时所需的资源(205)。由于移动终端会对即将启动预测应用程序进行预测,并预先加载预测应用程序启动时所需的启动资源至内存,当移动终端接收到用户的点击操作时,移动终端可以直接使用预先加载好的启动资源来启动该预测应用程序,减少了移动终端启动应用程序所耗费的时间。

Description

启动资源加载方法及装置
本申请要求于2016年12月19日提交中国专利局、申请号为201611184027.6、申请名称为“启动资源加载方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及移动终端领域,特别涉及一种启动资源加载方法及装置。
背景技术
随着移动终端的普及和发展,移动终端中的应用程序的数量和种类越来越丰富,用户可以通过移动终端安装和使用各种应用程序。
当用户在使用应用程序时,需要通过触发移动终端来启动该应用程序。现有技术中,用户在移动终端中点击应用程序的图标,移动终端接收到用户的点击操作后,为了启动该应用程序,需要开始加载该应用程序启动时所需的启动资源。其中,该启动资源包括:该应用程序的代码、该应用程序在运行时所需要的用户界面(User Interface,UI)素材、该应用程序在运行时需要处理的数据等。
对于游戏类应用程序、画图类应用程序、邮箱类应用程序等应用程序来讲,由于启动时需要加载的启动资源较多,因此,移动终端从接收到用户的点击操作至完全启动该应用程序,需要耗费较多的时间。
发明内容
为了解决移动终端在接收到用户的点击操作至应用程序完全启动耗费的时间较多的问题,本申请实施例提供了一种启动资源加载方法及装置。所述技术方案如下:
第一方面,本申请提供了一种启动资源加载方法,所述方法包括:
获取当前情境对应的当前情境数据组,所述当前情境数据组包括用于描述当前运行情境下的至少一个当前情境数据;
获取多个历史情境各自对应的历史情境数据组,每个历史情境数据组包括用于描述历史运行情境的至少一个历史情境数据;
根据所述当前情境数据组和所述历史情境数据组,计算得到所述当前情境和每个所述历史情境之间的相似度;
确定与当前情境的相似度符合设定条件的历史情境,根据所述历史情境对应的历史应用程序确定所述当前情境对应的预测应用程序,所述历史应用程序是在所述历史情境下运行的应用程序;
将所述预测应用程序的启动资源预加载至内存中,所述启动资源是启动所述预测应用程序时所需的资源。
在第一方面的一种可能的实现方式中,所述根据所述当前情境数据组和所述历史情境 数据组,计算得到所述当前情境和每个所述历史情境之间的相似度,包括:
将所述当前情境数据组和每个所述历史情境数据组输入相似度模型,计算所述当前情境和每个所述历史情境之间的相似度。
在第一方面的一种可能的实现方式中,所述确定与当前情境的相似度符合设定条件的历史情境包括:
确定与当前情境的相似度大于相似度阈值的历史情境;或者,
确定与当前情境的相似度最大的N个历史情境;其中,N为正整数。
在第一方面的一种可能的实现方式中,所述当前情境数据包括当前用户情境数据、当前环境情境数据和当前终端情境数据中的至少一种;
所述当前用户情境数据,用于描述在所述移动终端当前运行时,使用所述移动终端的用户的信息,所述当前用户情境数据包括所述用户的姓名、性别、年龄、职业、心情和教育背景中的至少一种;
所述当前环境情境数据,用于描述所述移动终端当前运行时,使用所述移动终端的用户所处的环境,所述当前环境情境数据包括时间、位置、天气情况、温度、光照、声音和交通情况中的至少一种;
所述当前终端情境数据,用于描述所述移动终端的信息,所述当前终端情境数据包括所述移动终端的终端标识、网络信息和设备类型中的至少一种。
在第一方面的一种可能的实现方式中,所述当前情境数据组包括至少两个所述当前情境数据,所述历史情境数据组包括至少两个所述历史情境数据;
所述计算得到所述当前情境和每个所述历史情境之间的相似度,包括:
计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,所述子相似度是所述当前情境数据和所述历史情境数据组中属于同一个情境数据类型的所述历史情境数据之间的相似度;所述情境数据类型包括:用户情境数据、环境情境数据和终端情境数据中的至少一种;将每个所述当前情境数据对应的子相似度按照所述当前情境数据所属的情境数据类型对应的权重相加,得到所述当前情境与所述历史情境之间的相似度。
在第一方面的一种可能的实现方式中,所述将每个所述当前情境数据对应的子相似度按照所述当前情境数据所属的情境数据类型对应的权重相加,得到所述当前情境与所述历史情境之间的相似度,包括:
Figure PCTCN2017107457-appb-000001
其中,C1是所述当前情境数据组,C2是所述历史情境数据组,Sim(C1,C2)是所述当前情境和所述历史情境之间的相似度;n是所述当前情境数据组中的当前情境数据的个数,Si是所述当前情境数据组中的第i个所述当前情境数据,pi是Si所属的情境数据类型对应的权重,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Sim(Si,Sj)是所述当前情境数据对应的子相似度,n≥2且n为正整数。
在第一方面的一种可能的实现方式中,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
当所述当前情境数据采用数值进行表示时,按照如下公式计算所述子相似度:
Figure PCTCN2017107457-appb-000002
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据。
在第一方面的一种可能的实现方式中,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
当所述当前情境数据采用区间进行表示时,按照如下公式计算所述子相似度:
Figure PCTCN2017107457-appb-000003
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Si∈[ri s,ri e]、Sj∈[rj s,rj e];ri s是Si的最小值,ri e是Si的最大值,rj s是Sj的最小值,rj e是Sj的最大值。
在第一方面的一种可能的实现方式中,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
当所述当前情境数据采用坐标进行表示时,按照如下公式计算所述子相似度:
Figure PCTCN2017107457-appb-000004
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Si=(xi,yi)、Sj=(xj,yj)。
在第一方面的一种可能的实现方式中,所述根据相似度最大的N个历史情境对应的历史应用程序确定所述当前情境对应的预测应用程序,包括:
根据所述N个历史情境各自对应的相似度,确定出所述历史情境对应的所述历史应用程序的推荐权重;
获取已经处于运行状态的当前应用程序;
确定每个所述历史应用程序相对于所述当前应用程序的关联启动概率,所述关联启动概率用于表征在所述当前应用程序的运行过程中,所述历史应用程序被关联启动的概率;
在所述N个历史应用程序中,将所述推荐权重和对应的所述关联启动概率的乘积最大的P个历史应用程序确定为所述预测应用程序,P为大于等于1的整数。
在第一方面的一种可能的实现方式中,所述获取当前情境对应的当前情境数据组,还包括:
采集所述当前情境对应的原始情境数据组,所述原始情境数据组包括用于描述所述当前运行情境下的至少一个原始情境数据;使用预定算法对所述原始情境数据组中的所述原始情境数据进行预处理,生成所述当前情境对应的所述当前情境数据组,所述预定算法用于对所述原始情境数据进行数据清理、数据集成、数据规约和数据转换中的至少一种;
和/或,
采集所述当前情境对应的原始情境数据组,所述原始情境数据组包括用于描述所述当前运行情境下的至少一个原始情境数据;使用语义分析方法对所述原始情境数据组中的所述原始情境数据进行预处理,生成所述当前情境对应的所述当前情境数据组。
在第一方面的一种可能的实现方式中,所述将所述当前情境数据组和每个所述历史情境数据组输入相似度模型之前,还包括:
获取至少一组历史情境样本,每组所述历史情境样本包括两个所述历史情境各自对应的所述历史情境数据,以及两个所述历史情境之间的情境相似度分值,所述情境相似度分值是根据两个所述历史情境对应的历史应用程序之间的相似度标定的;
根据所述历史情境样本对初始相似度模型进行训练,得到训练后的所述相似度模型,所述初始相似度模型中包括至少一个情境数据类型以及每个情境数据类型对应的初始权重。
在第一方面的一种可能的实现方式中,所述方法还包括:
确定所述当前情境下实际启动的应用程序;
当所述实际启动的应用程序与所述预测应用程序不同时,根据所述当前情境数据组与所述实际启动的应用程序更新所述相似度模型中的各个情境数据类型对应的所述权重。
第二方面,本申请提供了一种启动资源加载设备,所述设备包括处理器以及存储器:
所述存储器用于存储程序代码,并将所述程序代码传输给所述处理器;
所述处理器用于根据所述程序代码中的指令执行第一方面中任一项所述的启动资源加载方法。
第三方面,本申请提供了一种存储介质,所述存储介质用于存储程序代码,所述程序代码用于执行第一方面中任一项所述的启动资源加载方法。
第四方面,本申请提供了一种包括指令的计算机程序产品,当其在计算机上运行时,使得所述计算机执行第一方面中任一项所述的启动资源加载方法。
第五方面,本申请提供了一种启动资源加载方法,应用于移动终端,所述移动终端执行第一方面中任一项所述的启动资源加载方法。
本申请实施例提供的技术方案带来的有益效果是:
通过获取当前情境对应的当前情境数据组,并获取多个历史情境各自对应的历史情境数据组。然后,根据该当前情境数据组和该历史情境数据组,计算得到该当前情境和每个该历史情境之间的相似度。接着,确定与当前情境的相似度符合设定条件的历史情境,并 根据该历史情境对应的历史应用程序确定该当前情境对应的预测应用程序,其中,该历史应用程序是在该历史情境下运行的应用程序。最后,将该预测应用程序的启动资源预加载至内存中。由于移动终端在处于相似的情境下时有较大的概率会启动相同的应用程序,因此,该移动终端可以将相似度较高的历史情境对应的历史应用程序预测为移动终端即将启动的预测应用程序,并预先加载该预测应用程序启动时所需的启动资源至内存。这样,当移动终端接收到用户的点击操作时,移动终端就可以直接使用预先加载好的启动资源来启动该预测应用程序,避免了在接收到用户的点击操作后,移动终端才开始加载启动资源来启动该预测应用程序,从而减少了移动终端启动应用程序所耗费的时间。
附图说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本申请各个实施例涉及的一种实施环境的结构示意图;
图2是本申请一示例性实施例示出的一种启动资源加载方法的流程图;
图3是本申请另一示例性实施例示出的一种启动资源加载方法的流程图;
图4是本申请另一示例性实施例示出的一种启动资源加载方法的流程图;
图5是本申请另一示例性实施例示出的一种启动资源加载方法的流程图;
图6是本申请另一示例性实施例示出的一种启动资源加载方法的流程图;
图7是本申请另一示例性实施例示出的一种启动资源加载方法的流程图;
图8是本申请一示例性实施例示出的一种启动资源加载装置的框图;
图9是本申请一示例性实施例示出的一种移动终端的结构示意图;
图10是本申请一示例性实施例示出的一种启动资源加载设备的结构示意图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚,下面将结合附图对本申请实施方式作进一步地详细描述。
本申请各个实施例提供的启动资源加载方法,可以由诸如智能手机、平板电脑、电子书阅读器、动态影像专家压缩标准音频层面3(Moving Picture Experts Group Audio Layer III,MP3)播放器和动态影像专家压缩标准音频层面4(Moving Picture Experts Group Audio Layer IV,MP4)之类的移动终端来实现,且移动终端中安装有应用程序。
本申请各个实施例涉及的一种启动资源加载方法的实施环境还可以如图1所示,该实施环境中可以包括:移动终端120和服务器140。
移动终端120可以通过无线网络与服务器140相连。
服务器140可以是一台服务器、多台服务器组成的服务器集群或云计算中心,服务器140具有模型训练功能。
其中,上述无线网络可以是使用标准通信技术和/或协议。网络通常可以为因特网,但也可以是任何网络,包括但不限于局域网(Local Area Network,LAN)、城域网(Metropolitan Area Network,MAN)、广域网(Wide Area Network,WAN)、移动网络、有线网络或者无线网络、专用网络或者虚拟专用网络的任何组合。在一些实施例中,可以使用包括超文本标记语言(Hyper Text Mark-up Language,HTML)、可扩展标记语言(Extensible Markup Language,XML)等的技术和/或格式来代表通过网络交换的数据。此外还可以使用诸如安全套接字层(Secure Socket Layer,SSL)、传输层安全(Transport Layer Security,TLS)、虚拟专用网络(Virtual Private Network,VPN)、网际协议安全(Internet Protocol Security,IPsec)等常规加密技术来加密所有或者一些链路。在另一些实施例中,还可以使用定制和/或专用数据通信技术取代或者补充上述数据通信技术。
图2是本申请公开的一个示例性实施例所示出的启动资源加载方法的流程图。本实施例以该方法用于上述移动终端120中来举例说明。该方法包括:
步骤201,移动终端120获取当前情境对应的当前情境数据组,该当前情境数据组包括用于描述该当前运行情境下的至少一个当前情境数据。
在本实施例中,当前情境数据可以根据情境数据类型进行分类,其中该情境数据类型可以包括:用户情境数据、环境情境数据和终端情境数据中的至少一种。因此,当前情境数据可以包括当前用户情境数据、当前环境情境数据和当前终端情境数据中的至少一种。
其中,当前用户情境数据可以用于描述在移动终端当前运行时,使用该移动终端的用户的信息,比如,该当前用户情境数据可以包括用户的姓名、性别、年龄、职业、心情和教育背景中的至少一种;当前环境情境数据可以用于描述移动终端当前运行时,使用该移动终端的用户所处的环境,比如,该当前环境情境数据可以包括时间、位置、天气情况、温度、光照、声音和交通情况中的至少一种;当前终端情境数据可以用于描述移动终端在当前运行时的信息,比如该当前终端情境数据可以包括该移动终端的终端标识、网络信息和设备类型中的至少一种。
比如,一组示意性的当前情境数据组可以以集合的形式表示为:{终端标识,时间,性别,年龄,位置,温度}。
在一种实施方式中,当前情境数据可以采用量化值的形式和/或文本形式进行表示。比如,当当前情境数据包括性别时,当前情境数据可以是“男”;当当前情境数据包括温度时,当前情境数据可以是“20℃”。
当一个当前情境数据采用量化值的形式进行表示时,该当前情境数据可以采用数值、区间和坐标中的任意一种形式进行表示。比如,若当前情境数据中包括用户的年龄,且该用户的年龄是以数值的形式进行表示,则该当前情境数据所包括的年龄可以为“18”;再比如,若当前情境数据包括温度,且该温度是以区间的形式进行表示,则该当前情境数据所包括的温度可以为“18℃-25℃”;再比如,若当前情境数据包括的位置,且该位置是以坐标形式进行标识,则该当前情境数据所包括的位置可以为“(40°N,116°E)”。
步骤202,移动终端120获取多个历史情境各自对应的历史情境数据组,每个历史情境数据组包括用于描述该历史运行情境的至少一个历史情境数据。
其中,历史情境数据也可以根据情境数据类型进行分类,因此,历史情境数据可以包括历史用户情境数据、历史环境情境数据和历史终端情境数据中的至少一种。具体地,历史情境数据的情境数据类型和表示形式均可以结合参考上述步骤201中的当前情境数据,本实施例对此不再赘述。
在本实施例中,移动终端内的数据库中可以存储有各个历史情境对应的历史情境数据组。因此,移动终端可以获取数据库中的各个历史情境对应的历史情境数据组,或者,可以获取数据库中的部分历史情境对应的历史情境数据组。
需要说明的是,当移动终端获取数据库中的部分历史情境对应的历史情境数据组时,移动终端可以获取数据库中最近的M个历史情境对应的历史情境数据组,或者,移动终端可以随机获取M个历史情境对应的历史情境数据组,其中,M为正整数。
需要注意的是,在实际实现时,步骤202和步骤201没有特定的先后顺序,且通常可以是同时执行的。
步骤203,移动终端120根据根据当前情境数据组和历史情境数据组,计算得到当前情境和每个历史情境之间的相似度。
具体地,可以将当前情境数据组和每个历史情境数据组输入相似度模型,并根据该相似度模型计算得到该当前情境和每个历史情境之间的相似度。
其中,相似度模型可以是移动终端和/或服务器预先根据至少一组历史情境样本训练得到的模型,每组历史情境样本可以包括两个历史情境各自对应的历史情境数据,以及这两个历史情境之间的情境相似度分值,该情境相似度分值可以是根据这两个历史情境对应的历史应用程序之间的相似度标定的。
步骤204,移动终端120确定与当前情境的相似度符合设定条件的历史情境,根据该历史情境对应的历史应用程序确定当前情境对应的预测应用程序。
其中,历史应用程序是在历史情境下运行的应用程序。在一种实施方式中,数据库中除了保存有历史情境对应的历史情境数据组之外,还可以保存有历史情境对应的历史应用程序,这样,移动终端可以从数据库中获取每个历史情境所对应的历史应用程序。
在一种实现方式中,移动终端可以先确定相似度最大的N个历史情境,接着,将该N个历史情境对应的历史应用程序作为预测应用程序。其中,N的取值可以由系统预设或用户自定义,需要说明的是,本实施例对N的取值方式不做限定。
在其他实现方式中,移动终端可以先确定与当前情境之间的相似度大于相似度阈值的历史情境,接着,将该历史情境对应的历史应用程序作为预测应用程序。其中,该相似度阈值可以是系统预设值或用户自定义值,需要说明的是,本实施例对该相似度阈值的设置方式不做限定。
在一个示例性的例子中,历史情境1-历史情境5分别与当前情境之间的相似度可以如下表一所示:
表一
Figure PCTCN2017107457-appb-000005
假设N=2,则移动终端确定的预测应用程序包括历史情境1对应的历史应用程序以及历史情境4对应的历史应用程序。再比如,假设相似度阈值为0.5,则移动终端确定的预测应用程序可以包括历史情境1对应的历史应用程序、历史情境3对应的历史应用程序,以及历史情境4对应的历史应用程序。
步骤205,移动终端120将预测应用程序的启动资源预加载至内存中,启动资源是启动预测应用程序时所需的资源。
综上所述,本申请实施例提供的启动资源加载方法,通过获取当前情境对应的当前情境数据组,以及M个历史情境各自对应的历史情境数据组,并根据该当前情境数据组和该历史情境数据组,得到该当前情境和每个所述历史情境之间的相似度。接着,根据相似度最大的N个历史情境对应的历史应用程序确定该当前情境对应的预测应用程序。最后,将该预测应用程序的启动资源预加载至内存中。由于移动终端在处于相似的情境下时有较大的概率会启动相同的应用程序,因此,该移动终端可以将相似度较高的历史情境对应的历史应用程序预测为移动终端即将启动的预测应用程序,并预先加载该预测应用程序启动时所需的启动资源至内存。这样,当移动终端接收到用户的点击操作时,移动终端就可以直接使用预先加载好的启动资源来启动该预测应用程序,避免了在接收到用户的点击操作后,移动终端才开始加载启动资源来启动该预测应用程序,从而减少了移动终端启动应用程序所耗费的时间。
