WO2019218465A1 - 一种用户行为提示生成方法、装置、终端设备及介质 - Google Patents

一种用户行为提示生成方法、装置、终端设备及介质 Download PDF

Info

Publication number
WO2019218465A1
WO2019218465A1 PCT/CN2018/097118 CN2018097118W WO2019218465A1 WO 2019218465 A1 WO2019218465 A1 WO 2019218465A1 CN 2018097118 W CN2018097118 W CN 2018097118W WO 2019218465 A1 WO2019218465 A1 WO 2019218465A1
Authority
WO
WIPO (PCT)
Prior art keywords
behavior
target
user
target completion
completion
Prior art date
Application number
PCT/CN2018/097118
Other languages
English (en)
French (fr)
Inventor
马豪华
张阳
Original Assignee
平安科技(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 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2019218465A1 publication Critical patent/WO2019218465A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/10Office automation; Time management
    • G06Q10/109Time management, e.g. calendars, reminders, meetings or time accounting
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"

Definitions

  • the present application belongs to the field of data processing technologies, and in particular, to a user behavior prompt generating method, device, terminal device and medium.
  • the traditional life assistant application in the market can only prompt the user to behave at a specific time according to the reminder rules set by the user. For example, the user is prompted to exercise at 9:00 in the morning, and the user is prompted to sleep at 10:00 in the evening. Wait.
  • these existing life assistant class applications are extremely passive to follow the user's settings or user state trigger prompts, and can not actively prompt the user for tips or suggestions according to the actual situation of the user. Therefore, the application of the life assistant class in the prior art is less intelligent, and it is impossible to intelligently provide behavior suggestion suggestions for the user's life and work according to the actual situation of the user.
  • the embodiment of the present application provides a user behavior prompt generation method and a terminal device, so as to solve the problem that the application program is less intelligent in the prior art, and cannot provide behavior prompts for the user's life and work according to the actual situation of the user. Suggested question.
  • a first aspect of the embodiment of the present application provides a method for generating a user behavior prompt, including:
  • An error behavior warning is generated according to a time point at which the target completion time is different from the corresponding target completion time corresponding to each of the target completion behaviors.
  • a second aspect of the embodiment of the present application provides a user behavior prompt generating apparatus, including:
  • a plan acquisition module configured to acquire a work behavior plan of the user in the first preset time period, where the work behavior plan stores the target completion behavior of the user in the first preset time period, and each Determining a target completion time of the target completion behavior within the first preset time period;
  • a behavior prediction module configured to filter target behavior history data corresponding to the target completion behavior from the historical behavior data of the user, and perform deep learning on the target behavior history data to obtain the first preset time period a behavior prediction data corresponding to each of the target completion behaviors, wherein the behavior prediction data includes a prediction completion time of the target completion behavior;
  • the prompt generating module is configured to generate an error behavior warning according to a time point at which the target completion time is different from the corresponding target completion time according to each of the target completion behaviors.
  • a third aspect of the embodiments of the present application provides a terminal device, including a memory, a processor, and the computer storing computer readable instructions executable on the processor, where the processor executes the computer The following steps are implemented when reading the instruction:
  • the target behavior history data corresponding to the target completion behavior from the historical behavior data of the user, and performing deep learning on the target behavior history data to obtain the first preset time period, and each of the a behavior prediction data corresponding to the target completion behavior, wherein the behavior prediction data includes a prediction completion time of the target completion behavior;
  • An error behavior warning is generated according to a time point at which the target completion time is different from the corresponding target completion time corresponding to each of the target completion behaviors.
  • a fourth aspect of the embodiments of the present application provides a computer readable storage medium storing computer readable instructions, wherein the computer readable instructions are implemented by at least one processor The following steps:
  • the target behavior history data corresponding to the target completion behavior from the historical behavior data of the user, and performing deep learning on the target behavior history data to obtain the first preset time period, and each of the a behavior prediction data corresponding to the target completion behavior, wherein the behavior prediction data includes a prediction completion time of the target completion behavior;
  • An error behavior warning is generated according to a time point at which the target completion time is different from the corresponding target completion time corresponding to each of the target completion behaviors.
  • the present invention has the beneficial effects that, in consideration of the user's own work schedule, it is often difficult to effectively implement these plans in actual situations, and the user's own behavioral habits are problematic.
  • the behavioral habits conflict with the behavioral behavior plan.
  • the user's current work habit is to eat dinner at 10 o'clock every night, but the plan is to eat on time every day at 6 o'clock in the afternoon, although there is a plan.
  • the user since the user is accustomed to eating at 10 o'clock, even if the user is reminded to eat at 6 o'clock, he still wants to eat at 10 o'clock, and at this time, it is very likely to undermine the plan.
  • the embodiment of the present application predicts the possible behavior of the user during the work plan according to the behavior data of the user's actual history, and predicts the behavior.
  • the completion time is different from the work schedule, and a corresponding behavior error warning is generated to inform the user that the conflict cannot be performed at this time.
  • the above plan is to eat dinner at 6 pm, but it is predicted that the user may still be at 10 pm. Will eat, at this time need to warn the user at 10 o'clock at night to eat, in order to ensure the normal implementation of the work schedule.
  • the behavior prompt of the specific event can be performed according to the time point set by the user.
  • the behavior of the user can be predicted according to the actual living habit of the user, and the behavior should not be performed.
  • the corresponding warning prompt is issued in time, the purpose of intelligently providing behavior suggestions for the user's life and work according to the actual situation of the user is realized, which greatly improves the intelligence degree of the application.
  • FIG. 1 is a schematic flowchart of an implementation process of a user behavior prompt generation method according to Embodiment 1 of the present application;
  • FIG. 2 is a schematic flowchart of an implementation process of a user behavior prompt generation method provided by Embodiment 2 of the present application;
  • FIG. 3 is a schematic flowchart of an implementation process of a user behavior prompt generation method according to Embodiment 3 of the present application;
  • FIG. 4 is a schematic structural diagram of a user behavior prompt generating apparatus provided in Embodiment 4 of the present application.
  • FIG. 5 is a schematic diagram of a terminal device according to Embodiment 5 of the present application.
  • FIG. 1 is a flowchart showing an implementation of a method for generating a user behavior prompt according to Embodiment 1 of the present application, which is described in detail as follows:
  • S101 Acquire a work behavior plan of the user in the first preset time period, where the work behavior plan stores the target completion behavior of the user in the first preset time period, and each target completion behavior is within the first preset time period.
  • the goal completion time Acquire a work behavior plan of the user in the first preset time period, where the work behavior plan stores the target completion behavior of the user in the first preset time period, and each target completion behavior is within the first preset time period. The goal completion time.
  • the work behavior plan refers to some behaviors that the user plans to complete in the future, including but not limited to, such as sports plans, diet plans and work schedules, such as plans to start at 8 am every day in the next month. Run for half an hour, go to bed at 11 o'clock every night, eat at 7:30, 12:00 and 17:30 every day.
  • the behavioral behavior plan may be set and input by the user, or may be automatically generated after analyzing the data according to the physical condition of the user, which is not limited herein, and may be determined by the actual application.
