WO2023151215A1

WO2023151215A1 - Prediction model establishment method and device, storage medium, and electronic device

Info

Publication number: WO2023151215A1
Application number: PCT/CN2022/100560
Authority: WO
Inventors: 刘建国; 胡百春
Original assignee: 青岛海尔科技有限公司; 海尔智家股份有限公司
Priority date: 2022-02-08
Filing date: 2022-06-22
Publication date: 2023-08-17
Also published as: CN114493028A

Abstract

The present disclosure provides a prediction model establishment method and device, a storage medium, and an electronic device. The method comprises: acquiring element information when a target object controls a household appliance to perform a target operation; dividing the element information into target sets, wherein the target sets at least comprise: a user attribute information set, a time attribute information set, a position attribute information set, a context information set, and an intention attribute information set; and establishing a behavior prediction model according to the user attribute information set, the time attribute information set, the position attribute information set, the context information set, and the intention attribute information set, so as to predict a behavior of the target object by means of the behavior prediction model.

Description

Method and device for establishing predictive model, storage medium and electronic device

This disclosure claims the priority of the Chinese patent application with the application number 202210118953.2 and the invention title "method and device for establishing a prediction model, storage medium and electronic device" submitted to the China Patent Office on February 08, 2022, the entire contents of which are incorporated by reference incorporated in this disclosure.

technical field

The present disclosure relates to the communication field, and in particular, to a method and device for establishing a prediction model, a storage medium, and an electronic device.

Background technique

The existing data modeling scheme about user behavior, as shown in Figure 3, mainly analyzes the relationship between user characteristics and behavior, as well as the relationship between behaviors, and builds a user behavior analysis model; however, the existing user behavior data modeling scheme, information collection Insufficient, only collecting basic user information and user historical behavior information; and incomplete analysis, only analyzing the relationship between user characteristics and behavior, and the relationship between behaviors.

In the related technologies, insufficient information collection and incomplete correlation lead to inaccurate prediction of user behavior by the established prediction model, and failure to provide users with reasonable and effective prediction behavior. No effective solution has been proposed yet.

Contents of the invention

Embodiments of the present disclosure provide a method and device for establishing a predictive model, a storage medium, and an electronic device, so as to at least solve the problem of insufficient information collection and incomplete association relationships in related technologies, which lead to inaccurate prediction of user behavior by the established predictive model , can not provide users with reasonable and effective predictive behavior and other issues.

According to an embodiment of the present disclosure, a method for establishing a predictive model is provided, including: acquiring element information when a target object controls a home appliance to perform a target operation, wherein the element information includes at least: the target object Basic information, the location information of the home appliance, the device information of the home appliance, the target operation that the target object controls the home appliance to perform, the behavior information when the target object controls the home appliance to perform the target operation, the target object The environment information when the home appliance is controlled to perform the target operation, the time information when the target object controls the home appliance to perform the target operation; the element information is divided into a target set, wherein the target set includes at least: a user attribute information set, Time attribute information set, location attribute information set, context information set, intent attribute information set; according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intent The attribute information set establishes a behavior prediction model, so as to predict the behavior of the target object through the behavior prediction model.

According to another embodiment of the embodiments of the present disclosure, there is also provided an apparatus for establishing a predictive model, including: an acquisition module configured to acquire element information when a target object controls a home appliance to perform a target operation, wherein the element information is at least Including: the basic information of the target object, the location information of the home appliance, the device information of the home appliance, the target operation that the target object controls the home appliance to perform, and the target operation when the target object controls the home appliance to perform the target operation. Behavior information, environment information when the target object controls the home appliance to perform the target operation, time information when the target object controls the home appliance to perform the target operation; a classification module, configured to divide the element information into target sets, wherein, The target set at least includes: a user attribute information set, a time attribute information set, a location attribute information set, a context information set, and an intent attribute information set; , the location attribute information set, the context information set and the intention attribute information set establish a behavior prediction model, so as to predict the behavior of the target object through the behavior prediction model.

According to yet another aspect of the embodiments of the present disclosure, there is also provided a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute the above-mentioned predictive model at runtime. Build method.

According to yet another aspect of the embodiments of the present disclosure, there is also provided an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the above-mentioned processor executes the above-mentioned How to build a predictive model.

