WO2012151831A1 - 一种预测用户操作的方法及移动终端 - Google Patents

一种预测用户操作的方法及移动终端 Download PDF

Info

Publication number
WO2012151831A1
WO2012151831A1 PCT/CN2011/079869 CN2011079869W WO2012151831A1 WO 2012151831 A1 WO2012151831 A1 WO 2012151831A1 CN 2011079869 W CN2011079869 W CN 2011079869W WO 2012151831 A1 WO2012151831 A1 WO 2012151831A1
Authority
WO
WIPO (PCT)
Prior art keywords
module
model
mobile terminal
operation model
call instruction
Prior art date
Application number
PCT/CN2011/079869
Other languages
English (en)
French (fr)
Inventor
张琦
毛可
王晓梅
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to US13/991,942 priority Critical patent/US9508041B2/en
Priority to EP11865183.5A priority patent/EP2672781B1/en
Publication of WO2012151831A1 publication Critical patent/WO2012151831A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724—User interfaces specially adapted for cordless or mobile telephones
    • H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724—User interfaces specially adapted for cordless or mobile telephones
    • H04M1/72469—User interfaces specially adapted for cordless or mobile telephones for operating the device by selecting functions from two or more displayed items, e.g. menus or icons
    • H04M1/72472—User interfaces specially adapted for cordless or mobile telephones for operating the device by selecting functions from two or more displayed items, e.g. menus or icons wherein the items are sorted according to specific criteria, e.g. frequency of use
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724—User interfaces specially adapted for cordless or mobile telephones
    • H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
    • H04M1/72451—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions according to schedules, e.g. using calendar applications
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724—User interfaces specially adapted for cordless or mobile telephones
    • H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
    • H04M1/72454—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions according to context-related or environment-related conditions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724—User interfaces specially adapted for cordless or mobile telephones
    • H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
    • H04M1/72457—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions according to geographic location
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M2250/00—Details of telephonic subscriber devices
    • H04M2250/10—Details of telephonic subscriber devices including a GPS signal receiver
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M2250/00—Details of telephonic subscriber devices
    • H04M2250/12—Details of telephonic subscriber devices including a sensor for measuring a physical value, e.g. temperature or motion