  • the preset time period refers to a future period after the current time, such as the next month, which is specifically selected by the user or the technician according to actual needs.
  • S102 Filter target behavior history data corresponding to the target completion behavior from the historical behavior data of the user, and perform deep learning on the target behavior history data, and obtain a one-to-one correspondence with each target completion behavior in the first preset time period.
  • Behavior prediction data which includes the predicted completion time of the target completion behavior.
  • the target behavior history data refers to the historical data of the completion status of the user's completion behavior before the current time, such as the historical data in which the user gets up, runs, and sleeps at any time.
  • the data source can be input by the user or obtained from other data sources.
  • the wearable device can be used to collect and record the daily life behavior of the user. Since the target behavior history data is used for the prediction of the user behavior, considering that the data of the long-term data has little practical significance for the user prediction, and may even adversely affect the accuracy of the prediction of the user behavior, the embodiment of the present application
  • the historical data of the target behavior refers to a certain period of time before the current time, and the specific time period can be selected by a technician according to the actual situation.
  • a behavioral action plan may include one or more behavioral plans, such as a running plan, a breakfast plan, and a sleep plan, which correspond to different goal completion behaviors. Due to the actual situation, the regularity of these target completion behaviors in the user's life is strong, such as the time of eating breakfast and the time of sleeping, the time of running every week, etc., the regularity is strong, therefore, in theory, it can be utilized
  • the user behavior history data is used to complete the analysis and prediction of the completion behavior of these goals, so as to realize the data such as the time for predicting the user to complete the completion behavior of these goals in the future time.
  • the independence of these target completion behaviors is strong, it is preferable to perform predictions in units of specific behaviors, and it is possible to achieve an independent and accurate prediction of the completion time of each target completion behavior of the user in the future.
  • the embodiment of the present application uses the user's behavior history data as a training sample to construct a prediction model for the user's target completion behavior, so as to implement the user for a period of time in the future.
  • the prediction of the internal target completion behavior lays the foundation for the subsequent intelligent prompts.
  • the specific deep learning prediction model construction method can be completed by using some existing network models.
  • S103 Generate an error behavior warning according to a time point at which the target completion time corresponding to each target completion behavior differs from the corresponding target completion time.
  • the wrong behavior warning is used to prompt the user, what kind of activities should not be performed at this point in time, so that the user can effectively carry out the work behavior plan, and eliminate some life behaviors that are not compatible with the work behavior plan.
  • the work behavior plan is only a plan, it is difficult to determine whether the user can complete the plan in a timely and effective manner, but it is certain that the user's historical behavior habits are difficult to change in time, such as the user.
  • I am used to the user's current work habits is to eat dinner at 10 o'clock every night, but the plan is to eat on time every day at 6 o'clock in the afternoon, although there is a plan, but because the user is used to eating 10 o'clock, even if the user is reminded to eat at 6 o'clock, At 10 o'clock, he still wants to eat. Therefore, if the behavioral event prompt is directly based on the work schedule, such as prompting the user to eat at 6 pm, but when the original work behavior event, such as the above 10 pm, the user may continue the historical behavior, especially For some users who have weak self-control ability, they are even more unaware.
  • the user behavior history data is used to identify that the predicted user performs the target completion behavior in a future period of time. Time, at the same time, compare it with the target completion time of the target completion behavior set in the work behavior plan. If there is a difference, it means that according to the user's habit, there is a time point that the target completion behavior should not be performed, and the user performs the target completion behavior.
  • the embodiment of the present application predicts the possible behavior of the user during the work plan according to the behavior data of the user's actual history, and completes the predicted behavior completion time.
  • a corresponding behavioral error warning is generated to inform the user that the conflict cannot be performed at this time, so as to ensure the normal execution of the work schedule.
  • the behavior prompt of the specific event can be performed according to the time point set by the user.
  • the behavior of the user can be predicted according to the actual living habit of the user, and the behavior should not be performed.
  • the corresponding warning prompt is issued in time, the purpose of intelligently providing behavior suggestions for the user's life and work according to the actual situation of the user is realized, which greatly improves the intelligence degree of the application.
  • the action plan of the user in the first preset time period is obtained, including:
  • personal information includes, but is not limited to, the user's gender, age, weight, current place of residence, place of birth, preferences, and hobbies.
  • the weather forecast data refers to the weather every day for the first preset time period in the future.
  • the executable probability refers to the probability probability that the work behavior can be executed in a future period of time.
  • the work behavior plan contains some behavior plans that the user hopes to complete in the future.
  • not all behaviors are suitable for users, such as user settings. Running at 8 o'clock every day in the next month, but the weather forecast data shows that there will be heavy rain every day in the next week, so the user is not suitable for running in the next week, and the executable probability of running is lower.
  • the user is set according to the user's personal information and weather forecast data.
  • the work behavior plan is revised so that what is ultimately obtained is a realistic work plan.
  • the specific calculation method of the executable probability can be set by the technician.
  • the method includes:
  • the technician may set corresponding restriction information for each target completion behavior, such as a pregnant woman or a user suffering from certain specific diseases, which is not suitable for running, etc., and directly according to these rules, it is possible to quickly determine whether the user is Suitable for goal completion behavior.
  • the executable probability of the target completion behaviors is set to one below the probability threshold. value. For example, the interest behavior plan is running at 8:00 every morning in the next month, but the weather forecast data shows that it will rain in the next week, so in the next week, it is not suitable for running at 8:00 in the morning.
  • the executable probability value of the run is set to a specific value below the probability threshold to inform subsequent modification of the target completion time of the run.
  • the target completion time of the target completion behavior whose executable probability is less than the probability threshold is adjusted. For the target completion behavior with an executable probability of 0, it will be deleted directly from the work behavior plan, that is, all its target completion time is set to null. Corresponding to the target completion behavior of non-zero but less than the probability threshold, the target completion time is modified according to the specific situation of the daily weather forecast data in the first preset time, for example, if one day is not suitable for running, it will directly change the day. Running cancellation, if it shows that it will rain when the original target is completed, but after one hour, when the weather is fine, the target completion time will be changed to the time after the original target completion time.
  • the specific modification method/rule can have The technicians set their own according to the actual situation.
  • the method when the target behavior history data is deeply learned, and the behavior prediction data corresponding to each target completion behavior in the first preset time period is obtained, the method includes:
  • each target behavior history data is carried out, and the behavior prediction model corresponding to the target completion behavior is obtained, and the behavior prediction model is used to analyze and predict each target completion behavior in the first preset time period. Obtain behavioral prediction data that correspond one-to-one with the target completion behavior.
  • the learning behavior model is separately performed based on the target behavior history data corresponding to each target completion behavior, such as the historical running data as a training sample, and the independent learning training modeling is performed to obtain the running corresponding prediction model. .