In an embodiment of the present disclosure, the element information when the target object controls the home appliance to perform the target operation is acquired, wherein the element information at least includes: the basic information of the target object, the location information of the home appliance, the home appliance The device information of the target object, the target operation performed by the target object to control the home appliance, the behavior information when the target object controls the home appliance to perform the target operation, the environment information when the target object controls the home appliance to perform the target operation, the target object Control the time information when the home appliance performs the target operation; divide the element information into a target set, wherein the target set includes at least: a user attribute information set, a time attribute information set, a location attribute information set, a context information set, an intent attribute information set; establish a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intention attribute information set, so as to pass the behavior prediction model Predict the behavior of the target object; adopt the above technical solution to solve the problems of insufficient information collection and incomplete correlation, which lead to inaccurate prediction of user behavior by the established prediction model and failure to provide users with reasonable and effective prediction behaviors, etc. , the embodiment of the present disclosure collects relevant information of target objects and behaviors in all aspects, analyzes the relationship between each tuple, builds a new prediction model, and realizes more effective and accurate behavior prediction.

Description of drawings

The drawings described here are used to provide a further understanding of the present disclosure, and constitute a part of the present disclosure. The schematic embodiments of the present disclosure and their descriptions are used to explain the present disclosure, and do not constitute improper limitations to the present disclosure. In the attached picture:

Fig. 1 is a block diagram of the hardware structure of a computer terminal of a method for establishing a prediction model according to an embodiment of the present disclosure;

FIG. 2 is a flowchart of a method for establishing a prediction model according to an embodiment of the present disclosure;

Fig. 3 is the flowchart of the establishment method of the predictive model of prior art;

4 is a schematic diagram of a method for establishing a prediction model according to an embodiment of the present disclosure;

FIG. 5 is a flowchart of a method for establishing a prediction model according to an optional embodiment of the present disclosure;

Fig. 6 is a structural block diagram of an apparatus for establishing a prediction model according to an embodiment of the present disclosure.

Detailed ways

In order to enable those skilled in the art to better understand the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only It is an embodiment of a part of the present disclosure, but not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present disclosure.

It should be noted that the terms "first" and "second" in the specification and claims of the present disclosure and the above drawings are used to distinguish similar objects, but not necessarily used to describe a specific sequence or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device comprising a sequence of steps or elements is not necessarily limited to the expressly listed instead, may include other steps or elements not explicitly listed or inherent to the process, method, product or apparatus.

The method embodiments provided by the embodiments of the present disclosure may be executed in mobile terminals, computer terminals or similar computing devices. Taking running on a computer terminal as an example, FIG. 1 is a hardware structural block diagram of a computer terminal according to a method for establishing a prediction model according to an embodiment of the present disclosure. As shown in Figure 1, the computer terminal may include one or more (only one is shown in Figure 1) processors 102 (processors 102 may include but not limited to processing devices such as microprocessor MCU or programmable logic device FPGA, etc.) and a memory 104 for storing data. In an exemplary embodiment, the above-mentioned computer terminal may further include a transmission device 106 and an input and output device 108 for communication functions. Those skilled in the art can understand that the structure shown in FIG. 1 is only for illustration, and it does not limit the structure of the above computer terminal. For example, the computer terminal may also include more or less components than those shown in FIG. 1 , or have a different configuration with functions equivalent to those shown in FIG. 1 or more functions than those shown in FIG. 1 .

The memory 104 can be set to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for establishing a predictive model in the embodiment of the present disclosure, and the processor 102 runs the computer program stored in the memory 104, Thereby executing various functional applications and data processing, that is, realizing the above-mentioned method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory that is remotely located relative to the processor 102, and these remote memories may be connected to a computer terminal through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

Transmission device 106 is configured to receive or transmit data via a network. The specific example of the above-mentioned network may include a wireless network provided by the communication provider of the computer terminal. In one example, the transmission device 106 includes a network interface controller (NIC for short), which can be connected to other network devices through a base station so as to communicate with the Internet. In an example, the transmission device 106 may be a radio frequency (Radio Frequency, RF for short) module, which is configured to communicate with the Internet in a wireless manner.

In this embodiment, a method for establishing a predictive model is provided, which is applied to the above-mentioned computer terminal. FIG. 2 is a flow chart of a method for establishing a predictive model according to an embodiment of the present disclosure. The process includes the following steps:

Step S202, acquiring element information when the target object controls the home appliance to perform the target operation, wherein the element information at least includes: basic information of the target object, location information of the home appliance, device information of the home appliance, The target object controls the target operation performed by the home appliance, the target object controls the behavior information when the home appliance performs the target operation, the target object controls the environment information when the home appliance performs the target operation, and the target object controls the home appliance. Time information when the target is operated;

Step S204, dividing the element information into a target set, wherein the target set at least includes: a user attribute information set, a time attribute information set, a location attribute information set, a context information set, and an intent attribute information set;

Step S206, establish a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intention attribute information set, so as to use the behavior prediction model to The behavior of the target object is predicted.