进一步地,如图3所示,在本申请实施例的一种实施方式中,移动终端在使用相似度模型计算当前情境与历史情境之间的相似度之前,该方法还可以包括模型训练的过程,具体可以包括如下几个步骤:
步骤301,移动终端120获取至少一组历史情境样本。
其中,每组历史情境样本可以包括两个历史情境各自对应的历史情境数据,以及这两个历史情境之间的情境相似度分值,该情境相似度分值可以是根据这两个历史情境对应的历史应用程序之间的相似度标定的。其中,该情境相似度分值与历史应用程序之间的相似度之间的关系可以是系统预设的,比如,该情境相似度分值与历史应用程序之间的相似度可以是成正比关系的,即该情境相似度分值越高,则历史应用程序之间的相似度越高,反之,该情境相似度分值越低,则历史应用程序之间的相似度越低。
在一种实施方式中,两个历史应用程序之间的相似度可以通过计算这两个应用程序各自所对应的标签之间的相似度确定,例如,获取到两个应用程序各自所对应的标签后,可以先确定这两个历史应用程序各自对应的标签中相同的标签,然后,计算该相同的标签在这两个历史应用程序对应的标签中所占的比例,接着,可以根据该比例确定这两个历史应用程序之间的相似度。
比如,假设相似度分值的取值范围为[0,1],当两个历史情境对应同一个历史应用程序 时,则两个历史情境之间的情境相似度分值为1;假设历史情境a对应历史应用程序a,历史应用程序a对应的标签包括美食、团购和手机,历史情境b对应历史应用程序b,历史应用程序b对应的标签包括手机、服装和KTV,由于历史应用程序a对应的标签和历史应用程序b对应的标签只有一个相同的标签即手机,则历史应用程序a与历史应用程序b之间的相似度可以为1/3=0.33,历史情境a和历史情境b之间的相似度可以等于历史应用程序a与历史应用程序b之间的相似度0.33。
步骤302,移动终端120根据历史情境样本对初始相似度模型进行训练,得到训练后的相似度模型。
其中,初始相似度模型中包括至少一个情境数据类型,以及每个情境数据类型对应的初始权重,其中,初始权重是系统预设或用户自定义的权重。一个可能的实现方式为,初始相似度模型中的每个情境数据类型对应初始权重均相等,比如,当初始相似度模型中有4个情境数据类型时,每个情境数据类型对应的初始权重均为0.25。
需要说明的是,初始相似度模型可以是人工神经网络、Adaboost、支持向量机(Support Vector Machine,SVM)、遗传算法和朴素贝叶斯模型中的任意一种模型。其中,人工神经网络可以是基于区域候选的深度卷积网络(Region-based Convolutional Neural Networks,R-CNN)、Fast R-CNN学习算法或Faster R-CNN学习算法。
当本实施例的实施环境如图1所示时,为了减少移动终端在训练相似度模型时的数据处理量,移动终端可以将各个历史情境样本发送给具有模型训练能力的服务器,由服务器根据历史情境样本训练得到相似度模型。
需要说明的是,步骤301和步骤302所涉及的模型训练过程,通常是在上述步骤201之前执行的,但在实际实现时,也可以是与步骤201和步骤202同时执行的,本实施例对此不做限定。
在实际实现时,移动终端所处的运行情境通常会包括大量的情境数据,则移动终端获取到的当前情境数据组可以包括至少两个当前情境数据,相应地,移动终端获取到的历史情境数据组中也可以包括至少两个历史情境数据,因此,在本申请实施例的一种实施方式中,如图4所示,上述步骤203可以包括如下几个步骤:
步骤401,移动终端120计算当前情境数据组中的每个当前情境数据对应的子相似度。
其中,当前情境数据对应的子相似度可以是当前情境数据和历史情境数据组中属于同一个情境数据类型的历史情境数据之间的相似度,情境数据类型的含义可以参见上述步骤201的相关介绍,本实施例对此不再赘述。比如,当前情境数据是位置,则与当前情境数据属于同一个情境数据类型的历史情境数据也为位置,当前情境数据对应的子相似度是当前情境数据“位置”与历史情境数据“位置”之间的相似度。
当当前情境数据以不同的形式进行表示时,移动终端120在计算当前情境数据对应的子相似度时使用的方法也不同,如下所示:
在第一种可能的实现方式中,当当前情境数据采用数值进行表示时,则移动终端120可以按照如下公式计算子相似度:
Figure PCTCN2017107457-appb-000006
其中,Sim(Si,Sj)是当前情境数据对应的子相似度;Si是当前情境数据组中的第i个当前情境数据;Sj是历史情境数据组中与Si属于同一个情境数据类型的历史情境数据;1≤i≤n,n是当前情境数据组中的当前情境数据的个数。
比如,当前情境数据“温度”为20℃,历史情境数据“温度”为25℃,则移动终端120计算当前情境数据“温度”对应的子相似度为1-|20℃-25℃|/25℃=0.8。
在第二种可能的实现方式中,当前情境数据采用区间进行表示时,则移动终端120可以按照如下公式计算子相似度:
Figure PCTCN2017107457-appb-000007
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度;Si是所述当前情境数据组中的第i个所述当前情境数据;Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据;Si∈[ri s,ri e]、Sj∈[rj s,rj e],ri s是Si表示的当前情境数据的最小值;ri e是Si表示的当前情境数据的最大值;rj s是Sj表示的历史情境数据的最小值;rj e是Sj表示的历史情境数据的最大值。
比如,当前情境数据“温度”为18℃-25℃,则ri s=18℃,ri e=25℃;历史情境数据“温度”为20℃-30℃,则rj s=20℃,rj e=30℃,移动终端120计算当前情境数据“温度”对应的子相似度为1-|18℃-20℃|/2*20℃-|25℃-30℃|/2*30℃≈0.867。
在第三种可能的实现方式中,当前情境数据采用坐标进行表示时,则移动终端120可以按照如下公式计算子相似度:
Figure PCTCN2017107457-appb-000008
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度;Si是所述当前情境数据组中的第i个所述当前情境数据;Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据;Si=(xi,yi)、Sj=(xj,yj)。
需要说明的是,当当前情境数据采用文本形式进行表示时,也可以将当前情境数据量化为以量化值的形式进行表示,再计算当前情境数据对应的子相似度。比如,当前情境数据是性别时,则可以将“男”量化为1,将“女”量化为0,本实施例对此不做限定,但本申请实施例将以量化值的形式进行表示的当前情境数据为例进行解释。
步骤402,移动终端120将每个当前情境数据对应的子相似度按照当前情境数据所属的情境数据类型对应的权重相加,得到当前情境与历史情境之间的相似度。
在一种实施方式中,移动终端120可以按照如下公式计算相似度:
Figure PCTCN2017107457-appb-000009
其中,C1是当前情境数据组,C2是历史情境数据组;Sim(C1,C2)是当前情境和历史情境之间的相似度;n是该当前情境数据组中的当前情境数据的个数,n≥2且n为正整数;Si是该当前情境数据组中的第i个所述当前情境数据;Sj是该历史情境数据组中与Si属于同一个情境数据类型的该历史情境数据;pi是Si所属的情境数据类型对应的权重,pi 的取值是在训练得到相似度模型的过程中训练得到的,Σpi=1,当前情境数据不同时;Sim(Si,Sj)是该当前情境数据对应的子相似度,Sim(Si,Sj)可以是使用上述三种可能的实现方式中的任意一种可能的实现方式计算得到的。
比如,C1和C2分别包括声音、温度和位置,C1={声音1,温度1,位置1},C2={声音2,温度2,位置2},其中,C1和C2中的声音均是以数值形式进行表示,温度以区间形式进行表示,位置以坐标形式进行表示,则移动终端120可以使用上述步骤401中的第一种可能的实现方式计算C1中的声音1对应的子相似度Sim(声音),假设Sim(声音)=0.7;移动终端120可以使用上述第二种可能的实现方式计算C1中的温度1对应的子相似度Sim(温度),假设Sim(温度)=0.9;移动终端120可以使用上述第三种可能的实现方式计算C1中的位置1对应的子相似度Sim(位置),假设Sim(位置)=0.4。具体的计算方式可以结合上述步骤401中的举例,本实施例对此不再赘述。
假设训练得到的相似度模型为Sim(C1,C2)=0.2*Sim(声音)+0.2*Sim(温度)+0.6*Sim(位置)=0.2*0.7+0.2*0.9+0.6*0.4=0.56。
需要说明的是,上述示例性的例子是以当前情境数据组中的每一个当前情境数据都存在与其属于同一个情境数据类型的历史情境数据为例,在实际实现时,对于某一个当前情境数据,历史情境数据组中可能不存在与其属于同一个情境数据类型的历史情境数据,比如在上述示例性的例子中,C2={声音2,温度2},C2中并不包括位置,则移动终端120可以不计算位置这个当前情境数据对应的子相似度。在本申请实施例中,可以以各个历史情境对应的历史情境数据中包括的数据类型都相同来举例说明,但对此不构成限定。