Definitions

  • the present invention relates to smart phone technology in mobile communication, and in particular to a method for predicting user operation and a mobile terminal. Background technique
  • Mobile terminals such as mobile phones
  • a habit is formed in the subtle way. This habit can be described as: The user is at a certain time, a certain place every day.
  • the existing mobile phone cannot memorize the user's operating habits, and cannot predict the user's upcoming operation.
  • an object of the present invention is to provide a method for predicting user operations and a mobile terminal, predicting an operation to be performed by a user, and providing an intelligent and detailed service.
  • the present invention provides a method for predicting user operations, the method comprising:
  • the acquired environment factor and the operation model are used to obtain the call instruction, and finally the call instruction is converted into the selection information and displayed to the user.
  • the training operation model is: modifying the operation model by using an operation record list.
  • the modifying the operation model by using the operation record list comprises: the mobile terminal calling the operation record in the operation record list one by one, using the environmental factor of the operation record as the input information of the operation model, and outputting the operation model information
  • the calling instructions in the operation record are compared. If the two are the same, the next operation record is continuously called for comparison. If the two are different, the error of the operation model is increased by one, and the operation model is corrected, and then the next one is called.
  • the operation records are compared until all the operation records in the operation record list are called.
  • the training operation model succeeds in: counting the error of the operation model, determining whether the error of the operation model is lower than a preset usage threshold, and if the threshold is lower than the usage threshold, the training operation model is successful; otherwise, the training operation model is not Successfully, retrain the operational model.
  • the method before the mobile terminal trains the operation model, the method further includes: the mobile terminal saves the call instruction and counts, when the count value of the call instruction is higher than a preset record threshold, the call instruction and the environmental factor are written as The operation record is saved in the operation record list.
  • the converting the calling instruction into the selection information and displaying it to the user comprises: writing the operation to be completed in the calling instruction into a piece of selection information that requires the user to select whether to execute.
  • the present invention also provides a mobile terminal, the mobile terminal comprising:
  • the operation model module is configured to input an environmental factor into the operation model after the training operation model is successful, and the operation model calculates an output call instruction, and then sends the call instruction to the output module;
  • An output module configured to receive a call instruction sent by the operation model module, and convert the call instruction into a selection information and send the message to the display interaction module;
  • the display interaction module is configured to receive selection information sent by the output module, and display the selection information to the user.
  • the mobile terminal further includes:
  • An input factor collection module configured to receive an instruction for displaying an environment of the collection environment sent by the interaction module, and send the environmental factor information to the operation model module;
  • the operation model module is further configured to receive environment factor information sent by the input factor collection module;
  • the display interaction module is further configured to send an instruction for collecting an environmental factor to the input factor collection module when detecting that the user uses the mobile terminal.
  • the mobile terminal further includes: an output event abstraction module and a storage module; wherein
  • the output event abstraction module is configured to determine whether the count value of the call instruction exceeds a record threshold. If the record threshold is exceeded, the call instruction is sent to the storage module, otherwise the operation is not performed;
  • the storage module is configured to receive a call instruction sent by the output event abstraction module.
  • the mobile terminal further includes:
  • the operating system application module is configured to send the call instruction to the output event abstraction module.
  • the output event abstraction module is further configured to save the call instruction sent by the operating system application module.
  • the storage module is further configured to: when receiving the calling instruction, obtain the environmental factor information from the input factor collecting module, and write the calling instruction and the environmental factor into the operation record, and fill in the operation record list;
  • the input factor collection module is further configured to provide environmental factor information for the storage module.
  • the operation model module is specifically used to call the operation record in the operation record list one by one, and the environmental factor of the operation record is used as the input information of the operation model, and the operation mode is The output information of the type is compared with the call finger in the operation record. If the two are the same, the next operation record is continuously called for comparison; if the two are not the same, the error of the operation model is increased by one, and the operation is corrected. The model, then call the next action record until all the action records in the action record list are called.
  • the operation model module is further configured to determine whether the error of the operation model is lower than a preset usage threshold. If the usage threshold is lower than the usage threshold, the operation model training is successful; otherwise, the operation model is retrained.
  • the method for predicting user operation and the mobile terminal provided by the present invention have the following advantages and features: After the operation model is successfully trained, when the user is about to use the mobile terminal, the mobile terminal displays the user using the environmental factor and the operation model. The predicted user's upcoming operation, the user only needs a simple selection to complete a series of identical operations repeated every day, without requiring the user to perform a large number of repeated operations every day, thereby providing users with smarter and more refined service.
  • FIG. 1 is a schematic flow chart of a method for predicting user operations according to the present invention
  • FIG. 2 is a schematic structural diagram of a mobile terminal for predicting user operations according to the present invention
  • FIG. 3 is a schematic structural diagram of a mobile terminal for predicting user operations applied to an Internet of Things environment according to the present invention
  • FIG. 4 is a schematic diagram showing the structure of a mobile terminal for predicting user operations applied to a non-IoT environment according to the present invention.
  • DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The basic idea of the present invention is: After the mobile terminal training operation model is successful, the calling instruction is obtained by using the environmental factor and the operation model, and finally the calling instruction is written into the selection information and displayed to the user.
  • the training refers to: using the operation record list saved by the mobile terminal to operate the model Line correction
  • the operation model refers to: a mathematical model preset in the mobile terminal, the input of the operation model is an environmental factor, and the output is a call instruction, which can be fabricated by using a technique in a neural network;
  • the environmental factors include date, time, location, temperature, and the like.
  • the method for predicting user operation of the present invention is as shown in FIG. 1 and includes the following steps:
  • Step 101 When the mobile phone performs an operation selected by the user, the mobile phone saves the calling instruction of the application and counts.
  • the calling instruction refers to an instruction that is invoked by the user when using one of the applications of the mobile phone, and specifically includes: an application to be called and an operation to be completed, for example: the user sets the mobile phone to become In the silent mode, the call instruction issued includes: calling the mobile phone mode management program, and selecting the silent mode; or the user selects to send the short message after the short message is completed, the call instruction includes: calling the short message application, and saving SMS and recipient number and send these two parts;
  • the saving the calling instruction of the application and counting comprises: the mobile phone first compares the application called by the calling instruction with the operation to be completed, and compares the application called by the calling instruction stored in the mobile phone with the operation to be completed, If they are inconsistent, there is no same call instruction, and the call instruction is counted and saved. If they match, the same call instruction exists, and the count value of the existing call instruction is incremented by one.
  • Step 102 The mobile phone writes the environmental factors of the calling instruction and the mobile phone collection as an operation record and saves it in the operation record list of the mobile phone.
  • Step 102 is further: the mobile phone determines whether the count value of the saved call instruction exceeds a preset recording threshold in the mobile phone, and if the recording threshold is exceeded, the call instruction and the environmental factor are written as an operation record and saved in the operation record list of the mobile phone; Otherwise, return to step 101.
  • the collecting environment factor needs to be collected by installing a Global Positioning System (GPS) module, a temperature and humidity sensing module, or the like, or installing an acquisition module and a wireless gateway loaded with the Internet of Things technology. The interaction can also realize the collection of environmental factors.
  • the specific collection method is determined according to the module selected by the user to be installed on the mobile phone; the collection environment factor can be collected for each time the instruction is saved, or can be judged. The collection is performed when the count value of the saved call instruction exceeds the record threshold;
  • GPS Global Positioning System
  • the recording threshold refers to: a preset value according to an actual application
  • the operation record list refers to: a list composed of a plurality of operation records.
  • Step 103 The mobile phone determines whether the number of operation records saved in the operation record list is higher than a preset training threshold. If it is higher than the training threshold, the process proceeds to step 104; otherwise, returns to step 101.
  • the training threshold refers to: starting the training of the operation model if the number of operation records is higher than the training threshold according to the preset value of the actual application.
  • Step 104 The mobile phone uses the operation record list to train the operation model, and the error of the operation model is statistically calculated.
  • the operation model refers to: a mathematical model preset in the mobile phone, the input is an environmental factor, and the output is a call instruction.
  • a reverse transmission (BP, Back Propagation) network technology in the neural network may be utilized.
  • the model is made.
  • the neural network is an information intelligent processing system established by simulating the information transmitted by the brain. It has the characteristics of self-learning, self-organization, self-adaptation and nonlinear dynamic processing. It is especially suitable for dealing with complex nonlinear processes.
  • BP network is one of the most widely used and successful neural networks. BP network usually includes input layer, hidden layer and output layer.
  • the data model established by BP network can only see the data model for users.
  • the content of the input layer and the output layer for example, where the input layer is an abstract value for each environment factor, and the output layer is a value abstracted by the call instruction;
  • the training operation model by using the operation record list includes: the mobile phone calls the operation record in the operation record list one by one, and uses the environmental factor of the operation record as the input information of the operation model, and compares the output information of the operation model with the call instruction in the operation record. Yes, if the two are the same, then Continue to call the next operation record for comparison. If the two are different, add the error of the operation model to one and correct the operation model; then call the next operation record for comparison until all the operation records in the operation record list will be operated. The call is completed; wherein the error refers to: the output information of the operation model is different from the call instruction in the operation record;
  • the forward propagation of the data model training is: the mobile phone calls the operation record in the operation record list one by one, and takes the value corresponding to each environmental factor of the operation record as the input of the data model, and outputs the value after being processed by the hidden layer. If the output value does not match the value corresponding to the call instruction in the operation record, the reverse propagation phase of the steering error is added, and the error is incremented by one; if the output value is the same as the value corresponding to the call instruction in the operation record, then the call is made. The next operation record is compared until all the operation records in the operation record list are called.
  • the back propagation of the error is: inputting an error between the output of the data model and the value corresponding to the call instruction in the operation record into the hidden layer in a specific form of the BP network data model, and the error is caused by the hidden layer Reverse the input layer, modify the weight of each unit of the hidden layer.
