CN118051625A - Method and device for optimizing scene generation model, storage medium and electronic device - Google Patents

Abstract

The application discloses a method and a device for optimizing a scene generation model, a storage medium and an electronic device, and relates to the technical field of smart families, wherein the method for optimizing the scene generation model comprises the following steps: after generating a scene by using the first scene generation model, acquiring satisfaction degree of a user on the generated scene; according to the satisfaction degree of the user, performing positive evaluation or negative evaluation on the generated scene; inputting the evaluation result into a knowledge base in the intelligent family field, and updating a first scene generation model; or inputting the evaluation result into the large language model to obtain correction data, and updating the first scene generation model by using the correction data. According to the method and the device for generating the scene, the satisfaction degree of the user on the generated scene can be obtained, and the model in the system can be optimized according to the satisfaction degree of the user, so that the scene generated by the model can be more in line with the requirements of the user in the subsequent use process of the user, and the user experience is improved.

Description

Method and device for optimizing scene generation model, storage medium and electronic device

Technical Field

The application relates to the technical field of smart families, in particular to a method and a device for optimizing a scene generation model, a storage medium and an electronic device.

Background

Currently, many existing smart home systems are capable of generating corresponding scenes according to the needs of users. Because there may be some bias in understanding the user's needs, after generating the scene, user feedback needs to be collected and analyzed, understanding the actual emotional changes of the user to the automatically generated scene, and updating and optimizing the scene generation rules accordingly to better meet the user's needs and provide a personalized user experience. Therefore, in order to make the generated scene more in line with the needs of the user, it is necessary to implement modification and optimization of the scene generation model in the smart home.

In order to optimize a scene generation model in an intelligent home, related art discloses an automated control system, which can automatically perform various tasks, such as playing music, setting an alarm clock, etc., according to a voice command of a user through a voice assistant. And, the system is able to gradually learn and adapt to the user's preferences by observing the user's habits.

In the process of implementing the embodiments of the present disclosure, it is found that at least the following problems exist in the related art:

Although the related art is capable of generating a scene from a user's voice feedback, the related art mainly relies on the user's direct feedback to learn and adapt to the user's preferences. However, users may not always be willing or able to provide direct feedback. In this case, if the system relies solely on direct feedback from the user, the user's needs and preferences may not be accurately understood. In addition, in the learning process, the system needs the user to feed back frequently, so that the user is interfered, and the user experience is reduced.

Further, the related art only updates the current scene according to the feedback after obtaining the direct feedback of the user, but cannot optimize the scene generation model.

It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the application and thus may include information that does not form the prior art that is already known to those of ordinary skill in the art.

Disclosure of Invention

The following presents a simplified summary in order to provide a basic understanding of some aspects of the disclosed embodiments. This summary is not an extensive overview, and is intended to neither identify key/critical elements nor delineate the scope of such embodiments, but is intended as a prelude to the more detailed description that follows.

The embodiment of the disclosure provides a method and a device for optimizing a scene generation model, a storage medium and an electronic device, and the scene generation model can be optimized according to feedback of a user.

In some embodiments, the method comprises: after generating a scene by using the first scene generation model, acquiring satisfaction degree of a user on the generated scene; according to the satisfaction degree of the user, performing positive evaluation or negative evaluation on the generated scene; inputting the evaluation result into a knowledge base in the intelligent family field, and updating a first scene generation model; or inputting the evaluation result into the large language model to obtain correction data, and updating the first scene generation model by using the correction data.

Optionally, obtaining satisfaction of the user with the generated scene includes: monitoring the behavior of a user in a generated scene, and obtaining the satisfaction degree of the user; and/or, collecting feedback of the user to obtain satisfaction of the user; and/or, obtaining satisfaction of the user through an A/B test; and/or obtaining satisfaction of the user according to the prediction model; and/or obtaining the satisfaction degree of the user according to the user loss rate and the revisit rate.

Optionally, inputting the evaluation result into a knowledge base in the smart home domain, and updating the first scene generation model includes: according to the evaluation result, the knowledge of the knowledge base in the intelligent family field is expanded, and an updated knowledge base is obtained; and training the first scene generating model according to the updated knowledge base.

Optionally, performing positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user, including: according to the satisfaction degree of the user, evaluating the scene corresponding to the satisfaction degree by using an evaluation model to obtain the evaluation score of the scene; determining to perform forward evaluation on the generated scene under the condition that the evaluation score is higher than or equal to a score threshold value; or in the case that the evaluation score is lower than the score threshold, determining to perform negative evaluation on the generated scene.

Optionally, before evaluating the scene corresponding to the satisfaction degree by using the evaluation model according to the satisfaction degree of the user, the method further includes: selecting a machine learning model or a statistical model; training the selected model according to the historical user feedback data to obtain an evaluation model.

Optionally, updating the first scene generation model with the correction data includes: applying the correction data to a scene vector library to update scene rules and obtain an updated scene vector library; and training through the updated scene vector library to obtain a second scene generation model.

Optionally, training through the updated scene vector library to obtain a second scene generation model, including: creating a training set according to the updated scene vector library; wherein the training set comprises scenes conforming to the updated scene vector library; a second scene generation model is obtained using training set training.

Optionally, training using the training set to obtain a second scene generation model includes: training the basic model by using a training set to obtain a second scene generation model; or adjusting the first scene generation model by using the training set to obtain a second scene generation model.

Optionally, generating the scene using the first scene generation model includes: acquiring the requirement of a user; inputting the requirements of a user into a first scene generation model to obtain a first scene scheme; the training data type of the first scene generation model comprises user data, equipment data, environment data and space data; the target device is controlled according to the first scenario scheme.

In some embodiments, the apparatus for scene generation comprises:

a model application module configured to generate a scene using the first scene generation model;

The user feedback module is configured to perform positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user; inputting the evaluation result into a knowledge base in the intelligent family field, and updating a first scene generation model; or inputting the evaluation result into the large language model to obtain correction data, and updating the first scene generation model by using the correction data.

In some embodiments, the computer-readable storage medium includes a stored program, wherein the program when run performs the method for optimizing a scene generation model described above.

In some embodiments, the electronic device comprises a memory in which a computer program is stored and a processor arranged to perform the above-described method for optimizing a scene generation model by means of the computer program.

The method and device for optimizing the scene generation model, the storage medium and the electronic device provided by the embodiment of the disclosure can realize the following technical effects:

Compared with the related art, the method and the device can acquire the satisfaction degree of the user on the generated scene, and optimize the model in the system according to the satisfaction degree of the user. Furthermore, the scene generated by the model can be more in line with the requirements of the user, so that the user experience is improved.

The foregoing general description and the following description are exemplary and explanatory only and are not restrictive of the application.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.

In order to more clearly illustrate the embodiments of the application or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.

FIG. 1 is a schematic diagram of a hardware environment for a method of scene generation according to an embodiment of the present disclosure;

FIG. 2 is a software system schematic diagram of a method for scene generation according to an embodiment of the present disclosure;

FIG. 3-1 is a software system schematic diagram of a method for optimizing a scene generation model according to an embodiment of the present disclosure;

FIG. 3-2 is a software system schematic diagram of another method for optimizing a scene generation model according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram of a method for optimizing a scene generation model according to an embodiment of the disclosure;

FIG. 5 is a schematic diagram of another method for optimizing a scene generation model according to an embodiment of the disclosure;

FIG. 6 is a schematic diagram of another method for optimizing a scene generation model according to an embodiment of the disclosure;

FIG. 7 is a schematic diagram of an apparatus for optimizing a scene generation model according to an embodiment of the disclosure;

FIG. 8 is a schematic diagram of an electronic device for optimizing a scene generation model according to an embodiment of the disclosure;

fig. 9 is a schematic diagram of a method for constructing a hierarchical knowledge-graph, in accordance with an embodiment of the present disclosure.