  • the resulting prediction modules are reused to predict a pair of target completion behaviors.
  • the specific deep learning modeling method can be set by the technician according to actual needs, including but not limited to learning modeling using BP neural network.
  • a behavior prediction model process is constructed for the target completion behavior, including:
  • S301 Analyze a historical completion time of the target completion behavior in the target behavior history data, determine a behavior repetition period corresponding to the target completion behavior, and divide the target behavior history data into N according to a time sequence in which the target behavior history data is stored.
  • the length of time is the target behavior history data segment of the behavior repetition period, and N period sample data is obtained.
  • S302 Perform model training based on N periodic sample data to obtain a behavior prediction model.
  • the cycle of many work activities is not one day, such as running plan and wake up time, often a cycle of one week, so if you use one day as a cycle for each goal completion behavior, it may make The accuracy of the forecast is low. Therefore, in the embodiment of the present application, the corresponding behavior repetition history is first identified according to the target behavior history data of each target completion behavior, for example, the minimum repetition of the running behavior is identified according to the historical data of the daily running of the user in the past year. According to the cycle, the data segment is divided into the target behavior history data of the target completion behavior, and the corresponding N cycle sample data is obtained. Since the regularity of user behavior in the repetition period is strong, training it as sample data will result in a relatively high prediction accuracy.
  • the data of the same day in the period of each period sample data may be separately extracted, for example, if the running period is one week, then At this time, the data of Monday, Tuesday, ..., and Sunday will be extracted separately, and the corresponding 7 sets of periodic sample data will be obtained. Then, based on the small periodic sample data, respectively, modeling is performed to obtain a plurality of corresponding prediction models, and finally, in the prediction, the prediction is performed separately in units of one day, as described above, for the first preset time, every Monday, Forecasts are made separately on Tuesday, ..., and Sunday to obtain corresponding forecast data for each day in the first preset time.
  • each target completion behavior corresponds to one or more behavior prediction sub-models for predicting behaviors of each day in the repetition period to improve the accuracy of the prediction.
  • the method includes: performing update training on the behavior prediction model based on the behavior data of the target completion behavior of the user in the first preset time period.
  • the actual behavior data of the user's completion behavior for the first preset time is recorded, such as the actual running data in the first preset time, and the prediction model is updated and trained based on the data. To improve the prediction accuracy of the prediction model.
  • FIG. 4 is a structural block diagram of the user behavior prompt generating apparatus provided by the embodiment of the present application. For the convenience of description, only parts related to the embodiment of the present application are shown.
  • the user behavior prompt generating device illustrated in FIG. 4 may be the execution subject of the user behavior prompt generating method provided in the foregoing first embodiment.
  • the user behavior prompt generating apparatus includes:
  • the plan acquisition module 41 is configured to acquire a work behavior plan of the user in the first preset time period, where the work behavior plan stores the target completion behavior of the user in the first preset time period, and each The target completion behavior is a target completion time within the first predetermined time period.
  • the behavior prediction module 42 is configured to filter target behavior history data corresponding to the target completion behavior from the historical behavior data of the user, and perform deep learning on the target behavior history data to obtain the first preset time. Within the segment, behavior prediction data corresponding to each of the target completion behaviors, the behavior prediction data including a prediction completion time of the target completion behavior.
  • the prompt generating module 43 is configured to generate an error behavior warning according to a time point in the prediction completion time corresponding to each of the target completion behaviors that is different from the target completion time corresponding thereto.
  • plan obtaining module 41 includes:
  • the behavior prediction module 42 includes:
  • the target completion behavior is separately analyzed and predicted, and the behavior prediction data corresponding to the target completion behavior is obtained one by one.
  • the behavior prediction module 42 includes:
  • the behavior history data is divided into N target behavior history data segments whose time lengths are all of the behavior repetition periods, and N period sample data are obtained.
  • Model training is performed based on the N period sample data to obtain the behavior prediction model.
  • the user behavior prompt generating device further includes:
  • first, second, and the like are used in the text to describe various elements in the embodiments of the present application, these elements should not be limited by these terms. These terms are only used to distinguish one element from another.
  • the first contact can be named a second contact, and similarly, the second contact can be named the first contact without departing from the scope of the various described embodiments. Both the first contact and the second contact are contacts, but they are not the same contact.
  • FIG. 5 is a schematic diagram of a terminal device according to an embodiment of the present application.
  • the terminal device 5 of this embodiment includes a processor 50, a memory 51 in which computer readable instructions 52 executable on the processor 50 are stored.
  • the processor 50 executes the computer readable instructions 52
  • the steps in the foregoing embodiments of the respective user behavior prompt generation methods are implemented, such as steps 101 to 103 shown in FIG.
  • the processor 50 executes the computer readable instructions 52
  • the functions of the modules/units in the various apparatus embodiments described above are implemented, such as the functions of the modules 41 to 43 shown in FIG.
  • the terminal device 5 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
  • the terminal device may include, but is not limited to, a processor 50 and a memory 51. It will be understood by those skilled in the art that FIG. 5 is only an example of the terminal device 5, does not constitute a limitation of the terminal device 5, may include more or less components than the illustrated, or combine some components, or different components.
  • the terminal device may further include an input transmitting device, a network access device, a bus, and the like.
  • the so-called processor 50 can be a central processing unit (Central Processing Unit, CPU), can also be other general purpose processors, digital signal processors (DSP), application specific integrated circuits (Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the memory 51 may be an internal storage unit of the terminal device 5, such as a hard disk or a memory of the terminal device 5.
  • the memory 51 may also be an external storage device of the terminal device 5, such as a plug-in hard disk equipped on the terminal device 5, a smart memory card (SMC), and a secure digital (SD). Card, flash card, etc. Further, the memory 51 may also include both an internal storage unit of the terminal device 5 and an external storage device.
  • the memory 51 is configured to store the computer readable instructions and other programs and data required by the terminal device.
  • the memory 51 can also be used to temporarily store data that has been sent or is about to be transmitted.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated modules/units if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium.
  • the present application implements all or part of the processes in the foregoing embodiments, and may also be implemented by computer readable instructions, which may be stored in a computer readable storage medium.
  • the computer readable instructions when executed by a processor, may implement the steps of the various method embodiments described above.
  • the computer readable instructions comprise computer readable instruction code, which may be in the form of source code, an object code form, an executable file or some intermediate form or the like.
  • the computer readable medium may include any entity or device capable of carrying the computer readable instruction code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), electrical carrier signals, telecommunications signals, and software distribution media. It should be noted that the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in a jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, computer readable media Does not include electrical carrier signals and telecommunication signals.