在实际实现时,移动终端120处于当前情境启动预测应用程序时,可能会因为突发事件的打断而改为启动其他应用程序,比如,移动终端120在将要启动播放类应用程序时,在后台运行的即时通信类客户端接收到即时通信消息,则用户可能会在回复即时通信消息的过程中打开图片类应用程序,则在基于上述各个实施例的其他可选实施例中,移动终端120可以并不直接将相似度最大的N个历史情境对应的历史应用程序确定为预测应用程序,则如图5所示,上述步骤204可以包括如下几个步骤:
步骤501,移动终端120根据N个历史情境各自对应的相似度,确定出历史情境对应的历史应用程序的推荐权重。
其中,推荐权重的取值范围可以为[0,1],推荐权重与历史情境和当前情境之间的相似度可以成正比,可以理解的是,历史情境与当前情境之间的相似度越大,该历史情境对应的历史应用程序的推荐权重越大,而相似度与推荐权重之间的对应关系可以是系统预设或用户自定义的。
为了减少移动终端120的数据处理量,移动终端120确定与当前情境之间的相似度最大的N个历史情境对应的历史应用程序的推荐权重,在实际实现时,移动终端120也可以确定每个历史情境对应的历史应用程序的推荐权重。
步骤502,移动终端120获取已经处于运行状态的当前应用程序。
其中,处于运行状态的当前应用程序可以包括在移动终端的前台运行的应用程序和在 移动终端的后台运行的应用程序。
需要强调的是,在实际实现时,步骤502与步骤501没有特定的先后顺序,且这两个步骤通常是同时运行的。
步骤503,移动终端120确定每个历史应用程序相对于当前应用程序的关联启动概率。
由于移动终端120在运行一个应用程序的过程中,可能会关联启动其他的应用程序,比如,应用程序在运行购物类应用程序的过程中,通常会触发启动支付类应用程序,则购物类应用程序与支付类应用程序有很大的概率会关联启动。
因此,移动终端可以采集应用程序之间互相关联启动的数据作为关联启动样本,并根据预定训练算法对关联启动样本训练得到关联启动模型。这样,就可以根据关联启动模型确定每个历史应用程序的关联启动概率。
其中,关联启动概率可以用于表征在当前应用程序的运行过程中,历史应用程序被关联启动的概率,该关联启动概率的取值范围可以为[0,1],可以理解的是,历史应用程序被当前应用程序关联启动的概率越大,该历史应用程序相对于当前应用程序的关联启动概率越大。
同理,在确定与当前情境之间的相似度最大的N个历史情境对应的历史应用程序之后,移动终端120可以确定这些历史应用程序相对于当前应用程序的关联启动概率,或者确定移动终端包括的各个应用程序相对于当前应用程序的关联启动概率,预定训练算法的含义可以参见上述步骤302的相关介绍,本实施例对此不再赘述。
步骤504,移动终端120在N个历史应用程序中,将推荐权重和对应的关联启动概率的乘积最大的P个历史应用程序确定为预测应用程序,P可以为大于或等于1的整数。
需要说明的是,当P=1时,移动终端可以通过计算
Figure PCTCN2017107457-appb-000010
确定预测应用程序,其中,Q是N个历史应用程序中使ωii取最大值时的历史应用程序,ωi是推荐权重,λi是关联启动概率。
比如,移动终端确定与当前情境之间的相似度最大的3个历史情境对应的历史应用程序分别为历史应用程序a、历史应用程序b和历史应用程序c,历史应用程序a对应的推荐权重为0.5,历史应用程序b对应的推荐权重为0.3,历史应用程序c对应的推荐权重为0.2。移动终端120可以根据关联启动模型计算得到历史应用程序a的关联启动概率为0.1,历史应用程序b的关联启动概率为0.3,历史应用程序c对应的关联启动概率为0.8。则移动终端120可以计算得到历史应用程序a对应的推荐权重和关联启动概率的乘积为0.05,历史应用程序b对应的推荐权重和关联启动概率的乘积为0.09,历史应用程序c对应的推荐权重和关联启动概率的乘积为0.16,假设P=1,则移动终端最终确定历史应用程序c为预测应用程序。
综上所述,本申请实施例提供的启动资源加载方法,由于移动终端并不是直接将与当前情境相似度最高的若干个历史情境对应的历史应用程序,确定为预测应用程序,而是综合分析已经处于运行状态的当前应用程序是否会关联启动其他应用程序,接着,将这些可能会关联启动的其他应用程序确定为预测应用程序,从而提高了确定得到的预测应用程序的准确性,避免了加载错误的预测应用程序的资源时引起的资源浪费。
在一种实施方式中,在基于上述实施例的其他可选实施例中,该方法还可以包括如下几个步骤,如图6所示:
步骤601,移动终端120确定当前情境下实际启动的应用程序。
步骤602,当实际启动的应用程序与预测应用程序不同时,移动终端120根据当前情境数据组与实际启动的应用程序更新相似度模型中的各个情境数据类型对应的权重。
综上所述,本申请实施例提供的启动资源加载方法,移动终端可以在预测到预测应用程序出现错误时,根据实际运行结果不断更新相似度模型中的权重,从而提高了确定得到的相似度模型的准确性,以及确定得到的预测应用程序的准确性,进而避免了加载错误的预测应用程序的资源时引起的资源浪费。
在上述各个实施例中,移动终端获取到的当前情境数据组中当前情境数据通常是经过数据预处理后的情境数据,则上述步骤201在实现时,如图7所示,可以包括如下几个步骤:
步骤701,移动终端120采集当前情境对应的原始情境数据组。
其中,原始情境数据组可以包括至少一个原始情境数据,该原始情境数据可以用于描述当前运行情境。可以理解的是,移动终端120可以确定当前运行情境下自身的终端标识、网络信息、设备类型和运行线程等终端信息,并将该终端信息作为原始情境数据,和/或,记录当前运行情境下使用该移动终端120的用户的浏览记录、下载记录、该用户填写的个人信息和该用户填写的评分反馈等运行数据,并将该运行数据作为原始情境数据。
在一种实施方式中,移动终端120可以通过传感器组件采集当前运行情境下的环境信息,并将该环境信息作为原始情境数据,其中,传感器组件可以包括以下任意一种或多种:重力传感器、加速度传感器、磁场传感器、陀螺仪、光线传感器、距离传感器、GPS(Global Positioning System,全球定位系统)、指纹传感器和温度传感器等。比如,移动终端可以通过GPS采集位置,可以通过温度传感器采集温度等。
在一种实施方式中,移动终端120也可以从服务器中获取原始情境数据,比如,从服务器中获取温度。
步骤702,移动终端120使用预定算法对原始情境数据组的原始情境数据进行预处理,生成当前情境对应的当前情境数据组。
其中,预定算法可以用于对原始情境数据进行数据清理、数据集成、数据规约和数据转换中的至少一种。其中:
预定算法可以用于对原始情境数据进行平滑噪声数据、填充缺失数据,以及删除错误数据等数据清理。在一种实施方式中,预定算法可以包括贝叶斯网络算法。
预定算法可以用于对移动终端采集到的原始情境数据进行数据集成,其中,移动终端120可以使用不同方法采集到属于同一情境数据类型的多个原始情境数据,该多个原始情境数据可以相同或者不同。比如,移动终端可以通过温度传感器采集到温度,也可以从服务器中获取到温度,移动终端通过温度传感器采集到的温度与从服务器中获取到的温度的 差值可以是在误差范围内,但是,这样可能会导致原始情境数据的数据冗余;另外,移动终端通过温度传感器采集到的温度与从服务器中获取到的温度的差值也可能超过误差范围,这样,可能会导致原始情境数据的数据不一致。为了解决数据冗余和数据不一致的问题,在一种实施方式中,预定算法包括诸如卡方检验、相关系数分析和协方差分析等其他相关分析法。
预定算法可以用于对原始情境数据进行数据规约,以减少原始情境数据的数量。在一种实施方式中,预定算法包括主成分分析法、小波变换和抽样算法等。
预定算法可以用于对原始情境数据进行数据转换,从而将原始情境数据转换成统一形式的数据,比如,可以将不同数量级的原始情境数据转换为同一数量级的情境数据,例如将各个原始情境数据都转换为[0,1]之间的情境数据;再比如,可以将不同单位的原始情境数据转换为同一单位的情境数据,例如将华氏温度数值表示的温度转换为摄氏温度数值表示的温度。
步骤703,移动终端120使用语义分析方法对原始情境数据组中的原始情境数据进行预处理,生成当前情境对应的当前情境数据组。
在本实施例中,移动终端120可以在使用语义分析方法对原始情境数据组中的各个原始情境数据进行预处理后,生成当前情境数据组,其中,该生成当前情境数据组可以包括:处理后的原始情境数据和/或逻辑推理生成的情境数据,比如,根据用户播放的音乐类型推理生成用户的心情。
实际实现时,移动终端120可以执行上述步骤702和步骤703中的至少一种,当移动终端120既执行上述步骤702又执行上述步骤703时,上述步骤702和步骤703没有特定的先后顺序,且通常情况下是同时执行的。
另外,在上述各个实施例中,移动终端120的数据库中存储的历史情境对应的历史情境数据组和历史应用程序,可以是移动终端120在历史运行情境下获取到并存储在数据库中的,移动终端120获取历史情境对应的历史情境数据组的方法可以结合上述图7所示的实施例,本实施例对此不再赘述。
在实际实现时,移动终端120会获取到大量历史情境的历史情境数据组,为了减少移动终端120在计算当前情境和历史情境之间的相似度时的运算量,在一种实施方式中,移动终端120可以先对各个历史情境的历史情境数据组进行聚类,然后,将每一个历史情境类对应的历史情境数据组存储在数据库中,接着,使用各个历史情境类对应的历史情境数据组训练得到相似度模型,并根据相似度模型计算当前情境与各个历史情境类之间的相似度。
具体的,移动终端120可以计算各个历史情境之间的相似度,并确定相似度大于预定阈值的历史情境,该历史情境与当前情境属于同一个历史情境类,其中,预定阈值可以是系统预设值或用户自定义值。移动终端120计算各个历史情境之间的相似度的方法可以结合上述实施例示出的计算当前情境与历史情境之间的相似度的方法,本实施例对此不再赘述。