Through the above steps, the element information when the target object controls the home appliance to perform the target operation is acquired, wherein the element information at least includes: the basic information of the target object, the location information of the home appliance, and the device information of the home appliance , the target operation performed by the target object to control the home appliance, the behavior information when the target object controls the home appliance to perform the target operation, the environment information when the target object controls the home appliance to perform the target operation, the target object controls the home appliance time information when executing the target operation; divide the element information into target sets, wherein the target set at least includes: user attribute information set, time attribute information set, location attribute information set, context information set, intent attribute information set ; establish a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intention attribute information set, so as to use the behavior prediction model to predict the The behavior of the target object is predicted, which solves the problems of insufficient information collection and incomplete association in related technologies, which lead to inaccurate prediction of user behavior by the established prediction model and failure to provide users with reasonable and effective prediction behavior. This disclosure implements For example, by collecting relevant information about target objects and behaviors in all aspects, analyzing the relationship between each tuple, building a new prediction model, and realizing more effective and accurate behavior prediction.

In an exemplary embodiment, dividing the element information into target sets includes: obtaining at least basic information of the target object, location information of the home appliance, and control of the home appliance by the target object through the element information. The executed target operation, the environment information when the target object controls the home appliance device to perform the target operation, the device information of the home appliance device, and the time information when the target object controls the home appliance device to perform the target operation; the basic information of the target object Classify the information into the user attribute information set; classify the location information of the home appliance into the location attribute information set; control the environment information of the target object when the home appliance performs the target operation, and the device information of the home appliance . Classify the behavior information when the target object controls the home appliance to perform the target operation into the context information set; classify the time information when the target object controls the home appliance to perform the target operation into the time attribute information set; The target object controls the target operation performed by the home appliance to be classified into the intent attribute information set.

That is to say, based on each device, collect user information, family information, location information, environment information, device information, user behavior and other related information. Related information includes not only basic user information such as user age, birthday, and body fat, but also continuous or historical behavior information such as current behavior, previous behavior of the current behavior, and subsequent behavior of the current behavior; it also includes family information, environmental information, equipment Status information and other information. Analyze and organize the collected elemental information. Divide all element information into five tuple sets: user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set. Design a unified data model for each collection, including unified storage format, unified encoding, unified unit, etc.

In an exemplary embodiment, after the element information is divided into target sets, cleaning rules corresponding to the user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set are determined respectively ; According to the cleaning rules, the data classified into the user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set are respectively cleaned, so as to unify the classification into the user attribute information set, time attribute information set, and time attribute information set. The data format in attribute information set, location attribute information set, context information set, and intent attribute information set.

In an exemplary embodiment, acquiring the element information when the target object controls the home appliance device to perform the target operation includes: acquiring data information sent by multiple data sources when the target object controls the home appliance device to perform the target operation, wherein the data information Used to indicate the element information; perform deduplication and overwriting processing on the data information to obtain processed data information; analyze the processed data information, and obtain element information when the target object controls the home appliance to perform the target operation .

That is to say, through different device terminals, such as APP, AI, multi-screen, etc., and related systems, such as user center, IOT domain model, family model, etc., to collect information related to user behavior. The collected information includes but is not limited to user information, family information, location information, environmental information, device information, and user behavior and other related information. According to the information analysis, the collected information is deduplicated and overwritten to process the information.

In an exemplary embodiment, establishing a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set, and the intention attribute information set includes: establishing The association relationship among the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intent attribute information set; through the user attribute information set, the time attribute information set An association relationship among the attribute information set, the position attribute information set, the context information set and the intent attribute information set establishes a behavior prediction model.

The specific way to establish the association relationship between the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intent attribute information set is as follows: according to the target object Determine the first association relationship between the user attribute information set and the intention attribute information set; control the home appliance device to execute the target according to the target object Determine the second association relationship between the time attribute information set and the intent attribute information set based on the time information during operation and the behavior information when the target object controls the home appliance to perform the target operation; according to the location information of the home appliance Determine the third association relationship between the location attribute information set and the intent attribute information set with the behavior information when the target object controls the household electrical appliance to perform the target operation; the target object controls the target operation performed by the household electrical appliance, the The environment information when the target object controls the household electrical appliance to perform the target operation and the behavior information when the target object controls the household electrical appliance to perform the target operation determine the fourth association relationship between the context information set and the intent attribute information set.

