Data generation method, apparatus, device, and medium

Abstract

To provide a data generation method, an apparatus, an electronic device, a computer-readable storage medium, and a computer program product for realizing interaction with a user by generating an answer based on the content of a user query.SOLUTION: A data generation method comprises: generating first answer data based on first query data from a user; determining a first reconsideration result on the first answer data based on the first answer data and negative feedback in response to receiving the negative feedback of the user on the first answer data, wherein the first reconsideration result indicates a diagnostic reason why the user feedback on the first answer data is negative feedback; and generating second answer data on the first query data based on the first query data and the first reconsideration result.SELECTED DRAWING: Figure 2

Description

Technical Field

[0005] , , ,

[0004] , , ,

[0006] ,

[0001] The present disclosure relates to the field of artificial intelligence technology, particularly to fields such as natural language processing and deep learning. Specifically, it relates to a data generation method, apparatus, electronic device, computer-readable storage medium, and computer program product.

Background Art

[0002] Artificial intelligence is a subject that studies how to simulate some human thinking processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.) on a computer. There are both hardware technologies and software technologies. The hardware technologies of artificial intelligence generally include technologies such as sensors, artificial intelligence dedicated chips, cloud computing, distributed storage, and big data processing. The artificial intelligence software technologies mainly include several major directions such as computer vision technology, speech recognition technology, natural language processing technology, and machine learning / deep learning, big data processing technology, and knowledge graph technology.

[0003] The generative language megamodel can be applied to various natural language processing tasks. In particular, it can realize interaction with users by generating natural language texts for answers based on the content of the user's query.

[0004] The methods described in this part are not necessarily the methods previously assumed or adopted. Unless otherwise specified, none of the methods described in this part should be considered as prior art just because they are included in this part. Similarly, unless otherwise specified, the problems mentioned in this part should not be considered as approved by any prior art.

Summary of the Invention

[0005] The present disclosure provides a data generation method, apparatus, electronic device, computer-readable storage medium, and computer program product.

[0006] According to one aspect of the present disclosure, a data generation method is provided, the method comprising: generating first response data based on first query data from a user; determining a first reconsideration result for the first response data based on the first response data and the negative feedback in response to receiving negative feedback from the user for the first response data, the first reconsideration result indicating the reason for diagnosing that the user's feedback for the first response data is negative feedback; and generating second response data for the first query data based on the first query data and the first reconsideration result.

[0007] In another aspect of the present disclosure, a data generation apparatus is provided, the apparatus comprising: a first generation unit configured to generate first response data based on first query data from a user; a finalization unit configured to determine a first reconsideration result for the first response data based on the first response data and the negative feedback, in response to receiving negative feedback from the user for the first response data, the first reconsideration result indicating the diagnostic reason that the user's feedback for the first response data is negative feedback; and a second generation unit to generate second response data for the first query data based on the first query data and the first reconsideration result.

[0008] According to another aspect of the present disclosure, an electronic device is provided, the electronic device comprising at least one processor and a memory communicated to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the at least one processor to perform the data generation method.

[0009] According to another aspect of the present disclosure, a non-temporary computer-readable storage medium is provided which stores computer instructions, said computer instructions are used to cause the computer to execute the data generation method described above.

[0010] According to another aspect of this disclosure, a computer program product is provided which, when executed by a processor, can realize the data generation method described above.

[0011] According to one or more embodiments of this disclosure, the quality of response data generation can be improved.

[0012] It should be understood that the content described in this section is not intended to identify any essential or important features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure are readily apparent from the following specification. [Brief explanation of the drawing]

[0013] The drawings illustrate embodiments and constitute part of the specification, and are used to illustrate exemplary embodiments of the embodiments together with the textual description of the specification. The illustrated embodiments are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements. [Figure 1] This is a schematic diagram showing an exemplary system in which various methods described herein can be carried out according to exemplary embodiments of the present disclosure. [Figure 2] This flowchart shows a data generation method according to an exemplary embodiment of the present disclosure. [Figure 3] This is a schematic diagram illustrating the data generation process according to an exemplary embodiment of the present disclosure. [Figure 4] This is a block diagram showing a data generation device according to an exemplary embodiment of the present disclosure. [Figure 5]This is a block diagram showing an exemplary electronic device that can be used to realize an embodiment of the present invention. [Modes for carrying out the invention]

[0014] Hereinafter, exemplary embodiments of the present disclosure will be described in conjunction with the drawings, and the various details of the embodiments of the present disclosure contained herein should be considered as illustrative only, provided to aid understanding. Accordingly, those skilled in the art should be aware that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for clarity and conciseness, descriptions of known functions and structures have been omitted in the following description.