Abstract

本申请提供了一种用户行为提示生成方法、装置、终端设备及介质,适用于数据处理技术领域,该方法包括:获取用户在第一预设时间段内的作息行为计划;从用户的历史行为数据中筛选出目标完成行为对应的目标行为历史数据,并对目标行为历史数据进行深度学习,得到第一预设时间段内,与每个目标完成行为一一对应的行为预测数据,行为预测数据中包含目标完成行为的预测完成时间;根据每个目标完成行为对应的预测完成时间中,与其对应的目标完成时间存在差异的时间点,生成错误行为警告。本申请实施例实现了根据用户的实际情况智能化地为用户生活作息提供行为建议的目的,极大地提升了应用程序的智能化程度。

Description

一种用户行为提示生成方法、装置、终端设备及介质
本申请要求于2018年05月14日提交中国专利局、申请号为201810457094.3 、发明名称为“一种用户行为提示生成方法及终端设备”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
本申请属于数据处理技术领域,尤其涉及一种用户行为提示生成方法、装置、终端设备及介质。
背景技术
市面上传统的生活助手类的应用程序,都是只能根据用户自行设定的提醒规则,在特定的时间点对用户进行行为提示,如早上9点提示用户进行锻炼,晚上10点提示用户睡觉等。然而,这些现有的这些生活助手类的应用程序都是极其被动的跟随用户的设定或者用户状态触发提示,而无法根据用户的实际情况来主动为用户发出提示或建议。因此,现有技术中的生活助手类的应用程序智能化程度较为低下,无法根据用户的实际情况智能化地为用户生活作息提供行为提示建议。
技术问题
有鉴于此,本申请实施例提供了一种用户行为提示生成方法及终端设备,以解决现有技术中应用程序智能化程度低下,无法根据用户的实际情况智能化地为用户生活作息提供行为提示建议的问题。
技术解决方案
本申请实施例的第一方面提供了一种用户行为提示生成方法,包括:
获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
本申请实施例的第二方面提供了一种用户行为提示生成装置,包括:
计划获取模块,用于获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
行为预测模块,用于从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
提示生成模块,用于根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
本申请实施例的第三方面提供了一种终端设备,包括存储器、处理器,所述存储器上存储有可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时实现如下步骤:
获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
本申请实施例的第四方面提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机可读指令,其特征在于,所述计算机可读指令被至少一个处理器执行时实现如下步骤:
获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
有益效果
本申请实施例与现有技术相比存在的有益效果是:考虑到用户虽然会有自己的作息行为计划,但实际情况中往往难以有效的执行这些计划,并且,由于用户本身作息行为习惯的问题,实际情况经常会存在作息行为习惯与作息行为计划相冲突的地方,如用户现在作息行为习惯是每天晚上10点吃晚饭,但计划希望每天下午6点准时吃饭准时,此时虽然有了计划,但由于用户习惯了10点吃饭,即使6点提醒了用户吃饭,到10点其仍会想吃东西,此时就极有可能破坏制定的计划。
因此,为了实现根据用户的实际情况进行智能化的提示,以保证作息行为计划的正常执行,本申请实施例会根据用户实际历史的行为数据来预测用户在作息计划期间可能的行为,并对预测行为完成时间对与作息行为计划不同的时间点,生成对应的行为错误警告的提示,告知用户此时不能进行冲突的行为,如上述的计划是晚上6点吃晚饭,但预测晚上10点用户仍可能会吃东西,此时需要在晚上10点警告用户不能吃东西,以保证作息行为计划的正常执行。相对现有技术中仅能根据用户设定的时间点来进行特定事件的行为提示,本申请实施例中还能根据用户实际的生活习惯,来预测用户的行为,并在不应当进行某项行为时及时发出相应的警告提示,从而实现了根据用户的实际情况智能化地为用户生活作息提供行为建议的目的,极大地提升了应用程序的智能化程度。
附图说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例或现有技术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性的前提下,还可以根据这些附图获得其他的附图。
图1是本申请实施例一提供的用户行为提示生成方法的实现流程示意图;
图2是本申请实施例二提供的用户行为提示生成方法的实现流程示意图;
图3是本申请实施例三提供的用户行为提示生成方法的实现流程示意图;
图4是本申请实施例四提供的用户行为提示生成装置的结构示意图;
图5是本申请实施例五提供的终端设备的示意图。
本发明的实施方式
以下描述中,为了说明而不是为了限定,提出了诸如特定系统结构、技术之类的具体细节,以便透彻理解本申请实施例。然而,本领域的技术人员应当清楚,在没有这些具体细节的其它实施例中也可以实现本申请。在其它情况中,省略对众所周知的系统、装置、电路以及方法的详细说明,以免不必要的细节妨碍本申请的描述。
为了说明本申请所述的技术方案,下面通过具体实施例来进行说明。
图1示出了本申请实施例一提供的用户行为提示生成方法的实现流程图,详述如下:
S101,获取用户在第一预设时间段内的作息行为计划,作息行为计划中存储有用户在第一预设时间段的目标完成行为,以及每个目标完成行为在第一预设时间段内的目标完成时间。
其中,作息行为计划是指用户在未来的一段时间内,计划完成的一些行为,包括但不限于如运动计划、饮食计划和作息计划等,如计划在未来的一个月内,每天早上8点开始跑步半小时,每天晚上11点睡觉,每天7:30、12:00以及17:30准时吃饭等。该作息行为计划既可以是用户自行设置并输入的,也可以是根据用户的身体状况的一些数据进行分析后自动生成的,此处不予限定,具体可由实际应用情况决定。在本申请实施例中,预设时间段是指在当前时间之后的未来的一段时间,如未来一个月,具体需由用户或技术人员根据实际需求选取。
S102,从用户的历史行为数据中筛选出目标完成行为对应的目标行为历史数据,并对目标行为历史数据进行深度学习,得到第一预设时间段内,与每个目标完成行为一一对应的行为预测数据,行为预测数据中包含目标完成行为的预测完成时间。
其中目标行为历史数据,是指用户在当前时间之前对这些目标完成行为的完成情况历史数据,如历史数据中用户是每天什么时间点起床、跑步以及睡觉的。其数据来源既可以是用户自行输入的,或者是从其他数据来源获取的,如可以是利用可穿戴设备等,对用户日常生活行为进行采集记录的。由于目标行为历史数据是用于对用户行为的预测,但考虑到过久的数据对用户预测的实际意义较小,甚至可能会反过来影响对用户行为预测的准确性,因此,本申请实施例中的目标行为历史数据,是指当前时间前的某一段时间,其具体的时间段选取可有技术人员根据实际情况设定。