That is to say, the relationship between each collection is established in the following way:

1) Horizontal analysis: the basic information of the target object, the location information of the home appliance, the device information of the home appliance, the target operation executed by the home appliance controlled by the target object, the execution of the home appliance controlled by the target object Information such as the behavior information when the target operation is performed, the environment information when the target object controls the home appliance device to perform the target operation, and the time information when the target object controls the home appliance device to perform the target operation are horizontally connected, and the relationship between each element information and user behavior is analyzed. relationship between. For example, the relationship between "family size" in the family attribute and the user behavior attribute "turn on the air conditioner";

2) Longitudinal analysis:

A) Analysis of different behavior attributes: according to the time series of behaviors, perform correlation analysis on different user behavior attributes; for example, the correlation between the behaviors of 'turning on the air conditioner' and 'turning on the water heater'.

B) Analysis of the same behavior attributes: For the same user behavior, perform correlation analysis of multiple historical behaviors. For example, Zhang San "turned on the air conditioner" at 7:05 pm on January 1st; "turned on the air conditioner" at 7:15 pm on January 2nd, and analyzed the relationship between historical behaviors. It should be noted that the above numbers are only for better understanding of the embodiments of the present disclosure, and are not limited by the embodiments of the present disclosure.

In an exemplary embodiment, after the behavior prediction model is established according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intent attribute information set, the obtained According to the current behavior of the target object, and input the current behavior into the prediction model; according to the user attribute information set, time attribute information set, location attribute information set, context information set and intent attribute information in the prediction model Predict the to-be-executed behavior of the target object, and control the target device to perform corresponding operations according to the to-be-executed behavior.

That is to say, predict the next behavior of the target object according to the user attribute information set, time attribute information set, location attribute information set, context information set and intent attribute information in the prediction model, and according to the information push rule, the arrival of the prediction model The predicted behavior message is pushed to the target object.

In order to better understand the process of the above-mentioned method for establishing the prediction model, the flow of the implementation method for the above-mentioned establishment of the prediction model will be described below in conjunction with optional embodiments, but it is not intended to limit the technical solutions of the embodiments of the present disclosure.

In this embodiment, a method for establishing a prediction model is provided. FIG. 4 is a schematic diagram of a method for establishing a prediction model according to an embodiment of the disclosure, as shown in FIG. 4 , specifically as follows:

Obtain user data/device data/behavior log data/etc. from the GIO layer or AI device side, and input them into the operational data warehouse ODS, so that the ODS can process the data and send the processed data to the data warehouse DW , DW classifies the processed data and divides them into different tuples such as person, time, place, context, intention, etc.; sends the classified data to the data mart DM established with an application as the starting point, and DM passes Relevant algorithms analyze the classified data to obtain the corresponding knowledge graph, prediction model, etc.

In this embodiment, a method for establishing a prediction model is provided. FIG. 5 is a flowchart of a method for establishing a prediction model according to an optional embodiment of the present disclosure, as shown in FIG. 5 , and the details are as follows:

Step S501: start;

Step S502: collect user behavior-related information through different device terminals, such as APP, AI, multi-screen, etc., and related systems, such as user center, IOT domain model, family model, etc. The collected information includes but not limited to user information and family information , address information, weather and other third-party information, device information, and user behavior and other related information;

Step S503: information analysis;

Specifically, 1) data horizontal analysis:

Horizontally pull related information such as user information, family information, location information, environment information, device information, and user behavior to analyze the relationship between each element of information and user behavior. For example, the relationship between "family size" in the family attribute and the user behavior attribute "turn on the air conditioner";

2) Data longitudinal analysis

B) Analysis of the same behavior attributes: For the same user behavior, perform correlation analysis of multiple historical behaviors. For example, Zhang San "turned on the air conditioner" at 7:05 pm on January 1st; "turned on the air conditioner" at 7:15 pm on January 2nd, and analyzed the relationship between historical behaviors.

Step S504: information standardization;

Specifically: 3.1 Information classification

According to information analysis, various data are grouped and classified, and deduplication and coverage processing are performed; user attribute information set U, time attribute information set T, location attribute information set A, and context attribute information set L (equivalent to the above implementation example user attribute information collection, time attribute information collection, location attribute information collection, context information collection).

3.2 Standardization of tuple information

Design a unified data model for the classified data, including standardized processing such as unified storage format, unified coding, and unified unit. For example, the format of 'date' is unified as '2021-10-20'; the code of 'city' is unified as 'city', and the unit of 'water consumption' is unified as 'L', etc.

3.3 Standardization of behavioral information

For the same behavior events from different data sources, conduct behavior abstraction and classification processing to form coded and unified behavior information. For example, the behavior event collected on the AI side is 'turn on the air conditioner in the living room', the extracted location tuple information is 'living room', and the behavior event attribute of the context tuple is 'turn on the air conditioner'; for example, the collection function page on the APP side is 'air conditioner details page', execute If the operation is 'on', then the attribute of the behavior event merged into the context tuple is 'turn on the air conditioner'.