[0015] In this disclosure, unless otherwise specified, the use of terms such as “first,” “second,” etc., to describe various elements is not intended to limit the spatial, timing, or importance relationships of these elements. Such terms are used solely to distinguish one element from another. In some examples, the first element and the second element may refer to the same example of that element, or, depending on the contextual description, to different examples.

[0016] The terms used in describing the various examples in this disclosure are for illustrative purposes only and are not intended to limit them. Unless otherwise explicitly indicated in the context, such elements may be one or more, unless the number of elements is specifically limited. The terms "and / or" as used in this disclosure cover any of the listed items and all possible combinations thereof.

[0017] In related technologies, when generating response data based on user input data using generative language models, the model's performance can be improved by manually labeling the corpus during the model training phase or by adjusting the training method (for example, by optimizing the loss function or performing reinforcement learning). However, in the application phase of the model, i.e., the data generation process, response data is usually generated directly based only on the query data entered by the user. This makes it difficult to adjust the response data in response to user feedback, and as a result, the quality of the response data may not fully meet the user's needs.

[0018] Based on this, the disclosure provides a data generation method that generates initial response data for user query data, and then, upon receiving negative feedback from the user regarding this response data, self-diagnoses the reason why the response data received negative feedback based on this negative feedback, and further generates a reconsideration result for this response data, thereby generating new response data based on this reconsideration result, thereby better matching the response data to the user's needs and improving the quality of response data generation.

[0019] The embodiments of this disclosure will be described in detail below with reference to the drawings.

[0020] Figure 1 shows a schematic diagram of an exemplary system 100 in which various methods and apparatus described herein can be implemented according to embodiments of the present disclosure. Referring to Figure 1, the system 100 includes one or more client devices 101, 102, 103, 104, 105 and 106, a server 120, and one or more communication networks 110 that connect one or more client devices to the server 120. The client devices 101, 102, 103, 104, 105 and 106 can be configured to run one or more applications.

[0021] In an embodiment of the present disclosure, the server 120 can execute one or more services or software applications that enable the execution of a data generation method.

[0022] In some examples, the server 120 can also provide other services or software applications that can include a non-virtual environment and a virtual environment. In some examples, these services can be provided as web-based services or cloud services, for example, provided to users of the client devices 101, 102, 103, 104, 105, and / or 106 in a software as a service (SaaS) model.

[0023] In the arrangement shown in FIG. 1, the server 120 may include one or more assemblies that implement the functions executed by the server 120. These assemblies may include software assemblies, hardware assemblies, or combinations thereof that can be executed by one or more processors. Users operating the client devices 101, 102, 103, 104, 105, and / or 106 can interact with the server 120 by sequentially using one or more client applications to utilize the services provided by these assemblies. It should be understood that various different system arrangements are possible and may be different from the system 100. Therefore, FIG. 1 is an example of a system for implementing various methods described herein and is not intended to be limiting.

[0024] Users can send query data and feedback using the client devices 101, 102, 103, 104, 105, and / or 106. The client device can provide an interface through which a user of the client device can interact with the client device. The client device can also output information to the user through the interface. Although only six client devices are illustrated in FIG. 1, as will be understood by those skilled in the art, the present disclosure can support any number of client devices.

[0025] Client devices 101, 102, 103, 104, 105, and / or 106 may include various types of computer devices such as portable handheld devices, general-purpose computers (e.g., personal computers and notebook computers), workstation computers, wearable devices, smart screen devices, self-service terminal devices, service robots, game systems, thin clients, various messaging devices, sensors, or other detection devices. These computer devices can run various types and versions of software applications and operating systems such as MICROSOFT Windows, APPLE iOS, UNIX-like (registered trademark) operating systems, Linux (registered trademark) or Linux-like (registered trademark) operating systems (e.g., GOOGLE Chrome OS), or include various mobile operating systems such as MICROSOFT Windows Mobile OS, iOS, Windows Phone, Android. Portable handheld devices may include mobile phones, smartphones, tablets, personal digital assistants (PDAs), etc. Wearable devices may include head-mounted displays (e.g., smart glasses) and other devices. Game systems may include various handheld game devices, Internet-connected game devices, etc. Client devices can, for example, run Internet-related applications, communication applications (e.g., email applications), short message service (SMS) applications, and can run various applications and use various communication protocols.