  • the process of weight adjustment of each layer of forward propagation and error backpropagation is repeated, and the process of continuously adjusting the weight is the training process of the data model until the error is less than the preset threshold data model.
  • the backpropagation of the error corrects the data model.
  • Step 105 The mobile phone determines whether the error of the operation model is lower than the usage threshold. If the usage threshold is lower than the usage threshold, the operation model training is successful, and step 106 is performed; if the usage threshold is not lower, the process returns to step 104.
  • the use threshold refers to: according to the preset value of the actual application, if the error is lower than the use threshold, the operation model training is successful, and the error is higher than the use threshold, indicating that the operation model needs to continue training.
  • Step 106 When the user uses the mobile phone, the mobile phone inputs the collected environmental factors into the operation model, and acquires a call instruction output by the operation model.
  • the user uses the mobile phone, and refers to: any behavior that can illuminate the screen of the mobile phone, for example, the user can unlock the mobile phone, or the user can boot.
  • step 106 may be: the mobile phone collects environmental factors in real time, when the user uses the mobile phone, the mobile phone inputs the current environmental factor into the operation model to obtain the call instruction output by the operation model, and then proceeds to step 107;
  • step 106 may be: the mobile phone collects environmental factors in real time, and then inputs the current environmental factors into the operation model in real time to obtain a call instruction of the operation model, and when the user uses the mobile phone, step 107 is performed.
  • Step 107 The mobile phone writes the calling instruction into the selection information and displays it to the user, and operates according to the user's selection.
  • Step 107 is further: converting the call instruction into the selection information means writing the operation to be completed in the call instruction into a piece of selection information that requires the user to select whether to execute, and if the user selects "Yes", directly calling the application and completing the operation ; If the user selects "No", the operation ends.
  • the mobile phone after determining that the error of the operation model is lower than the usage threshold in the above step 105, that is, after the operation model is successfully trained, the mobile phone also deletes the operation record in the operation record list, and the deletion process mainly includes: the periodic operation of the mobile phone
  • the operation record in the record list is judged by the storage duration one by one. If the storage duration of the operation record exceeds the deletion threshold, the operation record is deleted, otherwise the operation is not performed.
  • the periodicity refers to a time set according to an actual situation, for example, may be 1 year;
  • the storage duration refers to a time calculated by using a date in the operation record from the current day period;
  • the deletion threshold refers to A threshold preset for the actual situation.
  • the environment can be utilized when the user is about to use the mobile phone.
  • the operation model the mobile phone will display the predicted user's upcoming operation for the user, and the user only needs to simply select and complete a series of the same operations repeated every day;
  • the mobile phone can obtain the latest user operation record, so that the mobile phone can make more accurate predictions by using the user's latest usage habits.
  • the present invention further provides a mobile terminal for predicting user operations, including: an operation model module 21, an output module 22, and a display.
  • Interaction module 23 wherein
  • the operation model module 21 is configured to: after the training operation model is successful, input an environmental factor input operation model to obtain a call instruction, and then send the call instruction to the output module 22;
  • the output module 22 is configured to receive the call instruction sent by the operation model module 21, and convert the call instruction into the selection information and send it to the display interaction module 23;
  • the display interaction module 23 is configured to receive the selection information sent by the output module 22, and display the selection information to the user.
  • the operation model module 21 is further configured to: after the operation model is successfully trained, send a notification that the operation model training is successful to the display interaction module 23; correspondingly, the display interaction module 23 is further configured to receive the operation model module 21.
  • the operational model trains the successful notification and then begins to detect whether the user is using the mobile terminal;
  • the use of the mobile terminal refers to any behavior that can illuminate the screen of the mobile terminal, for example, the user can be unlocked by the mobile terminal, or the user can open the mobile terminal or the like.
  • the mobile terminal further includes: an input factor collecting module 25, configured to receive an instruction for displaying an environment factor sent by the interaction module 23, and send the collected environmental factor information to the operation model module 21; correspondingly, the operation model module 21 And receiving the environmental factor information sent by the input factor collection module 25;
  • the display interaction module 23 is further configured to: when detecting that the user uses the mobile terminal, send an instruction for collecting the environmental factor to the input factor collection module 25, Collecting environmental factors by the trigger input factor acquisition module 25;
  • the input factor collection module 25 may be configured to collect environmental factors after receiving an instruction to collect environmental factors sent by the interaction module, and may also collect environmental factors in real time;
  • the environmental factor information refers to information composed of environmental factors such as date, time, place, and height, temperature and humidity.
  • the input factor collection module 25 is specifically configured to collect environmental factors through a GPS module, a temperature and humidity sensing module, or the like, or an acquisition module loaded with the Internet of Things technology, and collect environmental factors through interaction with the wireless gateway.
  • the mobile terminal further includes: an output event abstraction module 24 and a storage module 26, wherein the output event abstraction module 24 is configured to determine whether the count value of the call instruction exceeds a record threshold, and if the record threshold is exceeded, send the call instruction to the storage module. 26, otherwise no operation; correspondingly, the storage module 26 is configured to receive a call instruction sent by the output event abstraction module 24.
  • the output event abstraction module 24 is specifically configured to compare the application of the calling instruction and the operation to be completed with the application of the stored call instruction and the operation to be completed, if there is no identical call.
  • the instruction saves the call instruction after counting; if there is the same call instruction, the count value of the existing call instruction is incremented by one.
  • the storage module 26 is further configured to: when receiving the call instruction, obtain environmental factor information from the input factor collection module 25, write the call instruction and the environmental factor as an operation record, and fill in the operation record list; correspondingly, the input
  • the factor collection module 25 is further configured to provide the environment module information for the storage module 26;
  • the input factor collection module 25 may collect environmental factors after receiving the request for acquiring the environmental factor information of the storage module 26, or may collect environmental factors in real time.
  • the storage module 26 is further configured to count the number of operation records saved in the operation record list. Determining whether the number of operation records saved in the operation record list is higher than a preset training threshold. If the training threshold is higher than the training threshold, the operation record list is sent to the operation model module 21, otherwise the operation is not performed; correspondingly, the operation model
  • the module 21 is further configured to receive and save the operation record list sent by the storage module 26, specifically for calling the operation record in the operation record list one by one, using the environmental factor of the operation record as the input information of the operation model, and outputting the operation model The information is compared with the call instruction in the operation record. If the two are the same, the next operation record is continuously called for comparison.
  • the error of the operation model is increased by one, and the operation model is corrected, and then the next call is made.
  • An operation record is compared until all operation records in the operation record list are called, and then it is judged whether the error of the operation model is lower than a preset usage threshold. If the usage threshold is lower than the usage threshold, the operation model training is successful, otherwise retraining Operation model
  • the operation model refers to: a mathematical model preset in the operation model module 21, the input is an environmental factor, and the output is a call instruction.
  • the BP network technology in the neural network can be used to create the model.
  • the operation model module 21 can use the data model established by the BP network technology to train the data model by adjusting the weight between the layer and the layer, and the training process of the operation model module 21 on the data model is: Calling the operation record one by one
  • the operation record in the list, the value corresponding to the environmental factor of the operation record is taken as the input of the data model, and the value is output after being processed by the hidden layer. If the output value does not match the value corresponding to the call instruction in the operation record, the steering error The backpropagation phase, and increase the error by one; if the output value is the same as the value corresponding to the call instruction in the operation record, the next operation record is called for comparison until all the operation records in the operation record list are called.
  • the back-propagation of the error is: inputting the error between the output of the data model and the value corresponding to the call instruction in the operation record into the hidden layer in a specific form of the BP network data model, and the error is caused by the hidden layer Reverse the input layer, correct the weight of each unit of the hidden layer; above, forward propagation and error Cycle is performed to adjust the weights of the respective layers of the propagation process, the process of continuously adjusting weights i.e. data model training process, until the error is less than a pre- The threshold data model can be put into use; the back propagation of the error is the modified data model.
  • the operation model module 21 is further configured to notify the storage module 26 to operate the model training successfully after the operation model is successfully trained.
  • the storage module 26 is further configured to receive the operation model sent by the operation model module 21 and successfully train.
  • the notification of the operation record in the operation record list is periodically determined by the storage duration. If the storage duration of the operation record exceeds the deletion threshold, the operation record is deleted, otherwise the operation is not performed;
  • the periodicity refers to a time set according to an actual situation, for example, may be 1 year; the storage duration refers to a time calculated by using a date in the operation record from the date of the current day; the deletion threshold refers to A threshold preset for the actual situation.
  • the mobile terminal further includes: an operating system application module 27,
  • the operating system application module 27 is configured to send the received call instruction to the output event abstraction module 24; and the corresponding output event abstraction module 24 is further configured to save the call instruction sent by the operating system application module 27.
  • the operating system application module 27 is further configured to receive the calling instruction sent by the display interaction module 23, and send the called application and the response of the application to the display interaction module 23; correspondingly, the display interaction module 23 And displaying the interface for the user, and converting the application of the application to the application to the operating system application module 27, and then displaying the called application and the application response sent by the operating system application module 27 to the user user.
  • the operation model can be used to remember the user's operating habits, thereby achieving the effect of predicting the user's upcoming operation and making the mobile terminal more intelligent.
  • 3 is a structure of a mobile terminal provided in an Internet of Things environment, and the mobile terminal adds an input factor collection module, and can acquire a location, a height, a temperature, a noise, and a care from the Internet of Things through interaction with an external wireless gateway.
  • These environmental factors, through the clock with the outside world The interaction of the blocks can obtain the time and date, so the input factor acquisition module can obtain the environmental factor information, thus completing the implementation process given in the invention.
  • an input factor collection module which may specifically be a GPS module, a temperature and humidity sensing module, a noise sensing module, and a light sensing module. Or a clock module, or any combination of the above, through which the acquisition module can acquire environmental factors such as location, altitude, temperature, noise, illumination, time or date, and thus can utilize these environmental factors to accomplish the implementation described in the present invention. Process.