一个作息行为计划中可能包含一种或多种行为计划,如可能同时包含跑步计划,早餐计划和睡觉计划等,这些行为计划对应着不同的目标完成行为。由于实际情况中,这些目标完成行为在用户生活中的规律性较强,如每天吃早餐的时间和睡觉的时间,每周跑步的时间等,规律性均较强,因此,理论上是可以利用用户行为历史数据来完成对这些目标完成行为的分析预测的,以实现预知用户在未来时间内完成这些目标完成行为的时间等数据。同时,考虑到这些目标完成行为之间独立性较强,因此,优选地,以每个具体的行为为单位来进行预测,可以实现对用户未来一段时间各个目标完成行为完成时间的独立准确预测。
因此,为了为用户提供更加个性化智能化的行为提示,本申请实施例中会基于用户的行为历史数据来作为训练样本,构建对用户目标完成行为的预测模型,以实现对用户在未来一段时间内目标完成行为的预测,为后续智能化的提示奠定基础。其中,具体的深度学习预测模型构建的方法,既可以选用已有的一些网络模型来完成。
S103,根据每个目标完成行为对应的预测完成时间中,与其对应的目标完成时间存在差异的时间点,生成错误行为警告。
其中,错误行为警告用于提示用户,在该时间点不应当进行何种活动,以使得用户能够有效地执行作息行为计划,杜绝与作息行为计划不和的一些生活行为。由于作息行为计划仅仅是一项计划,实际情况中难以确定用户是否能及时有效地完成这个计划,但可以确定的是,无论作息计划如何,用户的历史作息行为习惯都难以及时发生变化,如用户习惯用户现在作息行为习惯是每天晚上10点吃晚饭,但计划希望每天下午6点准时吃饭准时,此时虽然有了计划,但由于用户习惯了10点吃饭,即使6点提醒了用户吃饭,到10点其仍会想吃东西。因此,若直接只是根据作息行为计划进行行为事件提示,如在下午6点提示用户吃饭,但到了原作息行为事件时,如上述的晚上10点时,用户仍可能继续历史的行为习惯,特别是对一些本身自制能力较弱的用户,更是不理。
因此,为了实现对用户个性化智能化的提示,以提升对用户生活作息提示建议的有效性,本申请实施例中会基于用户行为历史数据来识别预测用户在未来一段时间内执行目标完成行为的时间,同时将其与作息行为计划中设定的目标完成行为的目标完成时间进行对比,若存在差异,说明根据用户习惯,有一个不应当进行目标完成行为的时间点,是用户进行目标完成行为的习惯时间点,但此时用户不应当进行目标完成行为,这会对正常的作息行为计划造成极大影响,例如,发现根据预测用户在每天晚上10点会吃晚饭,但根据作息行为计划,应该是6点钟吃晚饭的,因此,即使是6点钟已经提示过了用户应当吃晚饭,为了防止用户在10点钟由于习惯继续吃东西,本申请实施例也会在每天晚上10点生成对应的错误行为警告,告知用户晚上10点不能吃东西。
为了实现根据用户的实际情况进行智能化的提示,以保证作息行为计划的正常执行,本申请实施例会根据用户实际历史的行为数据来预测用户在作息计划期间可能的行为,并对预测行为完成时间对与作息行为计划不同的时间点,生成对应的行为错误警告的提示,告知用户此时不能进行冲突的行为,以保证作息行为计划的正常执行。相对现有技术中仅能根据用户设定的时间点来进行特定事件的行为提示,本申请实施例中还能根据用户实际的生活习惯,来预测用户的行为,并在不应当进行某项行为时及时发出相应的警告提示,从而实现了根据用户的实际情况智能化地为用户生活作息提供行为建议的目的,极大地提升了应用程序的智能化程度。
作为本申请实施例二,如图2所示,获取用户在第一预设时间段内的作息行为计划,包括:
S201,获取用户的个人信息、第一预设时间段的天气预报数据以及用户输入作息行为计划。
其中,个人信息包括但不限于用户的性别、年龄、体重、现居地、出生地、偏好以及爱好的运动等。天气预报数据是指未来的第一预设时间段内每天的天气如何。
S202,基于个人信息以及天气预报数据,计算作息行为计划中包含的目标完成行为的可执行概率,并对可执行概率低于概率阈值的目标完成行为的目标完成时间进行调整。
其中,可执行概率,是指作息行为在未来的一段时间内可以执行的可能性概率。作息行为计划中包含着用户在未来一段时间内希望完成的一些行为计划,但实际情况中,由于用户个人身体情况的差异以及天气情况的不同,并不是所有行为都适合用户做的,如用户设置了未来一个月每天8点跑步,但天气预报数据显示未来一周内天天要下大雨,因此未来一周用户内并不适合跑步,此时跑步的可执行概率就较低。因此为了保证最终得到的作息行为计划能真正的满足用户的实际需求,以提高对用户生活作息提示的有效性,本申请实施例中,会根据用户的个人信息以及天气预报数据来对用户设置的作息行为计划进行修正,以使得最终得到的是真实可用的作息行为计划。
其中,可执行概率具体的计算方法,可以由技术人员自行设定。作为本申请计算可执行概率的一种具体实现方式,包括:
根据用户的个人信息以分析作息行为计划中,是否存在不适宜用户的目标完成行为,并将不适宜的目标完成行为的可执行概率设置为0。例如,可以由技术人员对每种目标完成行为设置对应的限制信息,如孕妇或患有某些特定疾病的用户不适合跑步等,此时直接根据这些规则,即可很快地判断出用户是否适宜进行目标完成行为。
对于非不适宜的目标完成行为,判断这些所有目标完成时间对应的天气预报数据,是否都适合进行这些目标完成行为,若非都适合,将这些目标完成行为的可执行概率设置为概率阈值以下的一个值。如,当作息行为计划中是未来一个月内每天早上8点跑步,但天气预报数据显示,未来一周内都会下雨,因此未来一周内,在早上8点都已经不适合跑步了,此时会将跑步的可执行概率值设置为低于概率阈值的一个具体值,以告知后续需要对跑步的目标完成时间进行修改。
对可执行概率小于概率阈值的目标完成行为的目标完成时间进行调整。对于可执行概率为0的目标完成行为,直接将从作息行为计划中删除,即将其所有的目标完成时间均设置为空。对应非0但小于概率阈值的目标完成行为,根据在第一预设时间内每天天气预报数据的具体情况,对目标完成时间进行修改,如,若某天不适合进行跑步,则直接将改天的跑步取消,若显示在原目标完成时间时会下雨,但一个小时之后,天气晴朗时候跑步,此时会将目标完成时间修改为原目标完成时间之后的时间,具体的修改方法/规则,可以有技术人员根据实际情况自行设定。
作为本申请的一个实施例,在对目标行为历史数据进行深度学习,得到第一预设时间段内,与每个目标完成行为一一对应的行为预测数据时,包括:
对每个目标行为历史数据分别进行深度学习,得到与目标完成行为一一对应的行为预测模型,并利用行为预测模型,对第一预设时间段内的每个目标完成行为分别进行分析预测,得到与目标完成行为一一对应的行为预测数据。
由上述说明可知,对每个目标完成行为进行独立预测,可提高对目标完成行为预测的准确性。因此,本申请实施例中会基于每个目标完成行为对应的目标行为历史数据分别进行学习建模,如对历史的跑步数据作为训练样本,进行独立学习训练建模,以得到跑步对应的预测模型。再利用得到的这些预测模块,一一对目标完成行为进行预测。其中,具体深度学习建模方法,可由技术人员根据实际需求进行设定,包括但不限于如利用BP神经网络进行学习建模。
作为本申请实施例三,如图3所示,对目标完成行为构建行为预测模型过程,包括:
S301,对目标行为历史数据中目标完成行为的历史完成时间进行分析,确定出目标完成行为对应的行为重复周期,并按照存储所述目标行为历史数据的时间顺序,将目标行为历史数据划分为N个时间长度均为行为重复周期的目标行为历史数据段,得到N个周期样本数据。
S302,基于N个周期样本数据进行模型训练,得到行为预测模型。