[0026] Network 110 may be any type of network known to those skilled in the art, and it may use any one of several available protocols (including but not limited to TCP / IP, SNA, IPX, etc.) to support data communication. For example, one or more networks 110 may be a local area network (LAN), an Ethernet-based network, a token loop, a wide area network (WAN), the internet, a virtual network, a virtual private network (VPN), an intranet, an extranet, a blockchain network, a public switched telephone network (PSTN), an infrared network, a wireless network (e.g., Bluetooth®, WIFI), and / or any combination of these and / or other networks.

[0027] Server 120 may include one or more general-purpose computers, dedicated server computers (e.g., PC (personal computer) servers, UNIX® servers, midrange servers), blade servers, large computers, server clusters, or any other suitable configuration and / or combination. Server 120 may include one or more virtual machines running a virtual operating system, or other computing architectures related to virtualization (e.g., one or more flexible pools of virtualized logical memory devices to maintain the server's virtual memory devices). In various embodiments, Server 120 may run one or more services or software applications that provide the functions described below.

[0028] The computing units in server 120 can run one or more operating systems, including any of the above-mentioned operating systems and any commercial server operating systems. Server 120 can also run any one of a variety of additional server applications and / or middle-tier applications, including HTTP servers, FTP servers, CGI servers, Java® servers, database servers, etc.

[0029] In some embodiments, the server 120 may include one or more applications for analyzing and integrating data feeds and / or event updates received from users of client devices 101, 102, 103, 104, 105, and 106. The server 120 may also include one or more applications for displaying data feeds and / or real-time events via one or more display devices of client devices 101, 102, 103, 104, 105, and 106.

[0030] In some embodiments, server 120 may be a server in a distributed system or a server incorporating blockchain. Server 120 may be a cloud server, or an intelligent cloud computing server or intelligent cloud host equipped with artificial intelligence technology. A cloud server is a host product in a cloud computing service system and solves the problems of high management difficulty and low business scalability that exist in conventional physical hosts and virtual private server (VPS) services.

[0031] System 100 may also include one or more databases 130. In some embodiments, these databases can be used to store data and other information. For example, one or more of the databases 130 can be used to store information such as audio files and video files. The databases 130 can be located in various locations. For example, a database used by server 120 may be located locally with server 120, or it may be located away from server 120 and communicate with server 120 via a network or a dedicated connection. The databases 130 may be of different types. In some embodiments, a database used by server 120 may be, for example, a relational database. One or more of these databases can store, update, and retrieve data from the databases in response to instructions.

[0032] In some embodiments, one or more of the databases 130 may be used by an application to store application data. The databases used by the application may be of different types, such as a key-value repository, an object repository, or a general-purpose repository supported by a file system.

[0033] The system 100 in Figure 1 can be configured and operated in various ways so as to allow the application of the various methods and apparatus described in this disclosure.

[0034] Figure 2 shows a flowchart of the data generation method 200 according to an exemplary embodiment of the present disclosure. As shown in Figure 2, the method 200 is: Step S201 generates first response data based on first query data from the user, Step S202, in response to receiving negative feedback from the user regarding the first response data, determines a first reconsideration result for the first response data based on the first response data and the negative feedback, and the first reconsideration result indicates the reason for diagnosing that the user's feedback regarding the first response data is negative feedback. The process includes step S203, which generates second response data for the first query data based on the first query data and the first reconsideration result.

[0035] According to the data generation method described in Method 200, after generating first response data for query data entered by the user, if negative feedback from the user is received regarding the first response data, a revised result for the first response data is generated based on this negative feedback, thereby achieving self-diagnosis. Furthermore, new second response data is generated based on this revised result, making the new second response data more aligned with the user's needs and improving the quality of response data generation.

[0036] In some examples, the first query data, first answer data, second answer data, and first reconsideration result may be natural language text. For example, the first query data may be "Please write a five-word car slogan," in which case the generated first answer data would be a slogan containing seven words. If the user gives negative feedback on the first answer data, the system can self-diagnose the reason for receiving the negative feedback and obtain a first reconsideration result such as "The user requested a five-word slogan, but the generated result contains seven words, and the number of words does not match the user's request." By using the first reconsideration result to indicate the possible causes of the first answer data receiving negative feedback, a new second answer data can be generated based on this, thereby avoiding the second answer data receiving negative feedback from the user again, and thus improving the quality of the generated answer data.