比如,移动终端获取到100个历史情境对应的历史情境数据组,若不进行聚类,则移动终端需要分别计算当前情境与这100个历史情境之间的相似度;若移动终端对这100个历史情境进行聚类,则可以聚类得到10个历史情境类,这样,移动终端只需要分别计算当前情境与这10个历史情境类之间的相似度,从而减少了移动终端的运算处理量。
请参考图8,其示出了本申请一个实施例中提供的启动资源加载装置的结构示意图。该装置可以通过软件、硬件或两者的结合实现成为上述方法实施例中的移动终端。该装置包括:
第一获取模块810,用于执行上述步骤201。
第二获取模块820,用于执行上述步骤202。
计算模块830,用于执行上述步骤203。
确定模块840,用于执行上述步骤204。
加载模块850,用于执行上述步骤205。
在一种实施方式中,计算模块830还包括如下几个单元:
计算单元,用于执行上述步骤401。
加权单元,用于执行上述步骤402。
在一种实施方式中,确定模块840还包括如下几个单元:
第一确定单元,用于执行上述步骤501。
获取单元,用于执行上述步骤502。
第二确定单元,用于执行上述步骤503。
第三确定单元,用于执行上述步骤504。
在一种实施方式中,第一获取模块810还用于执行上述步骤701-步骤703。
在一种实施方式中,该装置还包括:
第三获取模块,用于执行上述步骤301。
模型训练模块,用于执行上述步骤302。
在一种实施方式中,该装置还包括:
应用程序确定模块,用于执行上述步骤601。
更新模块,用于执行上述步骤602。
综上所述,本申请实施例提供的启动资源加载装置,移动终端根据相似度模型计算当前情境与各个历史情境之间的相似度,由于移动终端在处于相似的情境下时有较大的概率会启动相同的应用程序,因此,该移动终端可以将相似度较高的历史情境对应的历史应用程序预测为移动终端即将启动的预测应用程序,并预先加载该预测应用程序启动时所需的启动资源至内存。这样,当移动终端接收到用户的点击操作时,移动终端就可以直接使用预先加载好的启动资源来启动该预测应用程序,避免了在接收到用户的点击操作后,移动终端才开始加载启动资源来启动该预测应用程序,减少了移动终端启动应用程序所耗费的时间。
图9其示出了本申请一个实施例提供的移动终端900的框图,该移动终端可以包括射频(RF,Radio Frequency)电路901、包括有一个或一个以上计算机可读存储介质的存储器902、输入单元903、显示单元904、传感器905、音频电路906、无线保真(WiFi,Wireless Fidelity)模块907、包括有一个或者一个以上处理核心的处理器908、以及电源909等部件。本领域技术人员可以理解,图9中示出的终端结构并不构成对终端的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置。其中:
RF电路901可用于收发信息或通话过程中,信号的接收和发送,特别地,将基站的下行信息接收后,交由一个或者一个以上处理器908处理;另外,将涉及上行的数据发送给基站。通常,RF电路901包括但不限于天线、至少一个放大器、调谐器、一个或多个振荡器、用户身份模块(SIM,Subscriber Identity Module)卡、收发信机、耦合器、低噪声放大器(LNA,Low Noise Amplifier)、双工器等。此外,RF电路901还可以通过无线通信与网络和其他设备通信。所述无线通信可以使用任一通信标准或协议,包括但不限于全球移动通讯系统(GSM,Global System of Mobile communication)、通用分组无线服务(GPRS,General Packet Radio Service)、码分多址(CDMA,Code Division Multiple Access)、宽带码分多址(WCDMA,Wideband Code Division Multiple Access)、长期演进(LTE,Long Term Evolution)、电子邮件、短消息服务(SMS,Short Messaging Service)等。
存储器902可用于存储软件程序以及模块,处理器908通过运行存储在存储器902的软件程序以及模块,从而执行各种功能应用以及数据处理。存储器902可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等;存储数据区可存储根据终端的使用所创建的数据(比如音频数据、电话本等)等。此外,存储器902可以包括高速随机存取存储器,还可以包括非易失性存储器,例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。相应地,存储器902还可以包括存储器控制器,以提供处理器908和输入单元903对存储器902的访问。
输入单元903可用于接收输入的数字或字符信息,以及产生与用户设置以及功能控制有关的键盘、鼠标、操作杆、光学或者轨迹球信号输入。具体地,在一个具体的实施例中,输入单元903可包括触敏表面以及其他输入设备。触敏表面,也称为触摸显示屏或者触控板,可收集用户在其上或附近的触摸操作(比如用户使用手指、触笔等任何适合的物体或附件在触敏表面上或在触敏表面附近的操作),并根据预先设定的程式驱动相应的连接装置。在一种实施方式中,触敏表面可包括触摸检测装置和触摸控制器两个部分。其中,触摸检测装置检测用户的触摸方位,并检测触摸操作带来的信号,将信号传送给触摸控制器;触摸控制器从触摸检测装置上接收触摸信息,并将它转换成触点坐标,再送给处理器908,并能接收处理器908发来的命令并加以执行。此外,可以采用电阻式、电容式、红外线以及表面声波等多种类型实现触敏表面。除了触敏表面,输入单元903还可以包括其他输入设备。具体地,其他输入设备可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆等中的一种或多种。
显示单元904可用于显示由用户输入的信息或提供给用户的信息以及终端的各种图形 用户接口,这些图形用户接口可以由图形、文本、图标、视频和其任意组合来构成。显示单元904可包括显示面板,在一种实施方式中,可以采用液晶显示器(LCD,Liquid Crystal Display)、有机发光二极管(OLED,Organic Light-Emitting Diode)等形式来配置显示面板。进一步的,触敏表面可覆盖显示面板,当触敏表面检测到在其上或附近的触摸操作后,传送给处理器908以确定触摸事件的类型,随后处理器908根据触摸事件的类型在显示面板上提供相应的视觉输出。虽然在图9中,触敏表面与显示面板是作为两个独立的部件来实现输入和输入功能,但是在某些实施例中,可以将触敏表面与显示面板集成而实现输入和输出功能。
移动终端还可包括至少一种传感器905,比如光传感器、运动传感器以及其他传感器。具体地,光传感器可包括环境光传感器及接近传感器,其中,环境光传感器可根据环境光线的明暗来调节显示面板的亮度,接近传感器可在终端移动到耳边时,关闭显示面板和/或背光。作为运动传感器的一种,重力加速度传感器可检测各个方向上(一般为三轴)加速度的大小,静止时可检测出重力的大小及方向,可用于识别手机姿态的应用(比如横竖屏切换、相关游戏、磁力计姿态校准)、振动识别相关功能(比如计步器、敲击)等;至于终端还可配置的陀螺仪、气压计、湿度计、温度计、红外线传感器等其他传感器,在此不再赘述。
音频电路906、扬声器,传声器可提供用户与终端之间的音频接口。音频电路906可将接收到的音频数据转换后的电信号,传输到扬声器,由扬声器转换为声音信号输出;另一方面,传声器将收集的声音信号转换为电信号,由音频电路906接收后转换为音频数据,再将音频数据输出处理器908处理后,经RF电路901以发送给比如另一终端,或者将音频数据输出至存储器902以便进一步处理。音频电路906还可能包括耳塞插孔,以提供外设耳机与终端的通信。
WiFi属于短距离无线传输技术,终端通过WiFi模块907可以帮助用户收发电子邮件、浏览网页和访问流式媒体等,它为用户提供了无线的宽带互联网访问。虽然图9示出了WiFi模块907,但是可以理解的是,其并不属于终端的必须构成,完全可以根据需要在不改变发明的本质的范围内而省略。
处理器908是终端的控制中心,利用各种接口和线路连接整个手机的各个部分,通过运行或执行存储在存储器902内的软件程序和/或模块,以及调用存储在存储器902内的数据,执行终端的各种功能和处理数据,从而对手机进行整体监控。在一种实施方式中,处理器908可包括一个或多个处理核心;在一种实施方式中,处理器908可集成应用处理器和调制解调处理器,其中,应用处理器主要处理操作系统、用户界面和应用程序等,调制解调处理器主要处理无线通信。可以理解的是,上述调制解调处理器也可以不集成到处理器908中。
终端还包括给各个部件供电的电源909(比如电池),在一种实施方式中,电源可以通过电源管理系统与处理器908逻辑相连,从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。电源909还可以包括一个或一个以上的直流或交流电源、再充电系统、电源故障检测电路、电源转换器或者逆变器、电源状态指示器等任意组件。
尽管未示出,终端还可以包括摄像头、蓝牙模块等,在此不再赘述。具体在本实施例中,终端中的处理器908会运行存储在存储器902中的一个或一个以上的程序指令,从而实现上述各个方法实施例中所提供的启动资源加载方法。
本申请实施例还提供一种启动资源加载设备,参见图10所示,所述设备可以包括:
处理器1001、存储器1002、输入装置1003和输出装置1004。浏览器服务器中的处理器1001的数量可以一个或多个,图8中以一个处理器为例。在本发明的一些实施例中,处理器1001、存储器1002、输入装置1003和输出装置1004可通过总线或其它方式连接,输入装置1003和输出装置1004可以为通信模块的接口,如GSM模块的接口。其中,图8中以通过总线连接为例。
存储器1002可用于存储软件程序以及模块,处理器1001通过运行存储在存储器1002的软件程序以及模块,从而执行浏览器服务器的各种功能应用以及数据处理。存储器1002可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需的应用程序等。此外,存储器1002可以包括高速随机存取存储器,还可以包括非易失性存储器,例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。输入装置1003可用于接收输入的数字或字符信息,以及产生与浏览器服务器的用户设置以及功能控制有关的键信号输入。
具体在本实施例中,处理器1001会按照如下的指令,将一个或一个以上的应用程序的进程对应的可执行文件加载到存储器1002中,并由处理器1001来运行存储在存储器1002中的应用程序,从而实现各种功能:
获取当前情境对应的当前情境数据组,所述当前情境数据组包括用于描述当前运行情境下的至少一个当前情境数据;
获取多个历史情境各自对应的历史情境数据组,每个历史情境数据组包括用于描述历史运行情境的至少一个历史情境数据;
根据所述当前情境数据组和所述历史情境数据组,计算得到所述当前情境和每个所述历史情境之间的相似度;
确定与当前情境的相似度符合设定条件的历史情境,根据所述历史情境对应的历史应用程序确定所述当前情境对应的预测应用程序,所述历史应用程序是在所述历史情境下运行的应用程序;
将所述预测应用程序的启动资源预加载至内存中,所述启动资源是启动所述预测应用程序时所需的资源。
一种可能的实现方式中,所述根据所述当前情境数据组和所述历史情境数据组,计算得到所述当前情境和每个所述历史情境之间的相似度,包括:
将所述当前情境数据组和每个所述历史情境数据组输入相似度模型,计算所述当前情境和每个所述历史情境之间的相似度。
一种可能的实现方式中,所述确定与当前情境的相似度符合设定条件的历史情境包括:
确定与当前情境的相似度大于相似度阈值的历史情境;或者,
确定与当前情境的相似度最大的N个历史情境;其中,N为正整数。
一种可能的实现方式中,所述当前情境数据包括当前用户情境数据、当前环境情境数据和当前终端情境数据中的至少一种;
所述当前用户情境数据,用于描述在所述移动终端当前运行时,使用所述移动终端的用户的信息,所述当前用户情境数据包括所述用户的姓名、性别、年龄、职业、心情和教育背景中的至少一种;
所述当前环境情境数据,用于描述所述移动终端当前运行时,使用所述移动终端的用户所处的环境,所述当前环境情境数据包括时间、位置、天气情况、温度、光照、声音和交通情况中的至少一种;
所述当前终端情境数据,用于描述所述移动终端的信息,所述当前终端情境数据包括所述移动终端的终端标识、网络信息和设备类型中的至少一种。
一种可能的实现方式中,所述当前情境数据组包括至少两个所述当前情境数据,所述历史情境数据组包括至少两个所述历史情境数据;
所述计算得到所述当前情境和每个所述历史情境之间的相似度,包括:
计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,所述子相似度是所述当前情境数据和所述历史情境数据组中属于同一个情境数据类型的所述历史情境数据之间的相似度;所述情境数据类型包括:用户情境数据、环境情境数据和终端情境数据中的至少一种;将每个所述当前情境数据对应的子相似度按照所述当前情境数据所属的情境数据类型对应的权重相加,得到所述当前情境与所述历史情境之间的相似度。
一种可能的实现方式中,所述将每个所述当前情境数据对应的子相似度按照所述当前情境数据所属的情境数据类型对应的权重相加,得到所述当前情境与所述历史情境之间的相似度,包括:
Figure PCTCN2017107457-appb-000011
其中,C1是所述当前情境数据组,C2是所述历史情境数据组,Sim(C1,C2)是所述当前情境和所述历史情境之间的相似度;n是所述当前情境数据组中的当前情境数据的个数,Si是所述当前情境数据组中的第i个所述当前情境数据,pi是Si所属的情境数据类型对应的权重,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Sim(Si,Sj)是所述当前情境数据对应的子相似度,n≥2且n为正整数。
一种可能的实现方式中,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
当所述当前情境数据采用数值进行表示时,按照如下公式计算所述子相似度:
Figure PCTCN2017107457-appb-000012
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组 中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据。
一种可能的实现方式中,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
当所述当前情境数据采用区间进行表示时,按照如下公式计算所述子相似度:
Figure PCTCN2017107457-appb-000013
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Si∈[ri s,ri e]、Sj∈[rj s,rj e];ri s是Si的最小值,ri e是Si的最大值,rj s是Sj的最小值,rj e是Sj的最大值。
一种可能的实现方式中,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
当所述当前情境数据采用坐标进行表示时,按照如下公式计算所述子相似度:
Figure PCTCN2017107457-appb-000014
其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Si=(xi,yi)、Sj=(xj,yj)。
一种可能的实现方式中,所述根据相似度最大的N个历史情境对应的历史应用程序确定所述当前情境对应的预测应用程序,包括:
根据所述N个历史情境各自对应的相似度,确定出所述历史情境对应的所述历史应用程序的推荐权重;
获取已经处于运行状态的当前应用程序;
确定每个所述历史应用程序相对于所述当前应用程序的关联启动概率,所述关联启动概率用于表征在所述当前应用程序的运行过程中,所述历史应用程序被关联启动的概率;
在所述N个历史应用程序中,将所述推荐权重和对应的所述关联启动概率的乘积最大的P个历史应用程序确定为所述预测应用程序,P为大于等于1的整数。
一种可能的实现方式中,所述获取当前情境对应的当前情境数据组,还包括:
采集所述当前情境对应的原始情境数据组,所述原始情境数据组包括用于描述所述当前运行情境下的至少一个原始情境数据;使用预定算法对所述原始情境数据组中的所述原始情境数据进行预处理,生成所述当前情境对应的所述当前情境数据组,所述预定算法用于对所述原始情境数据进行数据清理、数据集成、数据规约和数据转换中的至少一种;
和/或,
采集所述当前情境对应的原始情境数据组,所述原始情境数据组包括用于描述所述当前运行情境下的至少一个原始情境数据;使用语义分析方法对所述原始情境数据组中的所述原始情境数据进行预处理,生成所述当前情境对应的所述当前情境数据组。
一种可能的实现方式中,所述将所述当前情境数据组和每个所述历史情境数据组输入相似度模型之前,还包括:
获取至少一组历史情境样本,每组所述历史情境样本包括两个所述历史情境各自对应的所述历史情境数据,以及两个所述历史情境之间的情境相似度分值,所述情境相似度分值是根据两个所述历史情境对应的历史应用程序之间的相似度标定的;
根据所述历史情境样本对初始相似度模型进行训练,得到训练后的所述相似度模型,所述初始相似度模型中包括至少一个情境数据类型以及每个情境数据类型对应的初始权重。
一种可能的实现方式中,所述方法还包括:
确定所述当前情境下实际启动的应用程序;
当所述实际启动的应用程序与所述预测应用程序不同时,根据所述当前情境数据组与所述实际启动的应用程序更新所述相似度模型中的各个情境数据类型对应的所述权重。
本申请实施例还提供一种存储介质,用于存储程序代码,该程序代码用于执行前述各个实施例所述的一种启动资源加载方法中的任意一种实施方式。
本申请实施例还提供一种包括指令的计算机程序产品,当其在计算机上运行时,使得计算机执行前述各个实施例所述的一启动资源加载方法中的任意一种实施方式。
需要说明的是:上述实施例提供的启动资源加载装置在进行加载启动资源时,仅以上述各功能模块的划分进行举例说明,实际应用中,可以根据需要而将上述功能分配由不同的功能模块完成,即将移动终端的内部结构划分成不同的功能模块,以完成以上描述的全部或者部分功能。另外,上述实施例提供的启动资源加载方法和启动资源加载装置实施例属于同一构思,其具体实现过程详见方法实施例,这里不再赘述。
上述本申请实施例序号仅仅为了描述,不代表实施例的优劣。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成,也可以通过程序来指令相关的硬件完成,所述的程序可以存储于一种计算机可读存储介质中,上述提到的存储介质可以是只读存储器,磁盘或光盘等。
以上所述仅为本申请的较佳实施例,并不用以限制本申请,凡在本申请的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本申请的保护范围之内。