由于实际情况中,很多作息行为的周期都不是一天,如跑步计划和起床时间,往往都是以一周为周期的行为,因此,若对每个目标完成行为都直接使用一天作为周期,可能会使得预测的准确率较低。因此,在本申请实施例中会首先根据每个目标完成行为的目标行为历史数据来识别出其对应的行为重复周期,如根据用户在过去年内每天跑步的历史数据,识别出跑步行为的最小重复周期,再根据这个周期,来对目标完成行为的目标行为历史数据进行数据段划分,得到对应的N个周期样本数据。由于重复周期内用户行为的规律性较强,因此,将其作为样本数据进行训练,得到的预测准确率也会相对较强。
作为本申请的一个优选实施例,在基于N个周期样本数据进行模型构建时,可以将每个周期样本数据中,周期内对应的同一天的数据分别提取,如假设跑步的周期是一周,那此时就会将每周一、周二、…、周日的数据分别提取,得到对应的7组周期样本数据。再基于这些小的周期样本数据分别进行建模,得到多个对应的预测模型,最后在预测时,以一天为单位进行分别进行预测,如上述的,对第一预设时间内的每周一、周二、…、周日分别进行预测,以得到第一预设时间内的每天对应的预测数据。在本申请实施例中,每个目标完成行为对应着一个或多个行为预测子模型,用于对重复周期内每一天的行为预测,以提升对预测的准确性。
作为本申请的一个实施例,在生成提示之后,包括:基于用户在第一预设时间段内目标完成行为的行为数据,对行为预测模型进行更新训练。
在生成提示之后,对第一预设时间内用户对这些目标完成行为的实际行为数据进行记录,如实际在第一预设时间内每天的跑步数据,并基于这些数据对预测模型进行更新训练,以提升预测模型的预测准确率。
对应于上文实施例的方法,图4示出了本申请实施例提供的用户行为提示生成装置的结构框图,为了便于说明,仅示出了与本申请实施例相关的部分。图4示例的用户行为提示生成装置可以是前述实施例一提供的用户行为提示生成方法的执行主体。
参照图4,该用户行为提示生成装置包括:
计划获取模块41,用于获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间。
行为预测模块42,用于从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间。
提示生成模块43,用于根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
进一步地,所述计划获取模块41,包括:
获取所述用户的个人信息、所述第一预设时间段的天气预报数据以及所述用户输入所述作息行为计划。
基于所述个人信息以及所述天气预报数据,计算所述作息行为计划中包含的所述目标完成行为的可执行概率,并对所述可执行概率低于概率阈值的所述目标完成行为的所述目标完成时间进行调整。
进一步地,所述行为预测模块42,包括:
对每个所述目标行为历史数据分别进行深度学习,得到与所述目标完成行为一一对应的行为预测模型,并利用所述行为预测模型,对所述第一预设时间段内的每个所述目标完成行为分别进行分析预测,得到与所述目标完成行为一一对应的所述行为预测数据。
进一步地,所述行为预测模块42,包括:
对所述目标行为历史数据中所述目标完成行为的历史完成时间进行分析,确定出所述目标完成行为对应的行为重复周期,并按照存储所述目标行为历史数据的时间顺序,将所述目标行为历史数据划分为N个时间长度均为所述行为重复周期的目标行为历史数据段,得到N个周期样本数据。
基于所述N个周期样本数据进行模型训练,得到所述行为预测模型。
该用户行为提示生成装置,还包括:
基于所述用户在所述第一预设时间段内所述目标完成行为的行为数据,对所述行为预测模型进行更新训练。
本申请实施例提供的用户行为提示生成装置中各模块实现各自功能的过程,具体可参考前述图1所示实施例一的描述,此处不再赘述。
应理解,上述实施例中各步骤的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
还应理解的是,虽然术语“第一”、“第二”等在文本中在一些本申请实施例中用来描述各种元素,但是这些元素不应该受到这些术语的限制。这些术语只是用来将一个元素与另一元素区分开。例如,第一接触可以被命名为第二接触,并且类似地,第二接触可以被命名为第一接触,而不背离各种所描述的实施例的范围。第一接触和第二接触都是接触,但是它们不是同一接触。
图5是本申请一实施例提供的终端设备的示意图。如图5所示,该实施例的终端设备5包括:处理器50、存储器51,所述存储器51中存储有可在所述处理器50上运行的计算机可读指令52。所述处理器50执行所述计算机可读指令52时实现上述各个用户行为提示生成方法实施例中的步骤,例如图1所示的步骤101至103。或者,所述处理器50执行所述计算机可读指令52时实现上述各装置实施例中各模块/单元的功能,例如图4所示模块41至43的功能。
所述终端设备5可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述终端设备可包括,但不仅限于,处理器50、存储器51。本领域技术人员可以理解,图5仅仅是终端设备5的示例,并不构成对终端设备5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,例如所述终端设备还可以包括输入发送设备、网络接入设备、总线等。
所称处理器50可以是中央处理单元(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
所述存储器51可以是所述终端设备5的内部存储单元,例如终端设备5的硬盘或内存。所述存储器51也可以是所述终端设备5的外部存储设备,例如所述终端设备5上配备的插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)等。进一步地,所述存储器51还可以既包括所述终端设备5的内部存储单元也包括外部存储设备。所述存储器51用于存储所述计算机可读指令以及所述终端设备所需的其他程序和数据。所述存储器51还可以用于暂时地存储已经发送或者将要发送的数据。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请实现上述实施例方法中的全部或部分流程,也可以通过计算机可读指令来指令相关的硬件来完成,所述的计算机可读指令可存储于一计算机可读存储介质中,该计算机可读指令在被处理器执行时,可实现上述各个方法实施例的步骤。其中,所述计算机可读指令包括计算机可读指令代码,所述计算机可读指令代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括:能够携带所述计算机可读指令代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、电载波信号、电信信号以及软件分发介质等。需要说明的是,所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在某些司法管辖区,根据立法和专利实践,计算机可读介质不包括电载波信号和电信信号。
以上所述实施例仅用以说明本申请的技术方案,而非对其限制;尽管参照前述实施例对本申请进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使对应技术方案的本质脱离本申请各实施例技术方案的精神和范围,均应包含在本申请的保护范围之内。