[0037] According to several embodiments, the first reconsideration result further includes an optimization policy for the first response data. By generating data based on the reconsideration result which further includes the optimization policy, it is possible to obtain a second response data that is more goal-oriented, richer in content, and more specific based on the reconsideration result.

[0038] In the example above, the rethinking result could be: "The user requested a 5-word catchphrase, but the generated result contains 7 words, which does not meet the user's request, and it is necessary to guarantee that the result will be 5 words when regenerated." By generating new second response data based on the rethinking result, which includes the optimization policy "it is necessary to guarantee that the result will be 5 words when regenerated," the quality of the second response data generation can be improved, further enhancing the user experience.

[0039] According to some embodiments, in step S201, generating first response data based on first query data from a user includes determining first input data to be used in a deep learning model for generating response data based on input data, based on the first query data, and inputting the first input data into the deep learning model to obtain the first response data. In step S203, generating second response data for the first query data based on the first query data and the first reconsideration result includes determining second input data to be used in the deep learning model, based on the first query data and the first reconsideration result, and inputting the second input data into the deep learning model to obtain the second response data. This allows response data to be generated using a deep learning model, and by generating first and second input data that indicate different needs and inputting them into a deep learning model, i.e., generating first and second response data using the same deep learning model, efficiency and convenience can be improved.

[0040] In some examples, a deep learning model for generating response data based on input data possesses end-to-end characteristics and can directly generate response data in natural language text format based on input data in natural language text format. In some examples, this deep learning model can employ an N-layer Transformer network structure with encoders and decoders, or a Unified pre-trained Language Model (UniLM) network structure. It should be understood that this deep learning model may also be a neural network model based on other Transformer network structures, and is not limited thereto.

[0041] In some cases, this deep learning model can be obtained by training it using a sample corpus, which may include, for example, sample input data and sample response data for this sample input data. During the training process, the sample input data can be input into the deep learning model to obtain predicted response data, a loss value can be calculated based on the predicted response data and the sample response data, and the parameters of the deep learning model can be adjusted based on the loss value. In some cases, the loss value of the deep learning model can be determined based on the Negative-Log Likelihood Loss (NLL) calculation method.

[0042] In some examples, different corpora are used to train the deep learning models applied to step S201 and step S203, and by using more oriented training corpora, the accuracy of the responses generated by the models can be improved.

[0043] According to some embodiments, determining the second input data to be used in the deep learning model based on the first query data and the first reconsideration result includes determining the second input data based on the first query data, the first reconsideration result, and task description information indicating that the second input data includes the first reconsideration result. This allows the current data generation needs to be explicitly indicated by adding task description information based on the input content, enabling the deep learning model to generate the second input data based on the first reconsideration result and the first query data, thereby improving the efficiency of data generation.

[0044] Referring to the example above, in one example, the first query data, the first response data, the second response data, and the first reconsideration result are all natural language text, and the deep learning model is also used to receive and generate natural language text. In this case, the task description information may be a pre-configured natural language text segment or template to indicate the presence of the first query data and the first reconsideration result in the input data. This task description information may, for example, state to generate response data for the first query data based on the first query data and the first reconsideration result. By filling in the contents of the first query data and the first reconsideration result in this template, input data that clearly indicates the data generation needs can be obtained, and the deep learning model can generate the second response data based on this.

[0045] The above embodiments are merely examples of input data for a deep learning model. Input data can be generated based on other methods, for example, second input data can be obtained by directly adding labels of a predefined generation pattern based on first query data and first reconsideration results, and the deep learning model can determine that the second input data contains the first reconsideration results based on these tags. The disclosure does not limit the method of determining input data as long as the deep learning model can sense the presence of reconsideration results in the input data and generate response data based on these reconsideration results.

[0046] According to some embodiments, in step S202, determining a first reconsideration result for the first response data based on the first response data and the first feedback includes inputting the first response data and the first feedback into a reconsideration generation network to obtain the first reconsideration result output by the reconsideration generation network, where the reconsideration generation network is obtained by training using a sample corpus, and the sample corpus includes sample response data, sample feedback, and sample reconsideration results for the sample response data. This makes it possible to obtain reconsideration results using a trained reconsideration generation network and improve the efficiency and convenience of generating reconsideration results.