Detailed Description

In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the present application (in the described embodiments) are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Currently, many existing smart home systems are capable of generating corresponding scenes according to the needs of users.

According to one aspect of an embodiment of the present disclosure, a method for scene generation is provided. The method for generating the scene is widely applied to full-house intelligent digital control application scenes such as intelligent Home (Smart Home), intelligent Home equipment ecology, intelligent Home (INTELLIGENCE HOUSE) ecology and the like. Alternatively, in the embodiment of the present disclosure, the method for generating a scene described above may be applied to a hardware environment composed of the smart home device 01 and the control center device 02 as shown in fig. 1. As shown in fig. 1, the control center device 02 is connected with the intelligent home appliance device 01 through a network, and can be used for providing services (such as application services and the like) for a terminal or a client installed on the terminal, a database can be set on a server or independent of the server, and used for providing data storage services for the control center device 02, and cloud computing and/or edge computing services can be configured on the server or independent of the server, and used for providing data computing services for the control center device 02.

The network may include, but is not limited to, at least one of: wired network, wireless network. The wired network may include, but is not limited to, at least one of: a wide area network, a metropolitan area network, a local area network, and the wireless network may include, but is not limited to, at least one of: WIFI (WIRELESS FIDELITY ), bluetooth. The intelligent household electrical appliance 01 can be not limited to a PC, a mobile phone, a tablet personal computer, an intelligent air conditioner, an intelligent smoke machine, an intelligent refrigerator, an intelligent oven, an intelligent cooking range, an intelligent washing machine, an intelligent water heater, an intelligent washing device, an intelligent dish washer, an intelligent projection device, an intelligent television, an intelligent clothes hanger, an intelligent curtain, an intelligent video, an intelligent socket, an intelligent sound box, an intelligent fresh air device, an intelligent kitchen and toilet device, an intelligent bathroom device, an intelligent sweeping robot, an intelligent window cleaning robot, an intelligent mopping robot, an intelligent air purifying device, an intelligent steam box, an intelligent microwave oven, an intelligent kitchen appliance, an intelligent purifier, an intelligent water dispenser, an intelligent door lock and the like.

The control center device 02 may not be limited to a cloud server, a central controller, an intelligent home gateway, or the like.

In connection with the hardware environment shown in fig. 1, a software system for scene generation is provided in an embodiment of the present disclosure. The control center device 02 is provided with a user interaction module 03, a model application module 04, a control module 05 and a model training module 06.

The user interaction module 03 is configured to obtain a requirement of a user, and input the requirement of the user into the first scene generation model.

Optionally, the user interaction module 03 includes a receiving module, an adding module, a processing module, and a classifying module. The receiving module is capable of receiving request data of a user. The adding module can add context information to the request data. Wherein the context information includes user data, device data, environment data, and spatial data. The processing module can process the request data added with the background information to obtain the fused request data. The classification module comprises a classifier, and can transmit the fused request data to the classifier. Wherein the classifier is used for classifying the request of the user.

The user interaction module 03 can acquire background information of the current smart home after the user sends out the request data. The background information contains a lot of information related to the scene. By adding such information to the request data, the content of the request data can be made richer, and more effective information can be contained for the classifier to understand and classify. The vectorization processing module can perform domain vectorization processing on the request data added with the background information, and finally input the request data into the classifier for classification. Therefore, the processed data is easier to be understood by the classifier, and the classification result of the classifier is more accurate. Finally, the accuracy of the classification model is improved.

A model application module 04 comprising a first scenario generation model configured to output a first scenario solution upon receiving a user's demand; wherein the training data type of the first scene generation model includes user data, device data, environment data, and spatial data.

Training data of the first scene generation model is stored in the background information base 09. The background information base 09 includes user data, device data, environment data, and space data.

The user data includes basic attributes, behavior habits, preferences and preferences, common equipment and the environment in which the user is located.

User data is extremely important in the smart home environment, and can help the system understand and predict the behavior of the user, so that the service is more personalized and efficient. Basic attributes include the sex, age, occupation, etc. of the user. These tags can help the system understand the general behavioral tendencies of the user. For example, a young white collar may be more familiar with smart phones to control home devices. The behavior habit comprises records of the past use of the intelligent home equipment by the user, and the behavior habit can describe a comprehensive behavior mode. For example, a user often uses a coffee machine in 7 a.m., the use time and device identification are taken as one of the user's behavioral habits data, which the system can learn and warm up the coffee machine in time before the use time. The preferences and preferences include a record of user control devices and configuration settings that the system can also learn. For example, a user may like to watch a movie in the evening, and the user may often dim the living room lights and turn on home theater mode. Common devices include records of the devices most commonly used by users. For example, if a user often cooks in a kitchen and prefers to hear an audio reading, the sound is determined to be the user's usual device, or one of the usual devices. The environment in which the user is located includes activity patterns in different spatial environments. For example, a user who is often working in a study will exhibit a different activity pattern than a user who is resting in a living room.

The device data includes: device basic information, device status, device capabilities, and device location.

Devices are a key component of a smart home environment. The basic information of the device includes the model, brand, model and the like of the device, and the information is helpful for identifying the type of the device and knowing the basic function of the device. The device state includes a switch state and a current operating state of the device. Such state information is critical to understanding the current situation of the device and predicting the operations that may need to be performed. For example, knowing that the air conditioner is running and is set at 22 ℃, the system may need to raise the temperature of the air conditioner setting if the user feels too cold. Device capabilities include functions that the device can perform. For example, intelligent lights may adjust brightness and color. And the intelligent sound box can play music, set timing and the like. The system needs to know the characteristics of the device and how the device responds to a particular command. Each device in a home environment will be in a particular spatial location. This location information will influence the understanding of the user instructions and subsequent operating strategies. For example, if it is known that a voice command is issued from a bedroom and the command content is "turn off the light" then the light located in the bedroom should be turned off.

The environmental data includes: somatosensory environment, indoor air quality, time and season, room traffic and special conditions.

The environment has a decisive influence on the system's understanding of the user's needs and making accurate decisions. The somatosensory environment comprises indoor temperature, humidity, illumination, noise and other environmental parameter information. For example, the indoor temperature may affect a thermostatic device. While indoor lighting conditions change may change the user's demand for intelligent lights. Indoor air quality may require a user to turn on an air purifier or a roller blind. The time and season include the current point in time, date, season, etc. For example, entering the night requires the light to be turned on, while in winter it may be necessary to raise the temperature of the thermostatic device. The room traffic includes whether someone is in the room and the movement of the person. For example, a room with no person may turn off the lights to save energy, while if someone enters it may be necessary to turn on the lighting. Special cases include special events that also affect the environment, such as whether a party is in progress or a guest is coming.

Spatial data includes room layout, room functions, device distribution, and spatial dimensions.