Landscapes

  • Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Telephone Function (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明公开了一种预测用户操作的方法，包括：移动终端训练操作模型成功后，利用环境因素和操作模型预测出调用指令，最后将调用指令编写成选择信息显示给用户。本发明还同时公开了一种移动终端，采用本发明能预测用户即将进行的操作，为用户提供智能且细化的服务。

Description

一种预测用户操作的方法及移动终端 技术领域
本发明涉及移动通信中的智能手机技术， 尤其涉及一种预测用户操作 的方法及移动终端。 背景技术
移动终端、 如手机， 已经作为一个生活必需品被随身携带， 在每个人 使用手机的过程中， 潜移默化中会形成一种习惯， 这种习惯可以描述为： 用户在每天的某段时刻、 某个地点和某个环境所进行的对手机习惯性的操 作行为。 换句话说， 用户在一天中的某些特定的时刻、 特定地点对手机的 使用具有一定规律。 但是， 现有的手机不能对用户的操作习惯进行记忆， 无法预测出用户即将进行的操作。
目前， 有些手机逐渐开始做出简单的改进用来解决这个问题， 比如： 添加记忆程序来记忆用户上次的操作， 或者是将应用程序的图标设置在主 界面上， 以供用户较快捷的使用。 但是， 上述的改进方法， 要么只能提供 用户上次的操作记录， 无法预测用户即将进行的操作； 要么需要用户进行 手动操作， 对经常使用的应用程序做设置， 无法为用户提供智能的服务。 可见， 目前没有移动终端能够预测用户即将进行的操作， 为用户提供智能 且细化的服务。 发明内容
有鉴于此， 本发明的目的在于提供一种预测用户操作的方法及移动终 端， 预测用户即将进行的操作， 提供智能且细化的服务。
为达到上述目的， 本发明的技术方案是这样实现的： 本发明提供了一种预测用户操作的方法， 该方法包括：
移动终端训练操作模型成功后，利用采集到的环境因素和操作模型获 取调用指令， 最后将调用指令转换为选择信息显示给用户。
上述方案中， 所述训练操作模型为： 利用操作记录列表对操作模型进 行修正。
上述方案中， 所述利用操作记录列表对操作模型进行修正， 包括： 移 动终端逐条调用操作记录列表中的操作记录， 将操作记录的环境因素作为 操作模型的输入信息， 将操作模型的输出信息跟操作记录中的调用指令做 比对， 如果二者相同， 则继续调用下一条操作记录进行比对； 如果二者不 相同， 则将操作模型的误差加一， 并修正操作模型， 然后调用下一条操作 记录进行比对， 直到将操作记录列表中的所有操作记录都调用完。
上述方案中， 所述训练操作模型成功， 为： 统计操作模型的误差， 判 断操作模型的误差是否低于预设的使用门限， 如果低于使用门限， 则训练 操作模型成功； 否则训练操作模型不成功， 重新训练操作模型。
上述方案中， 所述移动终端训练操作模型之前， 该方法还包括： 移 动终端保存调用指令并计数， 当该调用指令的计数值高于预设的记录门限 时， 将调用指令和环境因素编写为操作记录保存在操作记录列表中。
上述方案中， 所述将调用指令转换为选择信息显示给用户， 包括： 将 调用指令中的所要完成的操作编写成需要用户选择是否执行的一条选择信 息。
本发明还提供了一种移动终端， 该移动终端包括：
操作模型模块， 用于在训练操作模型成功后， 将环境因素输入操作模 型， 操作模型计算输出调用指令， 再将调用指令发送给输出模块；
输出模块， 用于接收操作模型模块发来的调用指令， 将调用指令转换 成选择信息发送给显示交互模块； 显示交互模块， 用于接收输出模块发来的选择信息， 将选择信息显示 给用户。
上述方案中， 该移动终端进一步包括：
输入因素采集模块， 用于接收显示交互模块的发来的采集环境因素的 指令， 将环境因素信息发送给操作模型模块；
相应的， 所述操作模型模块， 还用于接收输入因素采集模块发来的环 境因素信息；
所述显示交互模块， 还用于在检测到用户使用移动终端时， 将采集环 境因素的指令发送给输入因素采集模块。
上述方案中， 该移动终端进一步包括： 输出事件抽象模块和存储模块； 其中，
输出事件抽象模块， 用于判断调用指令的计数值是否超过记录门限， 如果超过记录门限， 则将调用指令发送给存储模块， 否则不做操作；
存储模块， 用于接收输出事件抽象模块发来的调用指令。
上述方案中， 该移动终端进一步包括：
操作系统应用模块， 用于将调用指令发送给输出事件抽象模块； 相应的， 所述输出事件抽象模块， 还用于保存操作系统应用模块发来 的调用指令。
上述方案中， 所述存储模块， 还用于接收到调用指令时， 从输入因素 采集模块获取环境因素信息， 将调用指令和环境因素编写为操作记录， 填 写在操作记录列表中；
相应的， 所述输入因素采集模块， 还用于为存储模块提供环境因素信 息。
上述方案中， 所述操作模型模块， 具体用于逐条调用操作记录列表中 的操作记录， 将操作记录的环境因素作为操作模型的输入信息， 将操作模 型的输出信息跟操作记录中的调用指 ^故比对， 如果二者相同， 则继续调 用下一条操作记录进行比对； 如果二者不相同， 则将操作模型的误差加一， 并修正操作模型， 然后调用下一条操作记录， 直到将操作记录列表中的所 有操作记录都调用完。
上述方案中， 所述操作模型模块， 还用于判断操作模型的误差是否低 于预设的使用门限， 如果低于使用门限， 则操作模型训练成功； 否则重新 训练操作模型。
本发明所提供的预测用户操作的方法及移动终端， 具有以下的优点和 特点： 在操作模型训练成功后， 用户即将使用移动终端的时候， 利用环境 因素以及操作模型， 移动终端会为用户显示出预测到的用户即将进行的操 作， 用户只需要简单的进行选择就可以完成每天重复进行的一系列相同的 操作， 不需要用户每天进行大量重复的操作， 从而为用户提供更为智能且 细化的服务。 附图说明
图 1为本发明预测用户操作的方法流程示意图；
图 2为本发明预测用户操作的移动终端的结构示意图；
图 3 为本发明预测用户操作的移动终端应用于物联网环境的结构示意 图；
图 4为本发明预测用户操作的移动终端应用于非物联网环境的结构示 意图。 具体实施方式 本发明的基本思想是： 移动终端训练操作模型成功后， 利用环境因 素和操作模型获取调用指令， 最后将调用指令编写成选择信息显示给用户。
其中， 所述训练指： 利用移动终端保存的操作记录列表对操作模型进 行修正；
所述操作模型指： 预置在移动终端中的数学模型， 操作模型的输入为 环境因素， 输出为调用指令， 可以利用神经网络中的技术来制作；
所述环境因素包括日期、 时间、 地点和温度等。
下面结合附图及具体实施例对本发明再作进一步详细的说明。
以下以手机为例， 本发明预测用户操作的方法如图 1 所示， 包括以下 步驟：
步驟 101 : 手机在执行用户选定的一项应用进行操作时，保存对该应用 程序的调用指令并计数。