[0047] In some examples, the rethinking generative network can employ an N-layer Transformer network structure with encoders and decoders, or a Unified pre-trained Language Model (UniLM) network structure. Similar to the training methods of the deep learning models described above, this rethinking generative network can be obtained by training using a sample corpus, and the sample corpus for training the rethinking generative network may include, for example, sample response data, sample feedback, and sample rethinking results for the sample response data. During the training process, by inputting the sample response data and sample feedback into this deep learning model, predictive rethinking results can be obtained, a loss value can be calculated based on the predictive rethinking results and sample rethinking results, and the parameters of the rethinking generative network can be further adjusted based on the loss value.

[0048] According to some embodiments, in step S202, in response to receiving negative feedback from the user regarding the first response data, determining a first reconsideration result for the first response data based on the first response data and the negative feedback includes, in response to receiving first feedback from the user regarding the first response data, determining a first reconsideration result for the first response data based on the first response data and the first feedback, in response to determining that the first feedback is negative feedback. This allows for further determination of whether the feedback data is negative after obtaining the user's feedback data, thereby allowing for the identification of negative feedback by the determination step and improving accuracy if the feedback data does not explicitly indicate whether the feedback is negative.

[0049] In some cases, a classifier-based user feedback recognition model can be used to distinguish whether this first feedback is negative or not. In some cases, the user feedback recognition model can be trained using a sample feedback dataset with actual attribute labels (including positive and negative feedback), thereby more efficiently and accurately determining whether the first feedback from the user is negative or not.

[0050] In some cases, it is also possible to determine whether negative feedback has been received from a user through other means. For example, it can be determined that negative feedback has been received based on user actions on the user interface. For instance, it can be determined that negative feedback has been received in response to the user clicking a negative rating button or giving a score below a threshold.

[0051] According to some embodiments, Method 200 further includes generating a third response data for the second query data based on the first query data, the second response data, and the second query data, in response to determining that the similarity between a second query data from a user and the first query data is greater than a preset threshold. This allows the generation of the current third response data to be enhanced using the first query data and the second response data, thereby improving the quality of the response data generation, when a user submits second query data that is similar to the first query data.

[0052] In some examples, the similarity between the first and second sets of query data can be determined by representing both sets of query data as text vectors and calculating vector similarity.

[0053] According to some embodiments, Method 200 further includes storing the first query data and the second answer data in a memory bank, wherein generating a third answer data for the second query data based on the first query data, the second answer data, and the second query data in response to the determination that the similarity between the second query data from the user and the first query data in the memory bank is greater than the predetermined threshold includes retrieving the second answer data from the memory bank in response to the determination that the similarity between the second query data from the user and the first query data in the memory bank is greater than the predetermined threshold, and generating the third answer data based on the first query data, the second answer data, and the second query data. This makes it possible to store query data-answer data pairs using a memory bank, and by using a memory bank, it is possible to store a larger number of historical interaction data over a longer period, and further enhance the generation of current answer data by referring to historical interactions and improving the quality of answer data generation.

[0054] In some examples, when generating response data using the deep learning model described above, a third input data used in the deep learning model can be determined based on the first query data, the second response data, and the second query data. The third response data can then be input into the deep learning model to obtain the third response data. In one example, this third response data may include explanatory information indicating the existence of the first query data and the second response data, for example, so that the deep learning model can generate the third response data based on it.

[0055] Figure 3 shows a schematic diagram of a data generation process according to an exemplary embodiment of the present disclosure. As shown in Figure 3, in one example, the data generation process may enable interaction with a user using an intelligent dialogue system 300.

[0056] Referring to Figure 3, the intelligent dialogue system 300 includes a user feedback recognition model 301, a rethinking generative network 302, a deep learning model 303, and a memory bank 304.

[0057] In one example, a data transmission path including a user feedback recognition model 301, a rethinking generation network 302, and a memory bank 304 is optional. If a user sends first query data and the memory bank 304 does not contain historical data similar to the first query data, the deep learning model 303 can directly generate first response data based on the first query data entered by the user. After receiving the user's first feedback on the first response data, the user feedback recognition model 301 can be used to determine whether this first feedback is negative feedback. If it is determined that this first feedback is negative feedback, the rethinking generation network 302 can be used to obtain a first rethinking result based on the first response data and the first feedback. Furthermore, the deep learning model 303 can determine new second response data based on the first query data and the first rethinking result, and the first query data and second response data can be stored in the memory bank 304.

[0058] When a second query data is received and it is determined that a first query data similar to the second query data 304 already exists in memory bank 304, the deep learning model 303 is instructed to determine a third response data for the second query data based on the first query data, the second response data, and the second query data. This allows the current generation of the third response data to be enhanced using the first query data and the second response data, thereby improving the quality of the response data generation.