Spatial data refers to physical attributes and features of the environment in which the smart home system is located. The room layout includes the positional relationship of the individual rooms in the home and the house type design. Such information may affect how the devices are distributed and installed, and may also affect the user's movement and use of the devices at home. Each room has specific functions, e.g. bedrooms for rest, study rooms for work or study, living rooms for social or recreational activities. Knowing the function of each room can help the system understand what activities a user may be doing at a particular location or what needs the device. The device distribution can learn the intelligent home devices contained in each room. And the relative position between the devices is also important for accurate control of the devices. Spatial dimensions include the size, angle, and height of a space or venue. For example, when controlling a window shade, it is necessary to consider the stroke required for window shade opening and closing control according to the width of the window.

A control module 05 configured to control the target device according to the first scenario scheme.

Model training module 06 is configured to train a first scene generation model. Model training module 06 includes: knowledge base 061, hierarchical knowledge graph 062, scene vector library 063 and reinforcement learning model 07 in the intelligent family domain.

In the knowledge base 061 of the smart home domain, structured and unstructured data are stored, including but not limited to: product design data, scene design, product planning, enterprise standardization data, product function list, etc. Such data may be obtained from product specifications, customer service orders, after-market orders, marketing orders, or e-commerce evaluations.

The hierarchical knowledge graph 062 is a knowledge graph subjected to deep clustering and grading treatment, and a knowledge structure meeting relevant requirements can be constructed. As shown in fig. 9, the knowledge base 061 of the smart home domain provides data support for the primary knowledge-graph such that the hierarchical knowledge-graph 062 obtained from the primary knowledge-graph includes multi-hierarchy attributes of the smart home domain. The knowledge in the hierarchical knowledge graph 062 is divided into a plurality of different hierarchies according to the degree of correlation and the importance level, so that the scene generation model is convenient to understand the use mode of each intelligent device and the relation between the intelligent devices. The hierarchical knowledge graph can be used as an effective input for training a lightweight model and provides a necessary knowledge system for the model. By constructing the hierarchical knowledge graph 062, new dimensions can be added on the basis of the original full-scale knowledge graph, so that the knowledge graph can better serve a specific business scene, and the value of the knowledge graph in practical application is improved.

The scene vector library 063 is composed of a large number of scene vectors, the scene vector library 063 comprises a large amount of data for scene generation, and the first scene generation model is trained through the scene vector library 063 so that the first scene generation model can combine the large amount of data for scene generation. And finally, enabling the generated scene to better meet the requirements of users.

The reinforcement learning model 07 is used to update the scene vector library, and the reinforcement learning model 07 can expand the scene vectors in the scene vector library 063. The reinforcement learning model 07 may include one or more of a condition generation countermeasure network, a markov decision process, or a Q value learning algorithm. In case of optimization of the countermeasure network by conditional generation, the authenticity of the scene vectors in the scene vector library 063 can also be tested, thereby further optimizing the scene vector library 063.

The control hub device 02 is further provided with a user feedback module 08 configured to collect and process feedback information of the user to update the first scenario generation model by the user.

As shown in fig. 2, the system is capable of training a first scenario generation model in conjunction with a knowledge base 061 of the smart home domain. In the process of generating a scene through the first scene generation model, various data such as user data, device data, environment data, space data and the like are combined. And, the first scene generation model can be updated with feedback from the user. And finally, enabling the generated scene to better meet the requirements of users.

The first scene generation model can control equipment in the smart home according to the requirements of the user, and generate a scene corresponding to the requirements of the user. The user interaction module 03 is configured to obtain a requirement of a user, and input the requirement of the user into the scene generating model. The knowledge base 061, the hierarchical knowledge graph 062, the scene vector library 063 and the reinforcement learning model 07 of the smart home domain are all used to train the first scene generation model.

Since there may be some deviation in understanding the user's needs, after generating the scene, user feedback needs to be collected, understanding the actual emotional changes of the user to the automatically generated scene, and updating and optimizing the scene generation rules accordingly to better meet the user's needs and provide a personalized user experience. Therefore, in order to make the generated scene more in line with the needs of the user, it is necessary to implement modification and optimization of the scene generation model in the smart home.

In generating a scene through a first scene generation model, a software system for optimizing the scene generation model is provided by embodiments of the present disclosure. As shown in fig. 3-1 and 3-2, various data such as user data, device data, environment data, and space data are combined. And, the first scene generation model can be updated with feedback from the user. And finally, enabling the generated scene to better meet the requirements of users.

As shown in fig. 3-1, the user feedback module 08 includes a collection module 081 and an evaluation model 082, and after the collection module 081 collects feedback of the user, feedback information is input into the evaluation model 082, so as to obtain an evaluation result. The control center device 02 can input the evaluation result to the knowledge base 061 in the smart home field to expand the knowledge in the knowledge base. And further, constructing a hierarchical knowledge graph 062 by using the expanded knowledge base, and training a lightweight model through the hierarchical knowledge graph 062 to update the first scene generation model.

As shown in fig. 3-2, the user feedback module 08 includes a collection module 081, an evaluation model 082 and a large language model 083, and the control center device 02 can input the evaluation result into the large language model 083, thereby obtaining correction data. And applying the correction data to the scene vector library 063 to update the scene rules and obtain an updated scene vector library 063. The first scene generation model is trained by the updated scene vector library 063.

In combination with the hardware environment shown in fig. 1 and the software systems shown in fig. 2, 3-1 and 3-2, a method for optimizing a scene generation model is provided in an embodiment of the present disclosure. The method is as shown in fig. 4, and comprises the following steps:

S001, after generating a scene by using the first scene generation model, the control center device obtains satisfaction degree of a user on the generated scene.

And S002, the control center equipment carries out positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user.

S003, the control center equipment inputs the evaluation result into a knowledge base in the intelligent family field, and updates the first scene generation model.

In the embodiments of the present disclosure, the scene generation model can be optimized. In the related art, after the direct feedback of the user is obtained, the current scene can only be updated according to the feedback, but the scene generation model cannot be updated, so that the generated scene still cannot meet the requirement of the user in the subsequent use process of the system. According to the method and the device for generating the scene, the satisfaction degree of the user on the generated scene can be obtained, and the model in the system can be optimized according to the satisfaction degree of the user, so that the scene generated by the model can be more in line with the requirements of the user in the subsequent use process of the user, and the user experience is improved.

The evaluation result is input into a knowledge base in the intelligent family field, the evaluation result can be used as a training material to construct a knowledge graph, and the knowledge graph is used for further training the first scene generation model, so that the scene generated by the first scene generation model meets the requirements of users.

As shown in connection with fig. 5, another method for optimizing a scene generation model provided by an embodiment of the present disclosure includes:

s101, after generating a scene by using a first scene generation model, the control center device acquires satisfaction degree of a user on the generated scene.

S102, the control center equipment carries out positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user.

S103, the control center equipment inputs the evaluation result into the large language model to obtain correction data, and updates the first scene generation model by using the correction data.

In the embodiment of the disclosure, the large language model is a deep learning model and is trained based on massive text data. Natural language text can be generated, meaning of the language text can be understood, and various natural language tasks such as text abstracts, questions and answers, translation and the like can be processed. Large language models are characterized by large scale, contain billions of parameters, and are capable of learning complex patterns in language data. The large language model has the advantages of naturalness, generalization capability, high efficiency, strong task processing capability, high-quality text content generation, strong dialogue system establishment capability, reduced dependence on domain data, promotion of development of the artificial intelligence field and the like. They are able to understand and generate natural language text, accommodating a variety of different languages and contexts.