Claims (17)

  1. 一种启动资源加载方法,所述方法包括:
    获取当前情境对应的当前情境数据组,所述当前情境数据组包括用于描述当前运行情境下的至少一个当前情境数据;
    获取多个历史情境各自对应的历史情境数据组,每个历史情境数据组包括用于描述历史运行情境的至少一个历史情境数据;
    根据所述当前情境数据组和所述历史情境数据组,计算得到所述当前情境和每个所述历史情境之间的相似度;
    确定与当前情境的相似度符合设定条件的历史情境,根据所述历史情境对应的历史应用程序确定所述当前情境对应的预测应用程序,所述历史应用程序是在所述历史情境下运行的应用程序;
    将所述预测应用程序的启动资源预加载至内存中,所述启动资源是启动所述预测应用程序时所需的资源。
  2. 根据权利要求1所述的方法,所述根据所述当前情境数据组和所述历史情境数据组,计算得到所述当前情境和每个所述历史情境之间的相似度,包括:
    将所述当前情境数据组和每个所述历史情境数据组输入相似度模型,计算所述当前情境和每个所述历史情境之间的相似度。
  3. 根据权利要求1所述的方法,其特征在于,所述确定与当前情境的相似度符合设定条件的历史情境包括:
    确定与当前情境的相似度大于相似度阈值的历史情境;或者,
    确定与当前情境的相似度最大的N个历史情境;其中,N为正整数。
  4. 根据权利要求1所述的方法,所述当前情境数据包括当前用户情境数据、当前环境情境数据和当前终端情境数据中的至少一种;
    所述当前用户情境数据,用于描述在所述移动终端当前运行时,使用所述移动终端的用户的信息,所述当前用户情境数据包括所述用户的姓名、性别、年龄、职业、心情和教育背景中的至少一种;
    所述当前环境情境数据,用于描述所述移动终端当前运行时,使用所述移动终端的用户所处的环境,所述当前环境情境数据包括时间、位置、天气情况、温度、光照、声音和交通情况中的至少一种;
    所述当前终端情境数据,用于描述所述移动终端的信息,所述当前终端情境数据包括所述移动终端的终端标识、网络信息和设备类型中的至少一种。
  5. 根据权利要求1或2所述的方法,所述当前情境数据组包括至少两个所述当前情境数据,所述历史情境数据组包括至少两个所述历史情境数据;
    所述计算得到所述当前情境和每个所述历史情境之间的相似度,包括:
    计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,所述子相似度是所述当前情境数据和所述历史情境数据组中属于同一个情境数据类型的所述历史情境数据之间的相似度;所述情境数据类型包括:用户情境数据、环境情境数据和终端情境数据中的至少一种;将每个所述当前情境数据对应的子相似度按照所述当前情境数据所属的情境数据类型对应的权重相加,得到所述当前情境与所述历史情境之间的相似度。
  6. 根据权利要求5所述的方法,所述将每个所述当前情境数据对应的子相似度按照所述当前情境数据所属的情境数据类型对应的权重相加,得到所述当前情境与所述历史情境之间的相似度,包括:
    Figure PCTCN2017107457-appb-100001
    其中,C1是所述当前情境数据组,C2是所述历史情境数据组,Sim(C1,C2)是所述当前情境和所述历史情境之间的相似度;n是所述当前情境数据组中的当前情境数据的个数,Si是所述当前情境数据组中的第i个所述当前情境数据,pi是Si所属的情境数据类型对应的权重,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Sim(Si,Sj)是所述当前情境数据对应的子相似度,n≥2且n为正整数。
  7. 根据权利要求5所述的方法,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
    当所述当前情境数据采用数值进行表示时,按照如下公式计算所述子相似度:
    Figure PCTCN2017107457-appb-100002
    其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据。
  8. 根据权利要求5所述的方法,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
    当所述当前情境数据采用区间进行表示时,按照如下公式计算所述子相似度:
    Figure PCTCN2017107457-appb-100003
    其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Si∈[ri s,ri e]、Sj∈[rj s,rj e];ri s是Si的最小值,ri e是Si的最大值,rj s是Sj的最小值,rj e是Sj的最大值。
  9. 根据权利要求5所述的方法,所述计算所述当前情境数据组中的每个所述当前情境数据对应的子相似度,包括:
    当所述当前情境数据采用坐标进行表示时,按照如下公式计算所述子相似度:
    Figure PCTCN2017107457-appb-100004
    其中,Sim(Si,Sj)是所述当前情境数据对应的子相似度,Si是所述当前情境数据组中的第i个所述当前情境数据,Sj是所述历史情境数据组中与Si属于同一个情境数据类型的所述历史情境数据,Si=(xi,yi)、Sj=(xj,yj)。
  10. 根据权利要求3所述的方法,所述根据相似度最大的N个历史情境对应的历史应用程序确定所述当前情境对应的预测应用程序,包括:
    根据所述N个历史情境各自对应的相似度,确定出所述历史情境对应的所述历史应用程序的推荐权重;
    获取已经处于运行状态的当前应用程序;
    确定每个所述历史应用程序相对于所述当前应用程序的关联启动概率,所述关联启动概率用于表征在所述当前应用程序的运行过程中,所述历史应用程序被关联启动的概率;
    在所述N个历史应用程序中,将所述推荐权重和对应的所述关联启动概率的乘积最大的P个历史应用程序确定为所述预测应用程序,P为大于等于1的整数。
  11. 根据权利要求1至10任一所述的方法,所述获取当前情境对应的当前情境数据组,还包括:
    采集所述当前情境对应的原始情境数据组,所述原始情境数据组包括用于描述所述当前运行情境下的至少一个原始情境数据;使用预定算法对所述原始情境数据组中的所述原始情境数据进行预处理,生成所述当前情境对应的所述当前情境数据组,所述预定算法用于对所述原始情境数据进行数据清理、数据集成、数据规约和数据转换中的至少一种;
    和/或,
    采集所述当前情境对应的原始情境数据组,所述原始情境数据组包括用于描述所述当前运行情境下的至少一个原始情境数据;使用语义分析方法对所述原始情境数据组中的所述原始情境数据进行预处理,生成所述当前情境对应的所述当前情境数据组。
  12. 根据权利要求2所述的方法,所述将所述当前情境数据组和每个所述历史情境数据组输入相似度模型之前,还包括:
    获取至少一组历史情境样本,每组所述历史情境样本包括两个所述历史情境各自对应的所述历史情境数据,以及两个所述历史情境之间的情境相似度分值,所述情境相似度分值是根据两个所述历史情境对应的历史应用程序之间的相似度标定的;
    根据所述历史情境样本对初始相似度模型进行训练,得到训练后的所述相似度模型,所述初始相似度模型中包括至少一个情境数据类型以及每个情境数据类型对应的初始权重。
  13. 根据权利要求1至12任一所述的方法,所述方法还包括:
    确定所述当前情境下实际启动的应用程序;
    当所述实际启动的应用程序与所述预测应用程序不同时,根据所述当前情境数据组与所述实际启动的应用程序更新所述相似度模型中的各个情境数据类型对应的所述权重。
  14. 一种启动资源加载设备,所述设备包括处理器以及存储器:
    所述存储器用于存储程序代码,并将所述程序代码传输给所述处理器;
    所述处理器用于根据所述程序代码中的指令执行权利要求1-13任一项所述的启动资源加载交互方法。
  15. 一种存储介质,所述存储介质用于存储程序代码,所述程序代码用于执行权利要求1-13任一项所述的启动资源加载方法。
  16. 一种包括指令的计算机程序产品,当其在计算机上运行时,使得所述计算机执行权利要求1-13任一项所述的启动资源加载方法。
  17. 一种启动资源加载方法,应用于移动终端,所述移动终端执行权利要求1-13任意一项所述的启动资源加载方法。
PCT/CN2017/107457 2016-12-19 2017-10-24 启动资源加载方法及装置 WO2018113409A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/368,451 US11169827B2 (en) 2016-12-19 2019-03-28 Resource loading at application startup using attributes of historical data groups