这里， 所述调用指令指， 用户在使用手机的其中一项应用时发出的调 用该应用程序的指令， 具体包括： 所要调用的应用程序和所要完成的操作 两部分内容， 比如： 用户设置手机成为静音模式， 所发出的调用指令中包 括： 调用手机模式管理程序、 和选定静音模式这两部分内容； 或者用户编 写短信完成后选择发送短信， 则调用指令中包括： 调用短信应用程序、 和 保存短信及收件人号码并发送这两部分内容；
所述保存对该应用程序的调用指令并计数包括： 手机首先将调用指令 调用的应用程序和所要完成的操作跟手机中已存的所有调用指令调用的应 用程序和所要完成的操作作比对， 如果不一致， 则没有相同的调用指令， 对调用指令进行计数后保存； 如果一致， 则存在相同的调用指令， 将已存 的调用指令的计数值加一。
步驟 102:手机将调用指令和手机采集的环境因素编写为一条操作记录 保存在手机的操作记录列表中。
步驟 102进一步为： 手机判断保存的调用指令的计数值是否超过手机 中预设的记录门限， 如果超过记录门限， 则将调用指令和环境因素编写为 一条操作记录保存在手机的操作记录列表中； 否则返回步驟 101。 这里，所述采集环境因素，需要通过在手机上安装全球定位系统（GPS, Global Positioning System )模块、 温湿度传感模块等来进行采集， 或者通过 安装加载了物联网技术的采集模块与无线网关的交互也可以实现环境因素 采集， 具体使用的采集方法根据用户选择安装在手机上的模块而定； 所述 采集环境因素， 可以为每次保存调用指令的同时进行采集， 也可以为判断 出进行保存的调用指令的计数值超过记录门限的时候进行采集；
其中， 所述记录门限指： 根据实际应用预设的数值；
所述操作记录列表指： 由多条操作记录组成的列表。
步驟 103:手机判断操作记录列表中所保存的操作记录的数量是否高于 预设的训练门限，如果高于训练门限， 则进入步驟 104; 否则返回步驟 101。
这里， 所述训练门限指： 根据实际应用预设的数值， 若操作记录的数 量高于训练门限就开始对操作模型进行训练。
步驟 104:手机利用操作记录列表训练操作模型，统计操作模型的误差。 这里， 所述操作模型指： 手机中预置的数学模型， 其输入为环境因素， 输出为调用指令， 在实际应用中， 可以利用神经网络中的反向传输（BP, Back Propagation ) 网络技术来制作该模型， 神经网络是以模拟脑神经传递 信息的方法建立起来的信息智能化处理系统， 具有自学习、 自组织、 自适 应和非线性动态处理等特性， 特别适合处理复杂的非线性过程， BP网络是 目前为止应用最为广泛和成功的神经网络之一， BP网络通常包括输入层、 隐含层、 输出层， 由 BP网路建立起的数据模型， 对于用户而言只能看到数 据模型的输入层和输出层的内容， 比如， 其中输入层为各个环境因素抽象 出的数值， 输出层为调用指令抽象出的数值；
所述利用操作记录列表训练操作模型包括： 手机逐条调用操作记录列 表中的操作记录， 将操作记录的环境因素作为操作模型的输入信息， 将操 作模型的输出信息跟操作记录中的调用指令做比对， 如果二者相同， 则继 续调用下一条操作记录进行比对， 如果二者不相同， 则将操作模型的误差 加一， 并修正操作模型； 然后调用下一条操作记录进行比对， 直到将操作 记录列表中的所有操作记录都调用完； 其中， 所述误差指： 操作模型的输 出信息跟操作记录中的调用指令不同的次数；
比如： 在实际应用中可以为对利用 BP网络技术制作的数据模型、 靠调 整层与层之间的权值进行训练， 具体的， 对数据模型的训练过程由正向传 播与误差的反向传播两个过程组成；
其中， 所述对数据模型训练的正向传播为： 手机逐条调用操作记录列 表中的操作记录， 将操作记录的各个环境因素所对应的数值作为数据模型 的输入， 经隐含层处理后输出数值， 若输出的数值与操作记录中的调用指 令对应的数值不符， 则转向误差的反向传播阶段， 并将误差加一； 若输出 的数值与操作记录中的调用指令对应的数值相同， 则调用下一条操作记录 进行比对， 直到将操作记录列表中的所有操作记录都调用完。 其中， 所述 误差的反向传播为： 将数据模型的输出和操作记录中的调用指令对应的数 值之间的误差以 BP网络数据模型的特定形式输入隐含层，由隐含层将该误 差向输入层反传， 修正隐含层各个单元的权值。 上述， 正向传播与误差反 向传播的各层权值调整过程是周而复始地进行， 权值不断调整的过程也就 是对数据模型的训练过程， 直到误差小于预设的门限数据模型就可以投入 使用； 所述误差的反向传播即修正数据模型。
步驟 105: 手机判断操作模型的误差是否低于使用门限，如果低于使用 门限， 则操作模型训练成功， 执行步驟 106; 如果不低于使用门限， 则返回 步驟 104。
这里， 所述使用门限指： 根据实际应用预设的数值， 误差低于使用门 限则说明操作模型训练成功， 误差高于使用门限则说明操作模型还需要继 续训练。 步驟 106:用户使用手机时，手机将所采集的环境因素输入操作模型中， 获取操作模型输出的调用指令。
这里， 所述用户使用手机， 指： 任何可以点亮手机屏幕的行为， 比如， 可以为用户解锁手机， 或者用户开机等。
上述步驟 106可以为： 手机实时采集环境因素， 当用户使用手机时， 手机将当前环境因素输入到操作模型中获取操作模型输出的调用指令， 然 后执行步驟 107;
或者步驟 106还可以为： 手机实时采集环境因素， 然后实时将当前环 境因素输入到操作模型中获取操作模型的调用指令， 当用户使用手机时执 行步驟 107。
步驟 107: 手机将调用指令编写成选择信息显示给用户， 并根据用户的 选择进行操作。
步驟 107进一步为： 将调用指令转换成选择信息指将调用指令中的所 要完成的操作编写成需要用户选择是否执行的一条选择信息， 如果用户选 择为 "是"， 则直接调用应用程序并完成操作； 如果用户选择为 "否"， 则 结束操作。
此外， 在上述步驟 105判断操作模型的误差低于使用门限后， 即操作 模型训练成功后， 手机还会对操作记录列表中的操作记录进行删除， 该删 除过程主要包括： 手机周期性的对操作记录列表中的操作记录逐条进行存 储时长判断， 如果操作记录的存储时长超过删除门限， 则删除该条操作记 录， 否则不做操作。
这里， 所述周期性指根据实际情况设定的时间， 比如可以为 1 年； 所 述存储时长指利用操作记录中的日期计算出的该记录距当前日期间的时 间； 所述删除门限指根据实际情预置的一个门限值。
可见， 通过上述步驟， 可以在用户即将使用手机的时候， 利用环境因 素以及操作模型， 手机会为用户显示出预测到的用户即将进行的操作， 用 户只需要简单的进行选择就可以完成每天重复进行的一系列相同的操作； 另外还能通过操作记录的删除操作， 使手机获取最新的用户操作记录， 从 而手机就能够利用用户最新的使用习惯做出更加准确的预测。
为实现上述预测用户操作的方法， 需要对移动终端进行改进， 因此， 如图 2所示， 本发明还提供了一种预测用户操作的移动终端， 包括： 操作 模型模块 21、 输出模块 22和显示交互模块 23; 其中，
操作模型模块 21 , 用于在训练操作模型成功后， 将环境因素输入操作 模型获取调用指令， 再将调用指令发送给输出模块 22;
输出模块 22, 用于接收操作模型模块 21发来的调用指令，将调用指令 转换成选择信息发送给显示交互模块 23;