The evaluation result is input to the large language model, and correction data corresponding to the positive evaluation or the negative evaluation obtained by the evaluation model can be output through the large language model. And applying the correction data to the scene vector library to update the scene rule, and obtaining an updated scene vector library. And training a first scene generating model through the updated scene vector library. The scene generated by the first scene generation model can be more in line with the requirements of users.

Optionally, obtaining satisfaction of the user with the generated scene includes: and monitoring the behavior of the user in the generated scene, and obtaining the satisfaction degree of the user. Illustratively, when the user manually turns off the lamp that has just been turned on, it is the user that is not satisfied with the current scenario.

Optionally, obtaining satisfaction of the user with the generated scene includes: and collecting feedback of the user to obtain satisfaction of the user. For example, user-implicit feedback may be collected. For example, users talk about smart home experiences on social media, or share use experiences with friends, family. These implicit feedback can represent user satisfaction with the scene.

Optionally, obtaining satisfaction of the user with the generated scene includes: and obtaining the satisfaction degree of the user through the A/B test. Illustratively, the satisfaction of the user is tested by creating two or more versions of the control scenario. For example, for the same device, different control strategies may be tested, testing a control strategy that is more popular with users.

Optionally, obtaining satisfaction of the user with the generated scene includes: and obtaining the satisfaction degree of the user according to the prediction model. Based on the user's historical behavior, personal characteristics, and other relevant factors, the user's reaction to a particular scenario may be predicted using machine learning or statistical models. And the satisfaction degree of the user to the scene can be deduced according to the reaction of the user.

Optionally, obtaining satisfaction of the user with the generated scene includes: and obtaining the satisfaction degree of the user according to the user loss rate and the return visit rate. The churn rate and revisit rate of the user can be counted, and if the user is not satisfied with the provided scene, the user can stop using the system, so that the churn rate is increased. If the user is very satisfied, the user may often return, resulting in an increase in return visit rate. Monitoring churn rate and return visit rate can help to learn user satisfaction to obtain user satisfaction.

In the embodiment of the disclosure, the satisfaction degree of the user on the generated scene can be obtained in various modes, so that the satisfaction degree of the user on the scene can be understood according to the satisfaction degree of the user, for example, the user immediately turns off the turned-on lamp to indicate that the user is not satisfied with the current scene.

Further, in the process of acquiring the satisfaction degree of the user on the generated scene, the control center equipment can acquire the satisfaction degree of the user according to the use condition of the user without additional operation of the user, so that the noninductive feedback of the user is realized, and the use experience of the user is improved.

Optionally, inputting the evaluation result into a knowledge base in the smart home domain, and updating the first scene generation model, including expanding knowledge of the knowledge base in the smart home domain according to the evaluation result, to obtain an updated knowledge base. And training the first scene generating model according to the updated knowledge base.

In the embodiment of the disclosure, user feedback can be collected and processed, and the satisfaction degree of the user on the generated scene can be obtained through the feedback information of the user. Through deep understanding of user satisfaction, the first scene generation model can be timely and accurately corrected and optimized, so that user experience is improved.

After the control center equipment obtains feedback information of the user, the feeling of the user and the execution effect of the first scene scheme can be reflected according to the feedback information. And evaluating the feedback information, so that the control center equipment can understand the feeling of the user and the execution effect of the first scene scheme conveniently. Inputting the evaluation results into a knowledge base in the smart home domain, where structured and unstructured data are stored, including but not limited to: product design data, scene design, product planning, enterprise standardization data, product function list, etc. The evaluation result can be stored as new data into a knowledge base in the smart home field. And then, a knowledge graph is constructed through a knowledge base with the evaluation result stored, and the knowledge graph is used for training a first scene generation model, and further training the first scene generation model, so that the scene generated by the first scene generation model meets the requirements of users.

Optionally, performing positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user, including: and according to the satisfaction degree of the user, evaluating the scene corresponding to the satisfaction degree by using an evaluation model to obtain the evaluation score of the scene. And in the case that the evaluation score is higher than or equal to the score threshold value, determining to perform forward evaluation on the generated scene. Or in the case that the evaluation score is lower than the score threshold, determining to perform negative evaluation on the generated scene.

In the embodiment of the disclosure, the scenes corresponding to the satisfaction degree need to be evaluated by using the evaluation model, the evaluation model can score the current scenes according to the satisfaction degree of the user, and the scenes are given an evaluation score. The rating score can represent how well the current scene meets the user's needs. The higher the degree of compliance, the higher the user satisfaction.

And sorting the evaluation scores according to the historical evaluation data by the evaluation model, and selecting a high-score scene and a low-score scene. The evaluation model can determine a score threshold based on the high score scene and the low score scene. And in the case that the evaluation score is higher than or equal to the score threshold value, determining to perform forward evaluation on the generated scene. Or in the case that the evaluation score is lower than the score threshold, determining to perform negative evaluation on the generated scene. In this way, by comparing the evaluation score with the score threshold, the generated scene can be evaluated positively or negatively.

Optionally, before evaluating the scene corresponding to the satisfaction degree by using the evaluation model according to the satisfaction degree of the user, the method further includes: a machine learning model or a statistical model is selected. Training the selected model according to the historical user feedback data to obtain an evaluation model.

Optionally, the machine learning model includes a decision tree, random forest, support vector machine, neural network, or clustering algorithm.

Alternatively, the statistical model comprises linear regression, logistic regression, analysis of variance, chi-square test, or survival analysis.

In the embodiment of the disclosure, training of the evaluation model is particularly important, and positive evaluation or negative evaluation can be performed on the generated scene through the evaluation model. Firstly, a proper model to be trained is selected, and under the condition of selecting a machine learning model, the machine learning model is better in performance when processing large-scale and high-dimensional data, can mine complex relations in the data, and accurately predicts unknown data. The machine learning model can continuously optimize its own parameters through an adaptive learning algorithm to better adapt to the change of data. In the case of selecting a statistical model, the statistical model can give an estimate of the parameter and its degree of influence on the target variable, making the result easier to interpret and understand. And the statistical model is more effective in processing data with obvious statistical rules, and the result is relatively stable. Secondly, the control center equipment can acquire historical user feedback data, and training the selected model according to the historical user feedback data. Therefore, the trained evaluation model can learn the habit of the historical user, and further, a more accurate evaluation result is obtained.

In practical applications, a suitable machine learning model or statistical model may be selected according to circumstances.

Optionally, updating the first scene generation model with the correction data includes: and applying the correction data to the scene vector library to update the scene rule, and obtaining an updated scene vector library. And training through the updated scene vector library to obtain a second scene generation model.

According to the embodiment of the disclosure, according to the correction data, the scene rules in the scene vector library can be updated, so that the scene vectors in the scene vector library are more real, and further, the requirements of users are met more. The scene vector library comprises a large amount of data for scene generation, and the second scene generation model is obtained through training of the scene vector library, so that the second scene generation model can be combined with the large amount of data for scene generation. And finally, enabling the generated scene to better meet the requirements of users.

Optionally, obtaining the scene vector library includes: constructing a hierarchical knowledge graph. The knowledge in the hierarchical knowledge graph is divided into a plurality of different hierarchies according to the degree of correlation and the importance level. And acquiring a plurality of first scene vectors according to the hierarchical knowledge graph. Training a plurality of first scene vectors to obtain a scene vector library.