Step S505: Establishing a quintuple model according to the association relationship between the information in the quintuple;

The five-tuple model includes: 1) user: user ID information, and user characteristic information.

2) Time: Behavior time series; including user behavior time stamp, year, month, day, hour and other information of behavior time.

3) Location: behavior location address information; including the space to which the user behavior belongs, such as 'living room'; also includes information such as the province, city, district, county, and community to which the behavior belongs.

4) Context: Information such as pre- and post-sequential behaviors, pre- and post-sequence behavior states, user or network device portraits, weather and air quality, etc.

5) Intent: Data predicts subsequent behavioral information.

Step S506: User behavior prediction: According to the user tuple, time tuple, location tuple, and context tuple in the five-tuple, predict the user's next behavior, and store the predicted user behavior information in the intention tuple;

Step S507: Predictive user behavior push: according to the information push rules, push the predicted behavior message of the predictive model to the user;

Step S508: end.

It should be noted that: U: user attribute information collection; u(x) represents a certain user attribute, and u(u_1,u_2,...,u_k) represents the 1st to kth attributes of the user. For example, a user includes attributes such as age, gender, and occupation information.

T: time attribute information set; t(x) represents the time attribute of a certain user behavior, and t(t_1,t_2,...,t_k) represents the 1st to k attributes of time. For example, attributes such as the year, month, day, and hour to which the user behavior belongs.

A: location attribute information collection; b(x) represents the location attribute of a certain user behavior, and b(b_1,b_2,...,b_k) represents the 1st to k attributes of the location. For example, attributes such as provinces, cities, districts and counties, and rooms to which user behaviors belong.

L: context attribute information set; l(x) represents the context attribute of a certain user behavior, and l(l_1,l_2,...,l_k) represents the 1st to k features of the context. For example, attributes such as the user's previous behavior, current behavior, current weather, and current device power-on status.

I: Intent attribute information set; l(x) represents a certain user intent attribute, and l(l_1,l_2,...,l_k) represents the 1st to k attributes of user intent. Properties such as opening curtains, turning on lights, increasing wind speed, etc.

F: five-tuple attribute information set; f_x(u_1,t_1,a_1,l_1,i_1,…) indicates that user l_1 and the corresponding user behavior time attribute 1 is t_1, address location attribute 1 is a_1, and context attribute 1 is l_1 ; Get the fifth tuple i_1 according to the first four tuples. For example, f_1(u_1, t_1, a_1, l_1, i_1) indicates that the user: 'Zhang San', the corresponding user behavior attribute is '2021-10-20', the location attribute is 'living room', and the context attribute is 'too cold' , to predict user behavior intention as 'turn on the air conditioner'.

In the embodiment of the present disclosure, each element information is collected from multiple client terminals, including but not limited to user information, family information, location information, environment information, device information, user behavior and other related information; through analyzing the characteristics of each element information, and Analysis of the relationship between each element and behavior; building a five-tuple data model of user behavior; thus enabling a more comprehensive analysis of the relationship between users and behavior, and obtaining more accurate user behavior prediction information.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation. Based on such an understanding, the technical solution of the present disclosure can be embodied in the form of a software product in essence or the part that contributes to the prior art, and the computer software product is stored in a storage medium (such as ROM/RAM, disk, CD) contains several instructions to make a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) execute the method of each embodiment of the present disclosure.

In this embodiment, a device for establishing a forecast model is also provided, and the device for establishing a forecast model is configured to implement the above-mentioned embodiments and preferred implementation modes, and those that have already been described will not be described in detail. As used below, the term "module" may be a combination of software and/or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementations in hardware, or a combination of software and hardware are also possible and contemplated.

Fig. 6 is a structural block diagram of a device for establishing a predictive model according to an embodiment of the present disclosure; as shown in Fig. 6 , it includes:

The obtaining module 62 is configured to obtain element information when the target object controls the home appliance to perform the target operation, wherein the element information at least includes: the basic information of the target object, the location information of the home appliance, the location information of the home appliance Device information, the target operation performed by the target object to control the home appliance, the behavior information when the target object controls the home appliance to perform the target operation, the environment information when the target object controls the home appliance to perform the target operation, the target object controls Time information when the home appliance performs the target operation;

The classification module 64 is configured to divide the element information into a target set, wherein the target set at least includes: a user attribute information set, a time attribute information set, a location attribute information set, a context information set, and an intent attribute information set;

The establishment module 66 is configured to establish a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intention attribute information set, so as to pass the behavior The prediction model predicts the behavior of the target object.