[0059] One aspect of the present disclosure further provides a data generation device. Figure 4 shows a block diagram of a data generation device 400 according to an exemplary embodiment of the present disclosure. As shown in Figure 4, the device 400 includes a first generation unit 401 configured to generate first response data based on first query data from a user; a finalization unit 402 configured to finalize a first reconsideration result for the first response data based on the first response data and the negative feedback in response to receiving negative feedback from the user for the first response data, wherein the first reconsideration result indicates the diagnostic reason why the user's feedback for the first response data is negative feedback; and a second generation unit 403 to generate second response data for the first query data based on the first query data and the first reconsideration result.

[0060] According to some embodiments, the first generation unit 401 includes a first determination subunit configured to determine first input data to be used in a deep learning model for generating response data based on input data, based on the first query data, and a first input subunit configured to obtain the first response data by inputting the first input data into the deep learning model, where the second generation unit 403 includes a second determination subunit configured to determine second input data to be used in the deep learning model, based on the first query data and the first reconsideration result, and a second input subunit configured to obtain the second response data by inputting the second input data into the deep learning model.

[0061] According to some embodiments, the second input subunit is configured to determine the second input data based on the first query data, the first reconsideration result, and task description information indicating that the second input data includes the first reconsideration result.

[0062] According to some embodiments, the final unit 402 inputs the first response data and the first feedback into the rethinking generation network to obtain the first rethinking result output by the rethinking generation network, where the rethinking generation network is obtained by training with a sample corpus, and the sample corpus is configured to include sample response data, sample feedback, and sample rethinking results for the sample response data.

[0063] According to some embodiments, the confirmation unit 402 is configured to confirm a first reconsideration result for the first response data based on the first response data and the first feedback, in response to receiving first feedback from the user for the first response data, and in response to confirming that the first feedback is negative feedback.

[0064] According to some embodiments, the apparatus 400 further includes a third generation unit configured to generate a third response data for the second query data based on the first query data, the second response data, and the second query data, in response to the determination that the similarity between a second query data from a user and the first query data is greater than a preset threshold.

[0065] According to some embodiments, the apparatus 400 further includes a storage unit configured to store the first query data and the second answer data in a memory bank, wherein the third generation unit includes an acquisition subunit configured to acquire the second answer data from the memory bank in response to the determination that the similarity between the second query data from the user and the first query data in the memory bank is greater than the preset threshold, and a generation subunit configured to generate the third answer data based on the first query data, the second answer data, and the second query data.

[0066] According to some embodiments, the first reconsideration result further includes an optimization policy for the first response data.

[0067] In the proposed technology described herein, all processing of relevant user personal information, including collection, storage, use, processing, transmission, provision, and disclosure, complies with the provisions of relevant laws and regulations and does not violate public order and morals.

[0068] Another aspect of the present disclosure provides an electronic device comprising at least one processor and a memory communicated to the at least one processor, the memory storing instructions executable by the at least one processor, and the instructions being executed by the at least one processor causing the at least one processor to perform the data generation method described above.

[0069] Another aspect of the present disclosure further provides a non-temporary computer-readable storage medium in which computer instructions are stored, the computer instructions being used to cause the computer to perform the data generation method described above.

[0070] According to another aspect of this disclosure, a computer program product is further provided, which includes a computer program, when executed by a processor, implements the data generation method described above.

[0071] Referring to Figure 5, a block diagram of the configuration of an electronic device 500 that functions as a server or client of the present disclosure is described, which is an example of hardware equipment applicable to each aspect of the present disclosure. The electronic device includes various forms of digital electronic computer devices, such as laptop computers, desktop computers, stages, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may further include various forms of mobile devices, such as personal digital processing devices, mobile phones, intelligent phones, wearable devices, and other similar computing devices. The components, their connections, and their functions shown herein are illustrative and do not limit the implementation of the present disclosure as described and / or claimed herein.

[0072] As shown in Figure 5, the device 500 includes a computing unit 501, which can perform various appropriate operations and processes by computer programs stored in read-only memory (ROM) 502 or by computer programs loaded from storage unit 508 into random access memory (RAM) 503. The RAM 503 may store various programs and data necessary to operate the device 500. The computing unit 501, ROM 502, and RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0073] Multiple components within the device 500 are connected to an I / O interface 505, which comprises an input unit 506, an output unit 507, a storage unit 508, and a communication unit 509. The input unit 506 may be any type of device capable of inputting information into the device 500, and may receive input numeric or character information and generate key signal inputs relating to user settings and / or function control of the electronic device, and may include, but is not limited to, a mouse, keyboard, touchscreen, trackboard, trackball, lever, microphone, and / or remote control. The output unit 507 may be any type of device capable of presenting information, and may include, but is not limited to, a display, speaker, video / audio output terminal, vibrator, and / or printer. The storage unit 508 may include, but is not limited to, a magnetic disk or an optical disk. The communication unit 509 enables the device 500 to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth® devices, 802.11 devices, WiFi devices, WiMAX devices, cellular communication devices, and / or similar devices.