In the embodiment of the disclosure, the knowledge graph can understand and describe the entities and the relations thereof in the intelligent home environment. The method can carry out visualization processing on elements such as various devices, users and the like and relations among the elements, and is convenient for obtaining scene vectors. The hierarchical knowledge graph is a knowledge graph subjected to deep clustering and grading treatment, and a knowledge structure meeting relevant requirements can be constructed. This optimized and refined knowledge graph can be used as an effective input for training a lightweight model and provide the necessary knowledge system for it. By constructing the hierarchical knowledge graph, new dimensions can be added on the basis of the original full-scale knowledge graph, so that the knowledge graph can better serve a specific business scene, and the value of the knowledge graph in practical application is improved.

By training the first scene vector, a scene vector library can be obtained and used for creating a first scene generation model, so that the generated scene is more in line with the requirements of users, and the use experience of the users is improved.

The scene vector can embody a scene mode which can be generated by the scene generating model, for example, a 'Xiaoming' vector, a vector of household appliances (intelligent sound box and air conditioner), a 'evening' vector and a 'living room' vector are combined to form a scene vector comprehensively considering all elements. Thus, the smart home scene can be described as "the smart home scene is used for watching television news at night by adjusting the air conditioning temperature to 24 degrees through the smart speakers in the living room".

Optionally, constructing a hierarchical knowledge graph includes: and acquiring a knowledge base in the intelligent family field. And constructing a primary knowledge graph according to the knowledge base. Training the primary knowledge graph to obtain a hierarchical knowledge graph.

In the embodiment of the disclosure, the knowledge graph not only can capture the multi-element information of the household electrical appliance, but also can integrate the information to form deep association, thereby providing valuable decision support for links such as design, production, sales and the like of the household electrical appliance. According to the embodiment of the disclosure, the knowledge graph is constructed according to the knowledge base in the intelligent home field, and the knowledge graph is used for training the first scene generation model, so that the first scene generation model can contain the multiple information in the intelligent home field, intelligent home appliances can be better utilized in the generation scene, the generated scene is further more in accordance with the requirements of users, and the use experience of the users is improved.

Further, the hierarchical knowledge graph obtained according to the primary knowledge graph contains multi-level attributes of the intelligent family field, so that the scene generation model is convenient to understand the use mode of each intelligent device and the relation between the intelligent devices, and further the intelligent devices can be better utilized to generate scenes which meet the requirements of users.

Optionally, constructing a primary knowledge graph according to the knowledge base, including: carrying out entity identification and relation establishment on the structured data in the knowledge base; carrying out entity identification and relation extraction on unstructured data lines in a knowledge base; and constructing a primary knowledge graph according to the entity and the relation.

In the embodiment of the disclosure, the structured data is data in a fixed format, for example, data in an Excel format. Some tools, such as the pandas library of Python, may be used to read Excel files and load the data into memory. In the data presented in Excel format, each column may represent an entity type and each row may represent a particular set of entities and relationships between entities. Thus, the primary knowledge-graph can be constructed by traversing the Excel file.

The format of unstructured data is not fixed, e.g., product introduction in the product specification. Entities and relationships may be obtained through natural language processing techniques. And then the obtained entity and relationship are used for constructing a primary knowledge graph.

Optionally, constructing a primary knowledge graph according to the entity and the relationship, including: and according to the entity and the relationship, carrying out entity link and relationship fusion to obtain a primary knowledge graph.

In the disclosed embodiment, entity linking is an important task in the field of natural language processing. The goal of entity linking is to link entities mentioned in the text, such as person names, place names, organization names, etc., with corresponding entities in the knowledge base. The semantics of sentences can be better understood through entity linkage, and text mining and analysis work can be performed on a richer and comprehensive knowledge base.

Optionally, before entity identification and relation extraction are performed on unstructured data in the knowledge base, the method further comprises: text cleansing is performed on unstructured data.

Text cleansing can delete unwanted characters, format text, and misspellings. And thus remove invalid information in the unstructured data. And further, accuracy and efficiency of entity identification and relation extraction are improved.

Optionally, before entity identification and relation extraction are performed on unstructured data in the knowledge base, the method further comprises: and word segmentation is carried out on unstructured data.

The segmentation may segment the continuous text into individual words or tokens. And further, accuracy and efficiency of entity identification and relation extraction are improved.

Optionally, before entity identification and relation extraction are performed on unstructured data in the knowledge base, the method further comprises: and performing stop word removal on unstructured data.

Stop words are words that occur frequently in text but contribute less to the meaning of the text, such as "have", etc. The stop word removal is to remove words that contribute less to the meaning of the text to reduce noise. And further, accuracy and efficiency of entity identification and relation extraction are improved.

Optionally, before entity identification and relation extraction are performed on unstructured data in the knowledge base, the method further comprises: and extracting word stems of unstructured data.

Stem extraction is used to unify words of different forms into the same form, thereby making the text representation more compact and consistent. The method is beneficial to improving the performance of tasks such as text classification, information retrieval and the like, and further improving the accuracy and efficiency of entity identification and relation extraction.

Optionally, training the primary knowledge-graph includes: and clustering the primary knowledge graph. And grading the clustered primary knowledge graph.

In the embodiment of the disclosure, the primary knowledge graphs are clustered, so that the related knowledge entities can be gathered together, and the control center equipment can find all information related to a certain theme or concept more easily. For the primary knowledge graph, clustering can hide the complexity of the underlying structure, thereby providing a more concise and intuitive knowledge view.

The clustered knowledge graph is classified, and each cluster is ordered or classified mainly according to the importance, the relevance or the hierarchical structure of the knowledge entity. By grading, a layering view from macroscopic to microscopic can be provided, and the structure of the knowledge graph can be better understood. Important entities or relationships can be highlighted by ranking, providing more valuable information. And the information or sub-fields required can be rapidly positioned according to the grading result, so that the information retrieval efficiency is improved. Further, the ranking may be updated according to time or other dynamic factors to reflect changes in the knowledge-graph. By grading, the structure and the content of the knowledge graph can be more easily explained, and the understandability of knowledge is improved.

Optionally, clustering the primary knowledge-graph includes: and carrying out relation prediction through a graph neural network. The predicted relationships are added to the primary knowledge-graph.

A graph neural network is a neural network that is dedicated to processing graph structure data. Graph convolution is a basic operation in a graph neural network to capture relationships between nodes. By updating node characteristics in the graph volume lamination, information of neighbor nodes can be aggregated, so that richer node representations are obtained. After the graph rolling, each node contains much information about its neighbors. Relationships between nodes may be predicted using prediction tasks. Common prediction tasks include link prediction between nodes, node attribute prediction, and the like. In the embodiment of the disclosure, the relationship among the nodes in the primary knowledge graph can be predicted through the prediction task, and the predicted relationship is added into the primary knowledge graph, so that the node representation in the primary knowledge graph is more abundant, and the primary knowledge graph is further conveniently layered, so that the hierarchical knowledge graph is obtained.

Optionally, ranking the clustered primary knowledge-graph includes ranking according to a learning algorithm.

In the embodiments of the present disclosure, the learning algorithm includes a machine learning algorithm and a deep learning algorithm. The machine learning algorithm can be directly classified according to the similarity among knowledge points, and the classification efficiency is higher. The deep learning algorithm can learn the nonlinear relation among the knowledge points, so that the grading result is more accurate, and the hierarchical knowledge graph which meets the requirements of users better is obtained.