Through the above apparatus, the element information when the target object controls the home appliance to perform the target operation is acquired, wherein the element information at least includes: basic information of the target object, location information of the home appliance, and device information of the home appliance , the target operation performed by the target object to control the home appliance, the behavior information when the target object controls the home appliance to perform the target operation, the environment information when the target object controls the home appliance to perform the target operation, the target object controls the home appliance time information when executing the target operation; divide the element information into target sets, wherein the target set at least includes: user attribute information set, time attribute information set, location attribute information set, context information set, intent attribute information set ; establish a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intention attribute information set, so as to use the behavior prediction model to predict the The behavior of the target object is predicted, which solves the problems of insufficient information collection and incomplete association in related technologies, which leads to inaccurate prediction of user behavior by the established prediction model, and cannot provide reasonable and effective prediction behavior for users. The embodiment collects relevant information of target objects and behaviors in all aspects, analyzes the relationship between each tuple, builds a new prediction model, and realizes more effective and accurate behavior prediction.

In an exemplary embodiment, the classification module is further configured to obtain at least the basic information of the target object, the location information of the home appliance, the target operation performed by the target object to control the home appliance, and the The environment information when the target object controls the home appliance to perform the target operation, the device information of the home appliance, and the time information when the target object controls the home appliance to perform the target operation; classify the basic information of the target object to the user attribute information set; classify the location information of the home appliance into the location attribute information set; control the environment information when the target object controls the home appliance to perform the target operation, the device information of the home appliance, the target object control The behavior information when the home appliance performs the target operation is classified into the context information set; the time information when the target object controls the home appliance to perform the target operation is classified into the time attribute information set; the target object controls the home appliance to perform the target operation. The target operations of are classified into the intent attribute information set.

In an exemplary embodiment, the obtaining module is further configured to determine cleaning rules respectively corresponding to the user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set; according to the cleaning The rules respectively clean the data classified into user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set, so as to unify the classification into user attribute information set, time attribute information set, location The data format in attribute information collection, context information collection, and intent attribute information collection.

In an exemplary embodiment, the obtaining module is further configured to obtain data information sent by multiple data sources when the target object controls the household electrical appliance to perform the target operation, wherein the data information is used to indicate the element information; The above data information is deduplicated and overwritten to obtain the processed data information; the processed data information is analyzed, and the element information when the target object controls the home appliance to perform the target operation is obtained.

In an exemplary embodiment, the establishment module is further configured to establish the relationship between the user attribute information set, the time attribute information set, the location attribute information set, the context information set, and the intent attribute information set The association relationship among the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intention attribute information set is used to establish a behavior prediction model.

In an exemplary embodiment, the establishment module is further configured to determine the user attribute information set and the intention attribute information according to the basic information of the target object and the behavior information of the target object when controlling the household electrical appliance to perform the target operation The first association relationship between the sets; according to the time information when the target object controls the home appliance device to perform the target operation and the behavior information when the target object controls the home appliance device to perform the target operation, determine the time attribute information set and the intention The second association relationship between attribute information sets; determine the relationship between the location attribute information set and the intention attribute information set according to the location information of the home appliance and the behavior information when the target object controls the home appliance to perform a target operation The third association relationship; the target operation performed by the target object to control the household electrical device, the environment information when the target object controls the household electrical device to perform the target operation, and the behavior information when the target object controls the household electrical device to perform the target operation determine the A fourth association relationship between the context information set and the intent attribute information set.

In an exemplary embodiment, the above apparatus further includes a prediction module configured to obtain the current behavior of the target object, and input the current behavior into the prediction model; according to the user attributes in the prediction model The information set, time attribute information set, location attribute information set, context information set and intent attribute information predict the to-be-executed behavior of the target object, and control the target device to perform corresponding operations according to the to-be-executed behavior.

An embodiment of the present disclosure also provides a storage medium, the storage medium includes a stored program, wherein the above-mentioned program executes any one of the above-mentioned methods when running.

Optionally, in this embodiment, the above-mentioned storage medium may be configured to store program codes for performing the following steps:

S1. Obtain element information when the target object controls the home appliance to perform the target operation, wherein the element information at least includes: basic information of the target object, location information of the home appliance, device information of the home appliance, and The target object controls the target operation performed by the home appliance, the behavior information when the target object controls the home appliance to perform the target operation, the environment information when the target object controls the home appliance to perform the target operation, the target object controls the home appliance to execute the target Time information at the time of operation;

S2. Divide the element information into a target set, wherein the target set includes at least: a user attribute information set, a time attribute information set, a location attribute information set, a context information set, and an intent attribute information set;

S3. Establish a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set, and the intent attribute information set, so as to use the behavior prediction model for all predict the behavior of the target audience.