Claims (20)

  1. 一种用户行为提示生成方法,其特征在于,包括:
    获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
    从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
    根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
  2. 如权利要求1所述的用户行为提示生成方法,其特征在于,所述获取用户在第一预设时间段内的作息行为计划,包括:
    获取所述用户的个人信息、所述第一预设时间段的天气预报数据以及所述用户输入所述作息行为计划;
    基于所述个人信息以及所述天气预报数据,计算所述作息行为计划中包含的所述目标完成行为的可执行概率,并对所述可执行概率低于概率阈值的所述目标完成行为的所述目标完成时间进行调整。
  3. 如权利要求1所述的用户行为提示生成方法,其特征在于,所述对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,包括:
    对每个所述目标行为历史数据分别进行深度学习,得到与所述目标完成行为一一对应的行为预测模型,并利用所述行为预测模型,对所述第一预设时间段内的每个所述目标完成行为分别进行分析预测,得到与所述目标完成行为一一对应的所述行为预测数据。
  4. 如权利要求3所述的用户行为提示生成方法,其特征在于,对所述目标完成行为构建所述行为预测模型过程,包括:
    对所述目标行为历史数据中所述目标完成行为的历史完成时间进行分析,确定出所述目标完成行为对应的行为重复周期,并按照存储所述目标行为历史数据的时间顺序,将所述目标行为历史数据划分为N个时间长度均为所述行为重复周期的目标行为历史数据段,得到N个周期样本数据;
    基于所述N个周期样本数据进行模型训练,得到所述行为预测模型。
  5. 如权利要求3所述的用户行为提示生成方法,其特征在于,还包括:
    基于所述用户在所述第一预设时间段内所述目标完成行为的行为数据,对所述行为预测模型进行更新训练。
  6. 一种用户行为提示生成装置,其特征在于,包括:
    计划获取模块,用于获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
    行为预测模块,用于从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
    提示生成模块,用于根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
  7. 如权利要求6所述的用户行为提示生成装置,其特征在于,所述计划获取模块,包括:
    获取所述用户的个人信息、所述第一预设时间段的天气预报数据以及所述用户输入所述作息行为计划;
    基于所述个人信息以及所述天气预报数据,计算所述作息行为计划中包含的所述目标完成行为的可执行概率,并对所述可执行概率低于概率阈值的所述目标完成行为的所述目标完成时间进行调整。
  8. 如权利要求6所述的用户行为提示生成装置,其特征在于,所述行为预测模块,还包括:
    对每个所述目标行为历史数据分别进行深度学习,得到与所述目标完成行为一一对应的行为预测模型,并利用所述行为预测模型,对所述第一预设时间段内的每个所述目标完成行为分别进行分析预测,得到与所述目标完成行为一一对应的所述行为预测数据。
  9. 如权利要求8所述的用户行为提示生成装置,其特征在于,所述行为预测模块,还包括:
    对所述目标行为历史数据中所述目标完成行为的历史完成时间进行分析,确定出所述目标完成行为对应的行为重复周期,并按照存储所述目标行为历史数据的时间顺序,将所述目标行为历史数据划分为N个时间长度均为所述行为重复周期的目标行为历史数据段,得到N个周期样本数据;
    基于所述N个周期样本数据进行模型训练,得到所述行为预测模型。
  10. 如权利要求8所述的用户行为提示生成装置,其特征在于,还包括:
    更新训练模块,用于基于所述用户在所述第一预设时间段内所述目标完成行为的行为数据,对所述行为预测模型进行更新训练。
  11. 一种终端设备,其特征在于,所述终端设备包括存储器、处理器,所述存储器上存储有可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时实现如下步骤:
    获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
    从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
    根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
  12. 如权利要求11所述的终端设备,其特征在于,所述获取用户在第一预设时间段内的作息行为计划,包括:
    获取所述用户的个人信息、所述第一预设时间段的天气预报数据以及所述用户输入所述作息行为计划;
    基于所述个人信息以及所述天气预报数据,计算所述作息行为计划中包含的所述目标完成行为的可执行概率,并对所述可执行概率低于概率阈值的所述目标完成行为的所述目标完成时间进行调整。
  13. 如权利要求11所述的终端设备,其特征在于,所述对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,包括:
    对每个所述目标行为历史数据分别进行深度学习,得到与所述目标完成行为一一对应的行为预测模型,并利用所述行为预测模型,对所述第一预设时间段内的每个所述目标完成行为分别进行分析预测,得到与所述目标完成行为一一对应的所述行为预测数据。
  14. 如权利要求13所述的终端设备,其特征在于,对所述目标完成行为构建所述行为预测模型过程,包括:
    对所述目标行为历史数据中所述目标完成行为的历史完成时间进行分析,确定出所述目标完成行为对应的行为重复周期,并按照存储所述目标行为历史数据的时间顺序,将所述目标行为历史数据划分为N个时间长度均为所述行为重复周期的目标行为历史数据段,得到N个周期样本数据;
    基于所述N个周期样本数据进行模型训练,得到所述行为预测模型。
  15. 如权利要求13所述的终端设备,其特征在于,所述处理器执行所述计算机可读指令时还实现如下步骤:
    基于所述用户在所述第一预设时间段内所述目标完成行为的行为数据,对所述行为预测模型进行更新训练。
  16. 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机可读指令,其特征在于,所述计算机可读指令被至少一个处理器执行时实现如下步骤:
    获取用户在第一预设时间段内的作息行为计划,所述作息行为计划中存储有所述用户在所述第一预设时间段的目标完成行为,以及每个所述目标完成行为在所述第一预设时间段内的目标完成时间;
    从所述用户的历史行为数据中筛选出所述目标完成行为对应的目标行为历史数据,并对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,所述行为预测数据中包含所述目标完成行为的预测完成时间;
    根据每个所述目标完成行为对应的所述预测完成时间中,与其对应的所述目标完成时间存在差异的时间点,生成错误行为警告。
  17. 根据权利要求16所述的计算机可读存储介质,其特征在于,所述获取用户在第一预设时间段内的作息行为计划,包括:
    获取所述用户的个人信息、所述第一预设时间段的天气预报数据以及所述用户输入所述作息行为计划;
    基于所述个人信息以及所述天气预报数据,计算所述作息行为计划中包含的所述目标完成行为的可执行概率,并对所述可执行概率低于概率阈值的所述目标完成行为的所述目标完成时间进行调整。
  18. 根据权利要求16所述的计算机可读存储介质,其特征在于,所述对所述目标行为历史数据进行深度学习,得到所述第一预设时间段内,与每个所述目标完成行为一一对应的行为预测数据,包括:
    对每个所述目标行为历史数据分别进行深度学习,得到与所述目标完成行为一一对应的行为预测模型,并利用所述行为预测模型,对所述第一预设时间段内的每个所述目标完成行为分别进行分析预测,得到与所述目标完成行为一一对应的所述行为预测数据。
  19. 根据权利要求18所述的计算机可读存储介质,其特征在于,对所述目标完成行为构建所述行为预测模型过程,包括:
    对所述目标行为历史数据中所述目标完成行为的历史完成时间进行分析,确定出所述目标完成行为对应的行为重复周期,并按照存储所述目标行为历史数据的时间顺序,将所述目标行为历史数据划分为N个时间长度均为所述行为重复周期的目标行为历史数据段,得到N个周期样本数据;
    基于所述N个周期样本数据进行模型训练,得到所述行为预测模型。
  20. 根据权利要求18所述的计算机可读存储介质,其特征在于,还包括:
    基于所述用户在所述第一预设时间段内所述目标完成行为的行为数据,对所述行为预测模型进行更新训练。
PCT/CN2018/097118 2018-05-14 2018-07-25 一种用户行为提示生成方法、装置、终端设备及介质 WO2019218465A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810457094.3 2018-05-14
CN201810457094.3A CN108876284B (zh) 2018-05-14 2018-05-14 一种用户行为提示生成方法及终端设备