Optionally, ranking according to a machine learning algorithm includes: extracting features of data in the primary knowledge graph to obtain feature vectors; according to the feature vector, acquiring a grading strategy corresponding to the feature vector; and grading the primary knowledge graph according to a grading strategy.

In the embodiment of the disclosure, feature extraction is performed on data in the primary knowledge-graph, wherein feature vectors can be extracted from nodes in the primary knowledge-graph. Based on the extracted feature vectors, a classification strategy may be formulated. Such as ranking based on the distance between feature vectors of the nodes or the similarity of feature vectors. In this way, the knowledge in the primary knowledge graph can be divided into different levels, and the attributes corresponding to the levels are obtained.

Optionally, ranking the clustered primary knowledge-graph includes ranking by a classifier model.

The classifier model can quickly classify new data, so that the classifier model is widely applied to scenes needing real-time response. And, the implementation and maintenance costs of the classifier model are relatively low compared to learning algorithms.

Optionally, after grading the clustered primary knowledge graph, the method further includes: and evaluating the grading result, and adjusting the grading strategy according to the evaluation result.

In the embodiment of the disclosure, due to the relative disorder of the data in the primary knowledge graph, the classification policy formulated for the first time may cause errors in classification, thereby affecting the classification result. By evaluating the grading result, the grading strategy can be timely adjusted, so that the knowledge in the primary knowledge graph can be more accurate when being graded, and the accuracy of the hierarchical knowledge graph is further improved.

Optionally, training through the updated scene vector library to obtain a second scene generation model, including: and creating a training set according to the updated scene vector library. Wherein the training set comprises scene vectors conforming to the updated scene vector library. A second scene generation model is obtained using training set training.

In the embodiment of the disclosure, the training set includes a scene conforming to the updated scene vector library, and training is performed through the training set, so that the second scene generating model obtained through training can be adjusted based on the updated scene vector library, and further, the scene generated by the scene generating model is more in accordance with the requirement of the user through feedback of the user.

Optionally, training through the updated scene vector library to obtain a second scene generation model, and further includes: and generating an countermeasure network by using the conditions, training a scene vector library, and obtaining a second scene vector. The second scene vector is added to the scene vector library.

The condition generating countermeasure network is a generation model in which condition data is combined with inputs of a generator and a arbiter to control the generated output. This enables the model to generate samples with constraints according to specific conditions, e.g. to generate images of specific categories according to category labels.

The condition generating countermeasure network consists of a generator and a discriminator. The task of the generator is to generate realistic samples from given random inputs and condition data. The task of the arbiter is to distinguish between the real samples and the samples generated by the generator and to try to improve their classification accuracy. In the training process, the generator and the discriminator can perform resistance training, and the respective parameters are continuously optimized to realize better generation and discrimination performance.

In the disclosed embodiments, the scene vector library may be trained by a conditional generation countermeasure network. The condition generation countermeasure network includes a generator and a discriminator. The generator can generate a second scene vector from the input scene vector, and the arbiter can determine whether the generated scene vector is authentic. During training, the generator generates scene vectors, and the arbiter distinguishes between real data and generated data. With such an antagonistic training, the generator will eventually be able to generate a second scene vector that is more closely related to the real data, obtaining a scene generation model. Therefore, the scene scheme generated by the scene generation model is more real, and the generated scene meets the requirements of users.

Optionally, after performing positive evaluation or negative evaluation on the generated scene according to the satisfaction of the user, the method further includes: inputting the evaluation result into a condition generation countermeasure network, and updating the first scene generation model.

In the embodiment of the disclosure, the condition generating countermeasure network can be used for training the scene vector library, and the evaluation result is input into the condition generating countermeasure network, so that the data which can be utilized by the condition generating network are more abundant, and further in the process of generating the scene vector by the generator, the input evaluation result can be utilized to generate a more real scene vector. Therefore, continuous training of the countermeasure network is generated through the conditions, so that scenes in the scene vector library are more real, and the requirements of users can be met.

Optionally, training using the training set to obtain a second scene generation model includes: and training the basic model by using the training set to obtain a second scene generating model.

The second scene generation model is a scene generation model which is obtained by training after integrating user feedback data on the basis of the basic model. Compared with the first scene generation model, the method has more reference data, and a more real scene scheme can be generated according to the requirements of the user, so that the requirements of the user are met.

In an embodiment of the disclosure, the base model includes a neural network-based language model, a recurrent neural network language model, a long-short-term memory network language model, or a gated loop unit language model. These models have wide application in the field of natural language processing, such as speech recognition, machine translation, text classification, and the like.

According to the embodiment of the disclosure, the basic model is trained through the training set, the obtained second scene generation model is integrated with feedback data of the user, and the generated scene meets the requirements of the user.

Optionally, training the first scene generation model using the training set includes: and adjusting the first scene generating model by using the training set to obtain a second scene generating model.

According to the embodiment of the disclosure, the training set is used as new training data, and the user data, the device data, the environment data and the space data are combined, so that the training set, the user data, the device data, the environment data and the space data can be simultaneously utilized when the first scene generation model is trained, and then parameters in the first scene generation model can be adjusted to obtain the second scene generation model. Optionally, generating the scene using the first scene generation model includes: the user's demand is obtained. And inputting the requirements of the user into a first scene generating model to obtain a first scene scheme. Wherein the training data type of the first scene generation model includes user data, device data, environment data, and spatial data. The target device is controlled according to the first scenario scheme.

As shown in connection with fig. 6, another method for optimizing a scene generation model provided by an embodiment of the present disclosure includes:

s201, the control center device obtains the requirements of the user.

S202, the control center equipment inputs the requirements of the user into a first scene generation model to obtain a first scene scheme. Wherein the training data type of the first scene generation model includes user data, device data, environment data, and spatial data.

S203, the control center device controls the target device according to the first scene scheme.

S204, after generating the scene by using the first scene generation model, the control center device obtains the satisfaction degree of the user on the generated scene.

S205, the control center equipment carries out positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user.

S206, the control center equipment inputs the evaluation result into a knowledge base in the intelligent family field to update the first scene generation model, or inputs the evaluation result into the large language model to obtain correction data, and updates the first scene generation model by using the correction data.

The embodiment of the disclosure trains the first scene generating model by utilizing various data such as user data, equipment data, environment data, space data and the like, so that the first scene generating model can better understand entities and relations among the entities in the home environment. Therefore, the first scene generation model can combine multiple factors such as a user, equipment, environment and space to generate a scene, and the generated scene is closer to the requirements of the user, so that the user experience is improved.

Optionally, obtaining the requirement of the user includes: request data of a user is received. Background information is added to the request data. Wherein the context information includes user data, device data, environment data, and spatial data. And carrying out field vectorization processing on the request data added with the background information to obtain the fused request data. And transmitting the fused request data to a classifier. Wherein the classifier is used for classifying the request of the user.

The embodiment of the disclosure adds background information in the request data, so that the content of the request data is richer. The background information includes structured data and unstructured data. The structured data is data in a fixed format, for example, data in an Excel format. The format of unstructured data is not fixed, e.g., product introduction in the product specification. Since the background information contains a large amount of unstructured information, the unstructured information is difficult to directly understand by a model, and therefore, after the background information is added to the request data, the data is subjected to subsequent processing. The processed data is easier to be understood by the classifier, and the classification result of the classifier is more accurate. Finally, the accuracy of the classification model is improved.