Embodiments of the present disclosure also provide an electronic device, including a memory and a processor, where a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

Optionally, the above-mentioned electronic device may further include a transmission device and an input-output device, wherein the transmission device is connected to the above-mentioned processor, and the input-output device is connected to the above-mentioned processor.

Optionally, in this embodiment, the above-mentioned processor may be configured to execute the following steps through a computer program:

Optionally, in this embodiment, the above-mentioned storage medium may include but not limited to: U disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), Various media that can store program codes such as removable hard disks, magnetic disks, or optical disks.

Optionally, for specific examples in this embodiment, reference may be made to the examples described in the foregoing embodiments and optional implementation manners, and details are not repeated in this embodiment.

Obviously, those skilled in the art should understand that each module or each step of the above-mentioned disclosure can be realized by a general-purpose computing device, and they can be concentrated on a single computing device, or distributed in a network composed of multiple computing devices Alternatively, they may be implemented in program code executable by a computing device so that they may be stored in a storage device to be executed by a computing device, and in some cases in an order different from that shown here The steps shown or described are carried out, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps among them are fabricated into a single integrated circuit module for implementation. As such, the present disclosure is not limited to any specific combination of hardware and software.

The above descriptions are only preferred embodiments of the present disclosure, and are not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

A method for establishing a predictive model, comprising:

Obtain element information when the target object controls the home appliance to perform the target operation, wherein the element information at least includes: basic information of the target object, location information of the home appliance, device information of the home appliance, the target The object controls the target operation performed by the home appliance, the behavior information when the target object controls the home appliance to perform the target operation, the environment information when the target object controls the home appliance to perform the target operation, and the target object controls the home appliance to perform the target operation. time information;

Dividing the element information into a target set, wherein the target set includes at least: a user attribute information set, a time attribute information set, a location attribute information set, a context information set, and an intent attribute information set;

Establish a behavior prediction model according to the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intention attribute information set, so as to predict the target through the behavior prediction model predict the behavior of objects.
The method for establishing a predictive model according to claim 1, wherein dividing the element information into target sets includes:

Obtain at least the basic information of the target object, the location information of the home appliance, the target operation performed by the target object to control the home appliance, the environment information when the target object controls the home appliance to perform the target operation, through the element information, The device information of the home appliance, the time information when the target object controls the home appliance to perform the target operation;

Classify the basic information of the target object into the user attribute information set;

classifying the location information of the home appliance into the location attribute information set;

Classify the environment information when the target object controls the home appliance to perform the target operation, the device information of the home appliance, and the behavior information when the target object controls the home appliance to perform the target operation into the context information set;

classify the time information when the target object controls the home appliance to perform the target operation into the time attribute information set;

The target operation performed by the target object to control the home appliance is classified into the intent attribute information set.
The method for establishing a predictive model according to claim 1, wherein, after dividing the element information into target sets, the method further comprises:

Determine the cleaning rules respectively corresponding to the user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set;

According to the cleaning rules, the data classified into the user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set are respectively cleaned, so as to unify the classification into the user attribute information set, time attribute Data format in information collection, location attribute information collection, context information collection, and intent attribute information collection.
The method for establishing a predictive model according to claim 1, wherein obtaining the element information when the target object controls the home appliance to perform the target operation includes:

Acquiring data information sent by multiple data sources when the target object controls the home appliance to perform the target operation, wherein the data information is used to indicate the element information;

Performing deduplication and overwriting processing on the data information to obtain processed data information;

The processed data information is analyzed, and the element information when the target object controls the household electrical appliance executes the target operation is acquired.
The method for establishing a predictive model according to claim 1, wherein the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intent attribute information set are established Behavioral predictive models, including:

establishing an association relationship between the user attribute information set, the time attribute information set, the location attribute information set, the context information set, and the intent attribute information set;

A behavior prediction model is established through the association relationship among the user attribute information set, the time attribute information set, the position attribute information set, the context information set and the intention attribute information set.
The method for establishing a prediction model according to claim 5, wherein establishing one of the set of user attribute information, the set of time attribute information, the set of location attribute information, the set of context information and the set of intent attribute information relationships, including:

determining a first association relationship between the user attribute information set and the intent attribute information set according to the basic information of the target object and the behavior information when the target object controls the household electrical appliance to perform a target operation;