Publications (1)

Publication Number Publication Date
WO2019218465A1 true WO2019218465A1 (zh) 2019-11-21

Family

ID=64333924

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/097118 WO2019218465A1 (zh) 2018-05-14 2018-07-25 一种用户行为提示生成方法、装置、终端设备及介质

Country Status (2)

Country Link
CN (1) CN108876284B (zh)
WO (1) WO2019218465A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111143350A (zh) * 2019-11-27 2020-05-12 深圳壹账通智能科技有限公司 企业数据监控方法、装置、计算机设备及存储介质

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109684543B (zh) * 2018-12-14 2021-04-02 北京百度网讯科技有限公司 用户行为预测和信息投放方法、装置、服务器和存储介质
CN109978491B (zh) * 2019-02-12 2024-02-06 平安科技(深圳)有限公司 提醒预测方法、装置、计算机设备及存储介质
WO2020168454A1 (zh) * 2019-02-18 2020-08-27 深圳市欢太科技有限公司 行为推荐方法、装置、存储介质及电子设备
CN110135794A (zh) * 2019-03-25 2019-08-16 阿里巴巴集团控股有限公司 智能审核提醒方法及装置、电子设备及存储介质
CN110362266B (zh) * 2019-07-19 2020-12-18 北京字节跳动网络技术有限公司 提示信息显示方法、系统、电子设备和计算机可读介质
CN110610752B (zh) * 2019-08-23 2023-01-17 北京卡路里信息技术有限公司 运动目标确定方法及装置
CN110719360B (zh) * 2019-09-04 2021-04-20 奇酷互联网络科技(深圳)有限公司 备忘事件的自动生成方法、装置以及存储装置
CN110648115B (zh) * 2019-09-20 2023-09-22 深圳Tcl新技术有限公司 事件提醒方法、装置、设备及存储介质
CN114518797B (zh) * 2020-11-20 2024-02-13 Oppo(重庆)智能科技有限公司 信息推送方法、装置、可穿戴设备及存储介质
CN112488657B (zh) * 2020-12-08 2023-07-28 中国航空结算有限责任公司 一种飞行信息处理方法、装置及设备
CN114564611A (zh) * 2022-02-28 2022-05-31 华南师范大学 一种用于音乐治疗的音乐播放方法、装置及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101620647A (zh) * 2008-07-03 2010-01-06 王晓毅 一种保健管理方法和实现该方法的保健管理系统
CN102804238A (zh) * 2011-12-15 2012-11-28 北京英福生科技有限公司 运动提醒设备及系统
CN106919776A (zh) * 2015-12-24 2017-07-04 中国移动通信集团公司 一种目标活动提醒的方法及装置
CN107239669A (zh) * 2017-06-15 2017-10-10 重庆柚瓣家科技有限公司 健康计划系统
CN107609324A (zh) * 2017-10-11 2018-01-19 上海展扬通信技术有限公司 一种基于智能设备的用户作息的管理方法及管理装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5297672B2 (ja) * 2008-03-25 2013-09-25 シスメックス株式会社 生活習慣改善提案装置、コンピュータプログラムおよび生活習慣改善提案システム
CN106650429A (zh) * 2016-09-30 2017-05-10 北京奇虎科技有限公司 一种用户行为管理方法和系统、以及一种安全桌面
CN106683340A (zh) * 2016-12-15 2017-05-17 歌尔股份有限公司 一种用户行为监测方法和可穿戴设备
CN107833610B (zh) * 2017-10-31 2021-08-06 咪咕互动娱乐有限公司 一种提醒方法、装置及计算机可读存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101620647A (zh) * 2008-07-03 2010-01-06 王晓毅 一种保健管理方法和实现该方法的保健管理系统
CN102804238A (zh) * 2011-12-15 2012-11-28 北京英福生科技有限公司 运动提醒设备及系统
CN106919776A (zh) * 2015-12-24 2017-07-04 中国移动通信集团公司 一种目标活动提醒的方法及装置
CN107239669A (zh) * 2017-06-15 2017-10-10 重庆柚瓣家科技有限公司 健康计划系统
CN107609324A (zh) * 2017-10-11 2018-01-19 上海展扬通信技术有限公司 一种基于智能设备的用户作息的管理方法及管理装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111143350A (zh) * 2019-11-27 2020-05-12 深圳壹账通智能科技有限公司 企业数据监控方法、装置、计算机设备及存储介质

Also Published As

Publication number Publication date
CN108876284B (zh) 2023-04-18
CN108876284A (zh) 2018-11-23

Similar Documents

Publication Publication Date Title
WO2019218465A1 (zh) 一种用户行为提示生成方法、装置、终端设备及介质
Bürkner et al. Ordinal regression models in psychology: A tutorial
Wangmo et al. Ethical concerns with the use of intelligent assistive technology: findings from a qualitative study with professional stakeholders
Garcia-Ceja et al. Mental health monitoring with multimodal sensing and machine learning: A survey
US10452816B2 (en) Method and system for patient engagement
Suhara et al. Deepmood: Forecasting depressed mood based on self-reported histories via recurrent neural networks
US11307892B2 (en) Machine-learning for state determination and prediction
Pashler et al. Editors’ introduction to the special section on replicability in psychological science: A crisis of confidence?
US20180107962A1 (en) Stress and productivity insights based on computerized data
US11508479B2 (en) Automated question generation and response tracking
Trella et al. Designing reinforcement learning algorithms for digital interventions: pre-implementation guidelines
WO2013158768A1 (en) Identifying and forecasting shifts in the mood of social media users
CN103457989A (zh) 信息处理设备、信息处理方法及程序
US20230011521A1 (en) System and method for adjusting hospital unit capacity based on patient-specific variables
US20150148616A1 (en) Systems and methods for probability based risk prediction
US20210151194A1 (en) Dynamic Behavioral Phenotyping for Predicting Health Outcomes
US20230237059A1 (en) Managing engagement methods of a digital assistant while communicating with a user of the digital assistant
Spruijt-Metz et al. Advancing behavioral intervention and theory development for mobile health: the HeartSteps II protocol
Mestdagh et al. m-Path: an easy-to-use and highly tailorable platform for ecological momentary assessment and intervention in behavioral research and clinical practice
Liu et al. Bidirectional imputation of spatial GPS trajectories with missingness using sparse online Gaussian process
US11436549B1 (en) Machine learning system and method for predicting caregiver attrition
US20230043921A1 (en) Sensor-based machine learning in a health prediction environment
US20230008936A1 (en) System and method for adaptive learning for hospital census simulation
US20220230753A1 (en) Techniques for executing transient care plans via an input/output device
Borges et al. Predicting and explaining absenteeism risk in hospital patients before and during COVID-19

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: 18918620

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 24/02/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18918620

Country of ref document: EP

Kind code of ref document: A1