Optionally, the classifier classifies the request of the user, including: acquiring a mapping relation between history input data and history output labels; and predicting the request of the user according to the mapping relation to obtain a prediction result.

In the embodiment of the disclosure, the basic working principle of the classifier is to achieve the purpose of predicting actual input data by learning and simulating the mapping relation between the input data and the output label, and further obtain a classification result of a request of a user according to the predicted output label.

Optionally, obtaining the mapping relationship between the history input data and the history output tag includes: acquiring historical input data and a historical output tag corresponding to the historical input data; and learning the history input data and the history output label to obtain a mapping relation.

In the embodiment of the disclosure, the classifier learns through the characteristics of the history input data, and can find the mapping relation between the history input data and the history output label. Specifically, if a deep learning model RNN or LSTM is selected, this process mainly adjusts the weight parameters in the network by back propagation and gradient descent methods, etc. for training.

Optionally, adding context information in the request data includes: background information is collected. Wherein the context information includes user data, device data, environment data, and spatial data. The request data and the context information are combined.

In the embodiment of the disclosure, the control center device can acquire the background information of the current smart home after the user sends out the request data. The background information contains a lot of information related to the scene. By adding such information to the request data, the content of the request data can be made richer, and more effective information can be contained for the classifier to understand and classify. The request data and the background information can be bound by combining the request data and the background information, the request data and the background information can be transmitted simultaneously in the data transmission process, and the request data and the background information can be understood and classified simultaneously when the classifier is used for understanding and classifying. By comprehensively understanding the request data and the background information, the classification result of the classifier is more accurate. Finally, the accuracy of the classification model is improved.

Optionally, merging the request data and the context information includes stitching the request data and the context information.

In the embodiment of the disclosure, the stitching is the most basic and intuitive method, and the background information can be stitched directly behind the text input by the user. For example, if the user's command is "turn on" and the current environment is "living room" obtained from the background information, the spliced text is "turn on" and the living room. This approach is simple and efficient.

Optionally, combining the request data and the context information includes combining the request data and the context information by a sequence generation model.

For more complex cases, such as when text lengths are different or information associations are high, sequence generation models such as RNN (Recurrent Neural Network ), LSTM (Long Short Term Memory, long short term memory network), or transfomer models may be used. RNN or LSTM can handle variable length sequences and capture dependencies in the time dimension of the sequence. In this scenario, the user command and the background may be input to RNN or LSTM, respectively, the time-dependent relationship inside each may be learned, and then the output of the two sequences may be connected or averaged to obtain the final representation. Transformer is a more powerful model that is advantageous over RNNs and LSTMs in handling long distance dependencies and capturing global information. Firstly, word embedding is carried out on a user command and a background respectively, corresponding vector representations are obtained, then the corresponding vector representations are input into a transducer model together, and finally the transducer can output sequence representations integrating vector information of all words.

It should be noted that the above methods all need to operate in conjunction with an efficient embedding manner, such as Word embedding or Positional Encoding, to ensure the quality of the incoming data.

Optionally, performing domain vectorization processing on the request data added with the background information, including: based on the background information, converted content is formed. And inserting the conversion content into the request data of the user to obtain the comprehensive data. The integrated data is converted into a fixed length vector.

The field vectorization processing is to vectorize the data in the household appliance field. Unstructured data can be converted into a structured vector representation by a domain vectorization process. Thus, the performance and accuracy of the model can be improved. Through field vectorization processing, the intrinsic rules and modes of the data can be deeply mined, and the intrinsic structure and characteristics of the data can be better understood, so that the result of the model can be better explained. In addition, the results of domain vectorization processing can be shared among different tasks and domains, thereby improving the reusability of the model. The field vectorization processing is carried out on the request data added with the background information, so that the processed data can be more easily understood by the classifier, and the classification result of the classifier is more accurate. Finally, the accuracy of the classification model is improved.

In the embodiment of the disclosure, the background information has various forms, and the direct insertion of the background information into the request data of the user may lead to the disorder of the inserted data form, which is not beneficial to the understanding and classification of the classifier. The conversion content obtained by the conversion based on the background information is more easily inserted into the request data of the user. And the comprehensive data is converted into a vector with a fixed length, so that the form of the comprehensive data can be unified, and the comprehensive data is input into the classifier, so that the classifier can understand and classify the comprehensive data conveniently, and the classification result of the classifier is more accurate. Finally, the accuracy of the classification model is improved.

Optionally, forming the conversion content according to the background information includes: acquiring a text in the background information according to a natural language processing algorithm; the text is converted into a numerical vector.

In the disclosed embodiments, text needs to be converted into numeric vectors for processing by a machine learning model or a deep learning model. Text conversion is typically aided by Word embedding techniques such as Word2Vec (Word to Vector), gloVe (Global Vectors for Word Representation, global Vector of Word representation) or BERT (Bidirectional Encoder Representations from Transformers, transform-based bi-directional encoder representation) models, and the like. BERT is a pre-trained model that can be fine-tuned to classify user intent. Word embedding is a method commonly used in natural language processing, mainly converting text data into numerical vectors for use by machine learning or deep learning models. Word2Vec is a common Word embedding method that can train a continuous vector space model in which semantically similar words are also located very close in space. The two modes are also classified into CBOW (Continuous Bag of Words, continuous word bag model) and Skip-gram (Skip-gram model). CBOW predicts the context around a word, while Skip-gram predicts the context around from a word. GloVe is another word embedding method, training on a global corpus. GloVe are directed to minimizing reconstruction errors of dense matrices. BERT employs a bi-directional transducer encoder. Unlike most previous pre-trained language models that can only be pre-trained using either left or right contexts, BERT allows the model to be encoded with both left and right contexts by introducing a masking language model objective function.

Each method is suitable for different scenarios, e.g. Word2Vec is suitable when there is a lot of training data and computational resources are limited. GloVe can fully mine global statistics. BERT achieves the best results when dealing with diverse inputs.

Optionally, inserting the conversion content into the request data of the user includes: vectorizing the request data of the user to obtain an input vector of the user; and splicing the input vector of the user with the numerical vector in the background information.

The most basic and intuitive method of stitching the user's input vector with the numeric vector in the background information is to stitch the numeric vector in the background vector information directly to the tail or head of the user's input vector. For example, if the user input is a vector of length N and the background is a vector of length M, then a new vector of length n+m will be obtained after stitching.

Optionally, inserting the conversion content into the request data of the user includes: vectorizing the request data of the user to obtain an input vector of the user; the user's input vector is weighted summed with the numeric vector of the background information.

The input vector of the user and the numerical vector of the background information are weighted and summed, and the other method is to weight and sum the input vector of the user and the numerical vector of the background information according to a certain weight. This method requires defining a weight parameter that indicates how important the two pieces of information are to the final result. The weighted summation can keep the dimension of the combined vector unchanged.

Optionally, inserting the conversion content into the request data of the user includes: vectorizing the request data of the user to obtain an input vector of the user; the user's input vector is mixed with the numeric vector of the background information.

The input vector of the user and the numerical vector of the background information are mixed, or the two vectors can be spliced together to form a longer vector, and then the longer vector is subjected to nonlinear transformation through one or more fully connected layers. The method can better fuse two parts of information, and can learn the optimal fusion method through training.

Optionally, after receiving the request data of the user, the method further includes: converting the request data; and complementing the converted request data.