[0074] The computing unit 501 may be a variety of general-purpose and / or dedicated processing components having processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that execute machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs each of the methods and processes described in the preamble, for example, the data generation method. For example, in some embodiments, the data generation method may be implemented as a computer software program tangibly contained in a machine-readable medium, for example, a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed in the device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the data generation method described above can be performed. Alternatively, in another embodiment, the computing unit 501 may be configured to perform the data generation method in any other suitable manner (for example, by firmware).

[0075] Various embodiments of the systems and technologies described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may be implemented in one or more computer programs, which may be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, which may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0076] Program code for carrying out the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, a dedicated computer, or other programmable data processing device, and when the program code is executed by the processor or controller, the functions / operations defined in the flowcharts and / or block diagrams are performed. The program code may be executed entirely by machine, partially by machine, partially by machine and partially by remote machine as a standalone software package, or entirely by remote machine or server.

[0077] In the context of this disclosure, a machine-readable medium may be a tangible medium that contains or stores a program used in or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any appropriate combination thereof. More specific examples of machine-readable storage media include one or more wired electrical connections, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any appropriate combination thereof.

[0078] To provide user interaction, the computer may implement the systems and technologies described herein, the computer comprising a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitoring monitor), and a keyboard and pointing device (e.g., a mouse or trackball), the user may input to the computer via the keyboard and pointing device. Other types of devices may further provide user interaction, for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback), and the computer may receive input from the user in any form (including sound input, voice input, or tactile input).

[0079] The systems and technologies described herein may be implemented in computing systems including backstage components (e.g., as data servers), computing systems including middleware components (e.g., application servers), computing systems including front-end components (e.g., user computers having a graphical user interface or web browser, through which users can interact with embodiments of the systems and technologies described herein), or in computing systems including any combination of such background components, middleware components, or front-end components. The components of the system may be interconnected by digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local networks (LANs), wide area networks (WANs), the internet, and blockchain networks.

[0080] A computer system may include a client and a server. The client and server are generally geographically distant from each other and typically interact via a communication network. The client-server relationship is created by running computer programs on corresponding computers that have a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server combined with a blockchain.

[0081] It should be understood that the steps may be reordered, added, or deleted using the various forms of flows described above. For example, each step described in this disclosure may be performed in parallel, sequentially, or in a different order, as long as the technical proposal disclosed herein achieves the desired result.

[0082] While examples of the embodiments or cases of this disclosure have been described with reference to the drawings, it should be understood that the methods, systems, and apparatus described above are merely illustrative examples, and the scope of the invention is not limited by these examples. Various elements of the embodiments or cases may be omitted or replaced by equivalent elements. Furthermore, each step may be performed in an order different from the order described herein. Moreover, various elements of the embodiments or cases may be combined in various ways. Importantly, as technology advances, many of the elements described herein can be replaced by equivalent elements appearing later in this disclosure.

Claims

1. A data generation method performed by a computer, wherein the method is: To generate first response data based on first query data from the user, In response to receiving negative feedback from the user regarding the first response data, a first reconsideration result for the first response data is determined based on the first response data and the negative feedback, and the first reconsideration result indicates the reason why the user's feedback regarding the first response data is negative feedback. This includes generating a second response data to the first query data based on the first query data and the first reconsideration result, Based on the first response data and the negative feedback, determining the first reconsideration result for the first response data is: A method comprising inputting the first response data and the negative feedback into a rethinking generation network to obtain the first rethinking result output by the rethinking generation network, wherein the rethinking generation network is obtained by training using a sample corpus, and the sample corpus includes sample response data, sample feedback, and sample rethinking results for the sample response data.

2. Generating the first response data based on the first query data from the user, as described above, Based on the first query data, the first input data used in a deep learning model for generating response data based on the input data is determined, This includes inputting the first input data into the deep learning model to obtain the first response data, Here, generating second response data for the first query data based on the first query data and the first reconsideration result is: Based on the first query data and the first reconsideration result, the second input data used in the deep learning model is determined, The method according to claim 1, further comprising inputting the second input data into the deep learning model to obtain the second response data.