显示交互模块 23 , 用于接收输出模块 22发来的选择信息，将选择信息 显示给用户。
所述操作模型模块 21 , 还用于在操作模型训练成功后， 向显示交互模 块 23发送操作模型训练成功的通知； 相应的， 所述显示交互模块 23 , 还用 于接收操作模型模块 21发来的操作模型训练成功的通知， 然后开始检测用 户是否使用移动终端；
所述使用移动终端， 指任何可以点亮移动终端屏幕的行为， 比如， 可 以为用户解锁移动终端， 或者用户打开移动终端等。
上述移动终端进一步包括： 输入因素采集模块 25 , 用于接收显示交互 模块 23发来的采集环境因素的指令， 将采集的环境因素信息发送给操作模 型模块 21 ; 相应的， 所述操作模型模块 21 , 还用于接收输入因素采集模块 25发来的环境因素信息； 所述显示交互模块 23 , 还用于在检测到用户使用 移动终端时， 将采集环境因素的指令发送给输入因素采集模块 25 , 以触发 输入因素采集模块 25采集环境因素； 其中， 所述输入因素采集模块 25 , 可以为接收到显示交互模快 23发来 的采集环境因素的指令后， 再进行环境因素的采集， 也可以为实时采集环 境因素；
这里， 所述环境因素信息指由日期、 时间、 地点以及高度、 温湿度等 环境因素组成的信息。
所述输入因素采集模块 25 , 具体用于通过 GPS模块、 温湿度传感模块 等来采集环境因素， 也可以是加载了物联网技术的采集模块， 通过与无线 网关的交互采集环境因素。
上述移动终端进一步包括： 输出事件抽象模块 24和存储模块 26， 所述输出事件抽象模块 24, 用于判断调用指令的计数值是否超过记录 门限， 如果超过记录门限， 则将调用指令发送给存储模块 26, 否则不做操 作； 相应的， 所述存储模块 26, 用于接收输出事件抽象模块 24发来的调用 指令。
所述输出事件抽象模块 24, 具体用于将调用指令的调用的应用程序和 所要完成的操作跟已存的所有调用指令的调用的应用程序和所要完成的操 作作比对， 如果没有相同的调用指令， 则对调用指令进行计数后保存； 如 果存在相同的调用指令， 则将已存的调用指令的计数值加一。
所述存储模块 26, 还用于接收到调用指令时， 从输入因素采集模块 25 获取环境因素信息， 将调用指令和环境因素编写为操作记录， 填写在操作 记录列表中； 相应的， 所述输入因素采集模块 25 , 还用于为存储模块 26提 供环境因素信息；
其中， 所述输入因素采集模块 25 , 可以为接收到存储模块 26的获取环 境因素信息的请求后， 再进行环境因素的采集， 也可以为实时采集环境因 素。
所述存储模块 26,还用于统计操作记录列表中保存的操作记录的数量， 判断操作记录列表中保存的操作记录的数量是否高于预设的训练门限， 如 果高于训练门限， 则将操作记录列表发给操作模型模块 21 , 否则不做操作； 相应的， 所述操作模型模块 21 ,还用于接收并保存存储模块 26发来的操作 记录列表， 具体用于逐条调用操作记录列表中的操作记录， 将操作记录的 环境因素作为操作模型的输入信息 , 将操作模型的输出信息跟操作记录中 的调用指令做比对， 如果二者相同则继续调用下一条操作记录进行比对， 如果二者不相同， 则将操作模型的误差加一， 并修正操作模型， 然后调用 下一条操作记录进行比对， 直到将操作记录列表中的所有操作记录都调用 完， 然后判断操作模型的误差是否低于预设的使用门限， 如果低于使用门 限则操作模型训练成功， 否则重新训练操作模型；
其中， 所述操作模型指： 预置在操作模型模块 21中的数学模型， 其输 入为环境因素， 输出为调用指令， 在实际应用中， 可以利用神经网络中的 BP网络技术来制作该模型。
所述操作模型模块 21 , 可以利用 BP网络技术建立的数据模型， 靠调 整层与层之间的权值对数据模型进行训练， 则操作模型模块 21对数据模型 的训练过程为： 逐条调用操作记录列表中的操作记录， 将操作记录的环境 因素所对应的数值作为数据模型的输入， 经隐含层处理后输出数值， 若输 出的数值与操作记录中的调用指令对应的数值不符， 则转向误差的反向传 播阶段， 并将误差加一； 若输出的数值与操作记录中的调用指令对应的数 值相同， 则调用下一条操作记录进行比对， 直到将操作记录列表中的所有 操作记录都调用完； 其中， 误差的反向传播为： 将数据模型的输出和操作 记录中的调用指令对应的数值之间的误差以 BP 网络数据模型的特定形式 输入隐含层， 由隐含层将该误差向输入层反传， 修正隐含层各个单元的权 值； 上述， 正向传播与误差反向传播的各层权值调整过程是周而复始地进 行， 权值不断调整的过程也就是对数据模型的训练过程， 直到误差小于预 设的门限数据模型就可以投入使用； 所述误差的反向传播即修正数据模型。 所述操作模型模块 21 , 还用于在操作模型训练成功后， 通知存储模块 26操作模型训练成功； 相应的， 所述存储模块 26, 还用于接收操作模型模 块 21发来的操作模型训练成功的通知， 然后周期性的对操作记录列表中的 操作记录逐条进行存储时长判断， 如果操作记录的存储时长超过删除门限， 则删除该条操作记录， 否则不做操作；
其中， 所述周期性指根据实际情况设定的时间， 比如可以为 1 年； 所 述存储时长指利用操作记录中的日期计算出的该记录距当前日期间的时 间； 所述删除门限指根据实际情预置的一个门限值。
所述移动终端进一步包括： 操作系统应用模块 27,
所述操作系统应用模块 27 ,用于将接收到的调用指令发送给输出事件抽 象模块 24; 相应的输出事件抽象模块 24,还用于保存操作系统应用模块 27 发来的调用指令。
所述操作系统应用模块 27 , 还用于接收显示交互模块 23发来的调用指 令， 并将调出的应用程序以及应用程序的响应发送给显示交互模块 23; 相 应的， 所述显示交互模块 23 , 还用于为用户显示界面， 并将用户对应用程 序的应用转化为调用指令发送给操作系统应用模块 27, 再将操作系统应用 模块 27发来的调出的应用程序以及应用程序响应显示给用户。
可见， 应用上述的方案， 就可以利用操作模型记住用户的操作习惯， 从而能够达到预测用户即将进行的操作的效果， 使移动终端更加智能。
本发明公开的技术方案可应用于物联网环境和非物联网环境， 下面对 具体实现做进一步详细说明：
图 3 为本发明提供的移动终端应用于物联网环境中的结构， 可见移动 终端添加了输入因素采集模块， 通过与外界无线网关的交互就可以从物联 网获取地点、 高度、 温度、 噪声和关照这些环境因素， 通过与外界时钟模 块的交互就能够获得时间和日期， 因此输入因素采集模块就能够获得环境 因素信息， 从而完成发明中所给出的实现流程。
图 4为本发明提供的移动终端应用于非物联网环境中的结构， 可见移 动终端添加了输入因素采集模块，具体可以是 GPS模块、温湿度传感模块、 噪声传感模块、 光线传感模块或时钟模块， 或以上任意的组合， 通过这些 输入因素采集模块就能够获取地点、 高度、 温度、 噪声、 光照、 时间或日 期这些环境因素， 进而能够利用这些环境因素完成本发明中所述的实现流 程。
以上所述， 仅为本发明的较佳实施例而已， 并非用于限定本发明的保 护范围。