When converting the request data, firstly, the original input submitted by the user through the interactive device is obtained. The user may submit the input in different ways, including but not limited to, voice commands, text commands, or user behavior. For example, the user may speak to the voice assistant: "turn on the living room air conditioner", or "turn on the air conditioner" with text input on the cell phone. After the interaction device receives the inputs, the control center device performs first processing. For example, if a user submits input using a speech recognition system, the received voice command may be converted to text form. Therefore, after conversion, the original input of the user in different forms can be converted into the same form, so that the original input of the user can be conveniently complemented and corrected.

The user may miss some important information for a number of reasons while interacting. For example, the user may simply say "turn on the air conditioner" without explicitly indicating which room is the air conditioner. To address such issues, the system needs to have the ability to complement some of the missing information. The system may infer and complement missing information according to preset rules. For example, if the air conditioner of the living room is controlled by default when no room is specified according to a certain rule, the system automatically interprets it as "air conditioner of the living room is turned on" when the user says "air conditioner is turned on". If there is no preset rule or the preset rule cannot cope with all cases, it is necessary to infer missing information using historical data and context information. For example, the system may determine which room air conditioner should be operated based on the user's past usage records, current device status, environmental conditions, and the like.

Optionally, after receiving the request data of the user, the method further includes: converting the request data; and correcting the converted request data.

The user may provide erroneous or ambiguous instructions requiring correction of the converted request data. For example, the user may sound unclear, resulting in a speech recognition result of "air conditioning on", and in fact, the user desires to "turn on the water heater". Many modern interactive systems are equipped with powerful language models and error detection algorithms that automatically detect and correct such errors.

As shown in connection with fig. 7, in some embodiments, an apparatus 60 for optimizing a scene generation model includes: the model application module 04 is configured to generate a scene using the first scene generation model. The user feedback module 08 is configured to perform positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user. And inputting the evaluation result into a knowledge base in the intelligent family field, updating the first scene generation model, or inputting the evaluation result into a large language model to obtain correction data, and updating the first scene generation model by utilizing the correction data.

The device 60 for optimizing the scene generation model provided by the embodiment of the present disclosure can optimize the scene generation model. In the related art, after the direct feedback of the user is obtained, the current scene can only be updated according to the feedback, but the scene generation model cannot be updated, so that the generated scene still cannot meet the requirement of the user in the subsequent use process of the system. According to the method and the device for generating the scene, the satisfaction degree of the user on the generated scene can be obtained, and the model in the system can be optimized according to the satisfaction degree of the user, so that the scene generated by the model can be more in line with the requirements of the user in the subsequent use process of the user, and the user experience is improved.

In some embodiments, a computer-readable storage medium comprising a scene conforming to an updated scene vector library comprises a stored program, wherein the program when run performs the method for optimizing a scene generation model described above.

Embodiments of the present disclosure may be embodied in a software product stored on a storage medium, including one or more instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of a method according to embodiments of the present disclosure. While the aforementioned storage medium may be a non-transitory storage medium, such as: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

As shown in connection with fig. 8, in some embodiments, the electronic device comprises a memory 701 and a processor 700, the memory 701 having stored therein a computer program, the processor 700 being arranged to perform the above-described method for optimizing a scene generation model by means of the computer program.

Optionally, the electronic device 70 may also include a communication interface (Communication Interface) 702 and a bus 703. The processor 700, the communication interface 702, and the memory 701 may communicate with each other through the bus 703. The communication interface 702 may be used for information transfer. The processor 700 may call logic instructions in the memory 701 to perform the method for optimizing a scene generation model of the above-described embodiments.

Further, the logic instructions in the memory 701 may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product.

The memory 701 is used as a computer readable storage medium for storing a software program, a computer executable program, and program instructions/modules corresponding to the methods in the embodiments of the present disclosure. The processor 700 executes the functional applications and data processing by running the program instructions/modules stored in the memory 701, i.e. implements the method for optimizing the scene generation model in the above-described embodiments.

The memory 701 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for a function. The storage data area may store data created according to the use of the terminal device, etc. In addition, the memory 701 may include a high-speed random access memory, and may also include a nonvolatile memory.

Claims (11)

1. A method for optimizing a scene generation model, comprising:

after generating a scene by using the first scene generation model, acquiring satisfaction degree of a user on the generated scene;

according to the satisfaction degree of the user, performing positive evaluation or negative evaluation on the generated scene;

inputting the evaluation result into a knowledge base in the intelligent family field, and updating a first scene generation model; or inputting the evaluation result into the large language model to obtain correction data, and updating the first scene generation model satisfaction degree by using the correction data.

2. The method of claim 1, wherein obtaining user satisfaction with the generated scene comprises:

Monitoring the behavior of a user in a generated scene, and obtaining the satisfaction degree of the user; and/or the number of the groups of groups,

Collecting feedback of a user to obtain satisfaction of the user; and/or the number of the groups of groups,

Obtaining satisfaction of a user through an A/B test; and/or the number of the groups of groups,

Obtaining satisfaction of the user according to the prediction model; and/or the number of the groups of groups,

And obtaining the satisfaction degree of the user according to the user loss rate and the return visit rate.

3. The method of claim 1, wherein inputting the evaluation result into a knowledge base of the smart home domain, updating the first scenario generation model, comprises:

According to the evaluation result, the knowledge of the knowledge base in the intelligent family field is expanded, and an updated knowledge base is obtained;

And training the first scene generating model according to the updated knowledge base.

4. The method of claim 1, wherein the positively evaluating or negatively evaluating the generated scene according to the satisfaction of the user comprises:

According to the satisfaction degree of the user, evaluating the scene corresponding to the satisfaction degree by using an evaluation model to obtain the evaluation score of the scene;

determining to perform forward evaluation on the generated scene under the condition that the evaluation score is higher than or equal to a score threshold value; or alternatively

And under the condition that the evaluation score is lower than the score threshold value, determining to carry out negative evaluation on the generated scene.

5. The method of claim 4, further comprising, before evaluating the scene corresponding to the satisfaction using the evaluation model according to the satisfaction of the user:

Selecting a machine learning model or a statistical model;

Training the selected model according to the historical user feedback data to obtain an evaluation model.

6. The method of claim 1, wherein updating the first scene generation model with the correction data comprises:

Applying the correction data to a scene vector library to update scene rules and obtain an updated scene vector library;

And training through the updated scene vector library to obtain a second scene generation model.

7. The method of claim 6, wherein obtaining the second scene generation model through training of the updated scene vector library comprises:

Creating a training set according to the updated scene vector library; wherein the training set comprises scenes conforming to the updated scene vector library;

a second scene generation model is obtained using training set training.

8. The method of claim 7, wherein training using the training set to obtain the second scene generation model comprises:

training the basic model by using a training set to obtain a second scene generation model; or alternatively

And adjusting the first scene generating model by using the training set to obtain a second scene generating model.

9. An apparatus for optimizing a scene generation model, comprising:

a model application module configured to generate a scene using the first scene generation model;

The user feedback module is configured to perform positive evaluation or negative evaluation on the generated scene according to the satisfaction degree of the user; inputting the evaluation result into a knowledge base in the intelligent family field, and updating a first scene generation model; or inputting the evaluation result into the large language model to obtain correction data, and updating the first scene generation model by using the correction data.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program when run performs the method of any one of claims 1 to 8.

11. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to perform the method of any of claims 1 to 8 by means of the computer program.