Determine the second association between the time attribute information set and the intention attribute information set according to the time information when the target object controls the household electrical appliance to perform the target operation and the behavior information when the target object controls the household electrical appliance to perform the target operation relation;

determining a third association relationship between the location attribute information set and the intention attribute information set according to the location information of the home appliance and the behavior information when the target object controls the home appliance to perform a target operation;

Determine the context information set and the A fourth association relationship between intent attribute information sets.
The method for establishing a predictive model according to claim 1, wherein the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intent attribute information set are established After the behavior prediction model, the method also includes:

Obtaining the current behavior of the target object, and inputting the current behavior into the prediction model;

Predict the to-be-executed behavior of the target object according to the user attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information in the prediction model, and control the target device to execute the corresponding action according to the to-be-executed behavior operate.
A device for establishing a predictive model, comprising:

An acquisition module, configured to acquire element information when the target object controls the household electrical appliance to perform the target operation, wherein the element information at least includes: basic information of the target object, location information of the household electrical appliance, device information of the household electrical appliance information, the target operation performed by the target object to control the home appliance, the behavior information when the target object controls the home appliance to perform the target operation, the environment information when the target object controls the home appliance to perform the target operation, the target object controls the home appliance Time information when the device performed the targeted operation;

A classification module, configured to divide the element information into a target set, wherein the target set at least includes: a user attribute information set, a time attribute information set, a location attribute information set, a context information set, and an intent attribute information set;

A building module, configured to establish a behavior prediction model according to the set of user attribute information, the set of time attribute information, the set of location attribute information, the set of context information and the set of intention attribute information, so as to pass the behavior prediction The model makes predictions about the behavior of the target object.
The device for establishing a predictive model according to claim 8, wherein,

The classification module is further configured to obtain at least basic information of the target object, location information of the home appliance, target operations performed by the target object to control the home appliance, and control of the home appliance by the target object through the element information. The environment information when the target operation is performed, the device information of the home appliance, and the time information when the target object controls the home appliance to perform the target operation; classify the basic information of the target object into the user attribute information set; The location information of the home appliance is classified into the location attribute information set; the environment information when the target object controls the home appliance to perform the target operation, the device information of the home appliance, the time when the target object controls the home appliance to perform the target operation Classify the behavior information of the target object into the context information set; classify the time information when the target object controls the home appliance to perform the target operation into the time attribute information set; classify the target operation performed by the target object to control the home appliance into the set A collection of intent attribute information.
The device for establishing a predictive model according to claim 8, wherein,

The acquisition module is further configured to determine cleaning rules corresponding to the user attribute information set, time attribute information set, location attribute information set, context information set, and intention attribute information set; The attribute information set, time attribute information set, location attribute information set, context information set, and intent attribute information set are cleaned to unify the classification into user attribute information set, time attribute information set, location attribute information set, and context information Collection, the data format in the intent attribute information collection.
The device for establishing a predictive model according to claim 8, wherein,

The obtaining module is further configured to obtain data information sent by multiple data sources when the target object controls the household electrical appliance to perform the target operation, wherein the data information is used to indicate the element information; and the data information is deduplicated and overlay processing to obtain processed data information; analyze the processed data information to obtain element information when the target object controls the household electrical appliance to perform the target operation.
The device for establishing a predictive model according to claim 8, wherein,

The establishing module is further configured to establish an association relationship between the user attribute information set, the time attribute information set, the location attribute information set, the context information set, and the intent attribute information set; through the Establish a behavior prediction model based on the association relationship between the user attribute information set, the time attribute information set, the location attribute information set, the context information set and the intent attribute information set.
The device for establishing a predictive model according to claim 12, wherein,

The establishment module is further configured to determine the first between the user attribute information set and the intention attribute information set according to the basic information of the target object and the behavior information of the target object when controlling the household electrical appliance to perform the target operation. Association relationship: determine the relationship between the time attribute information set and the intention attribute information set according to the time information when the target object controls the home appliance device to perform the target operation and the behavior information when the target object controls the home appliance device to perform the target operation The second association relationship: determine the third association relationship between the location attribute information set and the intention attribute information set according to the location information of the home appliance and the behavior information when the target object controls the home appliance to perform a target operation; Determine the context information set and the A fourth association relationship between intent attribute information sets.
The device for establishing a predictive model according to claim 8, wherein the device for establishing a predictive model further comprises:

A prediction module, configured to obtain the current behavior of the target object, and input the current behavior into the prediction model; according to the user attribute information set, time attribute information set, location attribute information set, The context information set and the intent attribute information predict the to-be-executed behavior of the target object, and control the target device to perform corresponding operations according to the to-be-executed behavior.
A computer-readable storage medium, the computer-readable storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 7 when running.
An electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the method described in any one of claims 1 to 7 through the computer program.