Claims

权利要求书
1、 一种预测用户操作的方法， 其特征在于， 该方法包括：
移动终端训练操作模型成功后，利用采集到的环境因素和操作模型获 取调用指令， 最后将调用指令转换为选择信息显示给用户。
2、 根据权利要求 1所述的方法， 其特征在于， 所述训练操作模型为： 利用操作记录列表对操作模型进行修正。
3、 根据权利要求 2所述的方法， 其特征在于， 所述利用操作记录列表 对操作模型进行修正， 包括： 移动终端逐条调用操作记录列表中的操作记 录， 将操作记录的环境因素作为操作模型的输入信息， 将操作模型的输出 信息跟操作记录中的调用指令做比对， 如果二者相同， 则继续调用下一条 操作记录进行比对； 如果二者不相同， 则将操作模型的误差加一， 并修正 操作模型， 然后调用下一条操作记录进行比对， 直到将操作记录列表中的 所有操作记录都调用完。
4、根据权利要求 1所述的方法，其特征在于，所述训练操作模型成功， 为： 统计操作模型的误差， 判断操作模型的误差是否低于预设的使用门限， 如果低于使用门限， 则训练操作模型成功； 否则训练操作模型不成功， 重 新训练操作模型。
5、根据权利要求 1所述的方法， 其特征在于， 所述移动终端训练操作 模型之前， 该方法还包括： 移动终端保存调用指令并计数， 当该调用指令 的计数值高于预设的记录门限时， 将调用指令和环境因素编写为操作记录 保存在操作记录列表中。
6、 根据权利要求 1至 5任一所述的方法， 其特征在于， 所述将调用指 令转换为选择信息显示给用户， 包括： 将调用指令中的所要完成的操作编 写成需要用户选择是否执行的一条选择信息。
7、 一种移动终端， 其特征在于， 该移动终端包括： 操作模型模块， 用于在训练操作模型成功后， 将环境因素输入操作模 型， 操作模型计算输出调用指令， 再将调用指令发送给输出模块；
输出模块， 用于接收操作模型模块发来的调用指令， 将调用指令转换 成选择信息发送给显示交互模块；
显示交互模块， 用于接收输出模块发来的选择信息， 将选择信息显示 给用户。
8、根据权利要求 7所述的移动终端， 其特征在于， 该移动终端进一步 包括：
输入因素采集模块， 用于接收显示交互模块发来的采集环境因素的指 令， 将环境因素信息发送给操作模型模块；
相应的， 所述操作模型模块， 还用于接收输入因素采集模块发来的环 境因素信息；
所述显示交互模块， 还用于在检测到用户使用移动终端时， 将采集环 境因素的指令发送给输入因素采集模块。
9、根据权利要求 7所述的移动终端， 其特征在于， 该移动终端进一步 包括： 输出事件抽象模块和存储模块； 其中，
输出事件抽象模块， 用于判断调用指令的计数值是否超过记录门限， 如果超过记录门限， 则将调用指令发送给存储模块， 否则不做操作；
存储模块， 用于接收输出事件抽象模块发来的调用指令。
10、根据权利要求 9所述的移动终端， 其特征在于， 该移动终端进一步 包括：
操作系统应用模块， 用于将调用指令发送给输出事件抽象模块； 相应的， 所述输出事件抽象模块， 还用于保存操作系统应用模块发来 的调用指令。
11、 根据权利要求 9所述的移动终端， 其特征在于， 所述存储模块， 还用于接收到调用指令时， 从输入因素采集模块获取 环境因素信息， 将调用指令和环境因素编写为操作记录， 填写在操作记录 列表中；
相应的， 所述输入因素采集模块， 还用于为存储模块提供环境因素信
12、 根据权利要求 7所述的移动终端， 其特征在于，
所述操作模型模块， 具体用于逐条调用操作记录列表中的操作记录， 将操作记录的环境因素作为操作模型的输入信息， 将操作模型的输出信息 跟操作记录中的调用指令做比对， 如果二者相同， 则继续调用下一条操作 记录进行比对； 如果二者不相同， 则将操作模型的误差加一， 并修正操作 模型， 然后调用下一条操作记录， 直到将操作记录列表中的所有操作记录 都调用完。
13、 根据权利要求 7至 12任一所述的移动终端， 其特征在于， 所述操作模型模块， 还用于判断操作模型的误差是否低于预设的使用 门限， 如果低于使用门限， 则操作模型训练成功； 否则重新训练操作模型。
PCT/CN2011/079869 2011-07-18 2011-09-20 一种预测用户操作的方法及移动终端 WO2012151831A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US13/991,942 US9508041B2 (en) 2011-07-18 2011-09-20 Method for predicting user operation and mobile terminal
EP11865183.5A EP2672781B1 (en) 2011-07-18 2011-09-20 Method and mobile terminal for predicting user operation