3. Based on the aforementioned first query data and the first reconsideration result, determining the second input data used in the deep learning model is: The method according to claim 2, comprising determining the second input data based on the first query data, the first reconsideration result, and task description information indicating that the second input data includes the first reconsideration result.

4. In response to receiving negative feedback from the user regarding the first response data, determining a first reconsideration result for the first response data based on the first response data and the negative feedback is: The method according to any one of claims 1 to 3, comprising determining a first reconsideration result for the first response data based on the first response data and the first feedback, in response to receiving first feedback from the user for the first response data, and in response to determining that the first feedback is negative feedback.

5. The method according to any one of claims 1 to 3, further comprising generating a third response data for the second query data based on the first query data, the second response data, and the second query data, in response to the determination that the similarity between second query data from a user and the first query data is greater than a preset threshold.

6. The method further includes storing the first query data and the second response data in a memory bank. In response to the determination that the similarity between the second query data from the user and the first query data is greater than a predetermined threshold, generating a third response data for the second query data based on the first query data, the second response data, and the second query data is: In response to the determination that the similarity between the second query data from the user and the first query data in the memory bank is greater than the preset threshold, the second response data is retrieved from the memory bank. The method according to claim 5, comprising generating the first query data, the second response data, and the third response data based on the second query data.

7. The method according to any one of claims 1 to 3, further comprising the first reconsideration result being an optimization policy for the first response data.

8. A data generation device, wherein the device is A first generation unit configured to generate first response data based on first query data from a user, A confirmation unit configured to confirm a first reconsideration result for the first response data based on the first response data and the negative feedback, in response to receiving negative feedback from the user for the first response data, wherein the first reconsideration result indicates the reason for diagnosing that the user's feedback for the first response data is negative feedback. The system includes a second generation unit that generates a second response data for the first query data based on the first query data and the first reconsideration result, The aforementioned confirmation unit is The device is configured to obtain the first reconsideration result output by the reconsideration generation network by inputting the first response data and the negative feedback into the reconsideration generation network, wherein the reconsideration generation network is obtained by training using a sample corpus, and the sample corpus includes sample response data, sample feedback, and sample reconsideration results for the sample response data.

9. The first generation unit is A first determinative subunit configured to determine first input data used in a deep learning model for generating response data based on input data, based on the first query data, It includes a first input subunit configured to obtain the first response data by inputting the first input data into the deep learning model, The second generation unit is, A second determinative subunit configured to determine the second input data used in the deep learning model based on the first query data and the first reconsideration result, The apparatus according to claim 8, further comprising a second input subunit configured to obtain the second response data by inputting the second input data into the deep learning model.

10. The second input subunit is, The apparatus according to claim 9, configured to determine the second input data based on the first query data, the first reconsideration result, and task description information indicating that the second input data includes the first reconsideration result.

11. The aforementioned confirmation unit is The apparatus according to any one of claims 8 to 10, configured to determine a first reconsideration result for the first response data based on the first response data and the first feedback, in response to receiving first feedback from the user for the first response data, and in response to the determination that the first feedback is negative feedback.

12. The apparatus according to any one of claims 8 to 10, further comprising a third generation unit configured to generate a third response data for the second query data based on the first query data, the second response data, and the second query data, in response to the determination that the similarity between a second query data from a user and the first query data is greater than a preset threshold.

13. The system further includes a storage unit configured to store the first query data and the second response data in a memory bank, The third generation unit is An acquisition subunit configured to acquire the second response data from the memory bank in response to the determination that the similarity between the second query data from the user and the first query data in the memory bank is greater than the preset threshold, The apparatus according to claim 12, comprising the first query data, the second response data, and a generating subunit configured to generate the third response data based on the second query data.

14. The apparatus according to any one of claims 8 to 10, further comprising the first reconsideration result, an optimization policy for the first response data.

15. It is an electronic device, At least one processor, Includes memory communicated to at least one processor, An electronic device wherein the memory stores instructions that can be executed by the at least one processor, and by executing the instructions by the at least one processor, the at least one processor can be made to perform the method according to claim 1.

16. A non-temporary computer-readable storage medium in which computer instructions are stored, wherein the computer instructions are used to cause a computer to perform the method according to claim 1.

17. A computer program, which, when executed by a processor, implements the method described in claim 1.

Images