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611184027.6 2016-12-19
CN201611184027.6A CN108228270B (zh) 2016-12-19 2016-12-19 启动资源加载方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/368,451 Continuation US11169827B2 (en) 2016-12-19 2019-03-28 Resource loading at application startup using attributes of historical data groups

Publications (1)

Publication Number Publication Date
WO2018113409A1 true WO2018113409A1 (zh) 2018-06-28

Family

ID=62624743

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/107457 WO2018113409A1 (zh) 2016-12-19 2017-10-24 启动资源加载方法及装置

Country Status (3)

Country Link
US (1) US11169827B2 (zh)
CN (1) CN108228270B (zh)
WO (1) WO2018113409A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110399213A (zh) * 2019-05-21 2019-11-01 腾讯科技(深圳)有限公司 确定应用程序的资源需求的方法、装置、电子设备及介质

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108228270B (zh) * 2016-12-19 2021-04-16 腾讯科技(深圳)有限公司 启动资源加载方法及装置
CN109960539A (zh) 2017-12-21 2019-07-02 广东欧珀移动通信有限公司 应用程序预加载方法、装置、存储介质及移动终端
CN108989887A (zh) * 2018-07-17 2018-12-11 奇酷互联网络科技(深圳)有限公司 移动终端和推送视频的方法、装置
CN109635199B (zh) * 2018-12-17 2021-06-01 广州大学 基于用户行为的应用列表动态推荐方法及系统
WO2020132991A1 (zh) * 2018-12-26 2020-07-02 深圳市欢太科技有限公司 应用推送方法、装置、移动终端及存储介质
CN109889656B (zh) * 2018-12-29 2021-04-20 深圳Tcl新技术有限公司 一种数据读取方法、装置、设备及存储介质
CN110209435A (zh) * 2019-04-28 2019-09-06 北京蓦然认知科技有限公司 一种应用预加载方法、装置
CN110569176A (zh) * 2019-09-17 2019-12-13 北京字节跳动网络技术有限公司 应用预测模型的训练方法及装置、应用控制方法及装置
CN110825818B (zh) * 2019-09-18 2023-06-27 平安科技(深圳)有限公司 多维特征构建方法、装置、电子设备及存储介质
CN110968425B (zh) * 2019-11-22 2022-12-06 中盈优创资讯科技有限公司 一种任务资源动态分配方法及系统
CN111381902B (zh) * 2020-03-10 2021-04-13 中南大学 基于带属性异构网络嵌入的app启动加速方法
US20220037004A1 (en) * 2020-07-31 2022-02-03 Hennepin Healthcare System, Inc. Healthcare worker burnout detection tool
CN112162796A (zh) * 2020-10-10 2021-01-01 Oppo广东移动通信有限公司 应用启动的方法、装置、终端设备以及存储介质
CN112631415B (zh) * 2020-12-31 2022-09-02 Oppo(重庆)智能科技有限公司 Cpu频率调整方法、装置、电子设备及存储介质
CN115016855B (zh) * 2021-11-17 2023-05-09 荣耀终端有限公司 应用预加载的方法、设备和存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2579153A1 (en) * 2010-06-02 2013-04-10 Minoru Yoshida Information generation system and method therefor
CN104168336A (zh) * 2014-09-05 2014-11-26 北京奇虎科技有限公司 数据获取方法、移动设备、计算设备及系统
CN104168670A (zh) * 2014-09-05 2014-11-26 北京奇虎科技有限公司 一种数据获取方法、设备及系统
CN105335178A (zh) * 2014-07-28 2016-02-17 重庆重邮信科通信技术有限公司 一种启动控制方法,及装置
CN105701108A (zh) * 2014-11-26 2016-06-22 阿里巴巴集团控股有限公司 一种信息推荐方法、装置及服务器

Family Cites Families (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8516114B2 (en) * 2002-03-29 2013-08-20 International Business Machines Corporation Method and apparatus for content pre-fetching and preparation
US9002828B2 (en) * 2007-12-13 2015-04-07 Seven Networks, Inc. Predictive content delivery
CN102104666B (zh) * 2009-12-17 2014-03-26 深圳富泰宏精密工业有限公司 应用跳转预测系统及方法
US8874129B2 (en) * 2010-06-10 2014-10-28 Qualcomm Incorporated Pre-fetching information based on gesture and/or location
US9485640B2 (en) * 2011-03-28 2016-11-01 Google Inc. Smart cache warming
US8706918B2 (en) * 2011-11-15 2014-04-22 International Business Machines Corporation External environment sensitive predictive application and memory initiation
US9189252B2 (en) * 2011-12-30 2015-11-17 Microsoft Technology Licensing, Llc Context-based device action prediction
CN102662690B (zh) * 2012-03-14 2014-06-11 腾讯科技(深圳)有限公司 应用程序启动方法和装置
KR101971715B1 (ko) * 2012-06-21 2019-04-23 삼성전자 주식회사 모바일 단말 사용자의 몰입 위험 판단 장치 및 방법
US20140280131A1 (en) * 2013-03-13 2014-09-18 Motorola Mobility Llc Recommendations for Applications Based on Device Context
US9508040B2 (en) * 2013-06-12 2016-11-29 Microsoft Technology Licensing, Llc Predictive pre-launch for applications
US9588897B2 (en) * 2013-07-19 2017-03-07 Samsung Electronics Co., Ltd. Adaptive application caching for mobile devices
CN105677025A (zh) * 2015-12-31 2016-06-15 宇龙计算机通信科技(深圳)有限公司 一种终端应用的启动方法、装置及终端
CN105939416A (zh) * 2016-05-30 2016-09-14 努比亚技术有限公司 移动终端及其应用预启动方法
CN106055369A (zh) * 2016-06-08 2016-10-26 维沃移动通信有限公司 一种移动终端应用程序的启动方法及移动终端
CN106228179A (zh) * 2016-07-13 2016-12-14 乐视控股(北京)有限公司 车辆比对的方法和系统
US10264050B2 (en) * 2016-10-03 2019-04-16 Paypal, Inc. Predictive analysis of computing patterns for preloaded data to reduce processing downtime
CN108228270B (zh) * 2016-12-19 2021-04-16 腾讯科技(深圳)有限公司 启动资源加载方法及装置
US20190042071A1 (en) * 2017-08-07 2019-02-07 Microsoft Technology Licensing, Llc Contextual experience based on location
US10372610B2 (en) * 2017-08-11 2019-08-06 Paypal, Inc. Prefetching data for application usage
US20190087205A1 (en) * 2017-09-18 2019-03-21 Microsoft Technology Licensing, Llc Varying modality of user experiences with a mobile device based on context
US11153285B2 (en) * 2018-11-07 2021-10-19 Citrix Systems, Inc. Systems and methods for application pre-launch

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2579153A1 (en) * 2010-06-02 2013-04-10 Minoru Yoshida Information generation system and method therefor
CN105335178A (zh) * 2014-07-28 2016-02-17 重庆重邮信科通信技术有限公司 一种启动控制方法,及装置
CN104168336A (zh) * 2014-09-05 2014-11-26 北京奇虎科技有限公司 数据获取方法、移动设备、计算设备及系统
CN104168670A (zh) * 2014-09-05 2014-11-26 北京奇虎科技有限公司 一种数据获取方法、设备及系统
CN105701108A (zh) * 2014-11-26 2016-06-22 阿里巴巴集团控股有限公司 一种信息推荐方法、装置及服务器

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110399213A (zh) * 2019-05-21 2019-11-01 腾讯科技(深圳)有限公司 确定应用程序的资源需求的方法、装置、电子设备及介质
CN110399213B (zh) * 2019-05-21 2024-05-10 腾讯科技(深圳)有限公司 确定应用程序的资源需求的方法、装置、电子设备及介质

Also Published As

Publication number Publication date
US11169827B2 (en) 2021-11-09
CN108228270B (zh) 2021-04-16
US20190220293A1 (en) 2019-07-18
CN108228270A (zh) 2018-06-29

Similar Documents

Publication Publication Date Title
WO2018113409A1 (zh) 启动资源加载方法及装置
TWI803514B (zh) 圖像描述生成方法、模型訓練方法、設備和儲存媒體
CN109492698B (zh) 一种模型训练的方法、对象检测的方法以及相关装置
KR102360659B1 (ko) 기계번역 방법, 장치, 컴퓨터 기기 및 기억매체
US10411945B2 (en) Time-distributed and real-time processing in information recommendation system, method and apparatus
EP3010015B1 (en) Electronic device and method for spoken interaction thereof
CN108027952B (zh) 用于提供内容的方法和电子设备
EP2816554A2 (en) Method of executing voice recognition of electronic device and electronic device using the same
CN109040182B (zh) 一种服务访问方法及装置、电子设备、存储介质
US20170109756A1 (en) User Unsubscription Prediction Method and Apparatus
CN111050370A (zh) 网络切换方法、装置、存储介质及电子设备
WO2017088434A1 (zh) 人脸模型矩阵训练方法、装置及存储介质
CN113723378B (zh) 一种模型训练的方法、装置、计算机设备和存储介质
WO2015188765A1 (en) Url error-correcting method, server, terminal and system
CN107765954B (zh) 一种应用程序图标更新方法、移动终端及服务器
CN111090877B (zh) 数据生成、获取方法及对应的装置、存储介质
CN105631059B (zh) 数据处理方法、数据处理装置及数据处理系统
KR102177203B1 (ko) 악성 코드 탐지 방법 및 컴퓨터 판독 가능한 저장매체
CN107807940B (zh) 信息推荐方法和装置
CN117332844A (zh) 对抗样本生成方法、相关装置及存储介质
CN117093766A (zh) 问诊平台的信息推荐方法、相关装置及存储介质
CN109754319B (zh) 信用分值确定系统、方法、终端及服务器
CN116933149A (zh) 一种对象意图预测方法、装置、电子设备和存储介质
EP3951622A1 (en) Image-based search method, server, terminal, and medium
US20210133013A1 (en) Method of monitoring closed system, apparatus thereof and monitoring device

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17883708

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17883708

Country of ref document: EP

Kind code of ref document: A1