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110200350.9A CN102891916B (zh) 2011-07-18 2011-07-18 一种预测用户操作的方法及移动终端
CN201110200350.9 2011-07-18

Publications (1)

Publication Number Publication Date
WO2012151831A1 true WO2012151831A1 (zh) 2012-11-15

Family

ID=47138673

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/079869 WO2012151831A1 (zh) 2011-07-18 2011-09-20 一种预测用户操作的方法及移动终端

Country Status (3)

Country Link
EP (1) EP2672781B1 (zh)
CN (1) CN102891916B (zh)
WO (1) WO2012151831A1 (zh)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10362125B2 (en) 2014-09-18 2019-07-23 Intel Corporation Technologies for pre-action execution
TWI549076B (zh) * 2014-11-06 2016-09-11 宏碁股份有限公司 電子裝置及其服務管理方法
CN106855796A (zh) 2015-12-09 2017-06-16 阿里巴巴集团控股有限公司 一种数据处理方法、装置和智能终端
CN106919314B (zh) * 2015-12-24 2021-04-02 小米科技有限责任公司 移动终端及亮屏方法
CN105975540A (zh) * 2016-04-29 2016-09-28 北京小米移动软件有限公司 信息显示方法及装置
CN106383832B (zh) * 2016-08-26 2019-10-22 用友网络科技股份有限公司 一种数据挖掘训练模型的生成方法
CN107885591A (zh) 2016-09-27 2018-04-06 华为技术有限公司 为应用分配系统资源的方法和终端
EP3627806A4 (en) * 2017-06-16 2020-05-13 Huawei Technologies Co., Ltd. USER PORTRAIT GENERATION METHOD, AND TERMINAL
CN107728772B (zh) * 2017-09-30 2020-05-12 Oppo广东移动通信有限公司 应用程序的处理方法、装置、存储介质及电子设备
CN110879750A (zh) 2017-10-13 2020-03-13 华为技术有限公司 资源管理的方法及终端设备
CN107831763B (zh) * 2017-10-18 2021-10-12 广州小鹏汽车科技有限公司 一种基于大数据学习的车载控制方法及系统

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1497434A (zh) * 2002-10-09 2004-05-19 ���µ�����ҵ��ʽ���� 信息终端设备、操作支持方法和操作支持程序
CN102037481A (zh) * 2008-03-19 2011-04-27 苹果核网络股份有限公司 用于检测行为模式的方法及设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8938688B2 (en) * 1998-12-04 2015-01-20 Nuance Communications, Inc. Contextual prediction of user words and user actions
US8990688B2 (en) * 2003-09-05 2015-03-24 Samsung Electronics Co., Ltd. Proactive user interface including evolving agent
JP5090661B2 (ja) * 2006-04-12 2012-12-05 株式会社エヌ・ティ・ティ・ドコモ ソフトウェア動作モデル化装置、ソフトウェア動作監視装置、ソフトウェア動作モデル化方法及びソフトウェア動作監視方法
WO2010010653A1 (ja) * 2008-07-24 2010-01-28 日本電気株式会社 ユーザモデル処理装置

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1497434A (zh) * 2002-10-09 2004-05-19 ���µ�����ҵ��ʽ���� 信息终端设备、操作支持方法和操作支持程序
CN102037481A (zh) * 2008-03-19 2011-04-27 苹果核网络股份有限公司 用于检测行为模式的方法及设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2672781A4 *

Also Published As

Publication number Publication date
CN102891916A (zh) 2013-01-23
EP2672781A1 (en) 2013-12-11
CN102891916B (zh) 2016-01-20
EP2672781A4 (en) 2014-02-19
EP2672781B1 (en) 2016-11-02

Similar Documents

Publication Publication Date Title
WO2012151831A1 (zh) 一种预测用户操作的方法及移动终端
JP7274418B2 (ja) 反復アクションに対するスマートアシスト
CN108076224B (zh) 应用程序控制方法、装置及存储介质和移动终端
US9508041B2 (en) Method for predicting user operation and mobile terminal
CN105025173B (zh) 移动计算设备上的自知简档切换
CN106453608B (zh) 一种基于云端的移动应用的后台请求自适应调度算法
CN107045528A (zh) 一种出行提醒的方法及终端
JP5870217B2 (ja) 適応システム
EP3231199B1 (en) Notifications on mobile devices
CN107209540A (zh) 对计算设备上的热修复进行管理
CN107548500A (zh) 基于用户例程模型的事件异常
WO2011088682A1 (zh) 一种通信终端日程设置方法、装置及终端
CN110503952B (zh) 一种语音处理方法、装置和电子设备
US9704167B2 (en) Electronic devices and service management methods for providing services corresponding to different situations
CN112751648A (zh) 一种丢包数据恢复方法和相关装置
US20150301677A1 (en) Device management system and method
CN110705646A (zh) 一种基于模型动态更新的移动设备流式数据识别方法
CN111797861A (zh) 信息处理方法、装置、存储介质及电子设备
CN111311014B (zh) 业务数据处理方法、装置、计算机设备和存储介质
CN107402777A (zh) 一种资源更新方法及电子设备
CN105611074A (zh) 一种根据闹钟时间优化推送信息时间的方法及系统
CN113055984B (zh) 终端控制方法、装置、移动终端及存储介质
US20180001140A1 (en) Exercise schedule optimizer system
CN111556155B (zh) 信息推送方法和装置
CN105472691B (zh) 无线上网控制方法、系统及电子装置

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: 11865183

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 13991942

Country of ref document: US

REEP Request for entry into the european phase

Ref document number: 2011865183

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2011865183

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE