Response data generation method and device and computer program product

Abstract

The invention provides a response data generation method and device, electronic equipment, a storage medium and a computer program product, relates to the technical field of artificial intelligence, in particular to the technical field of natural language understanding and intelligent medical treatment, and can be applied to an intelligent dialogue scene. According to the specific implementation scheme, the method comprises the following steps: determining a prediction level of an uncertainty result possibly brought by responding to to-be-responded data; determining target knowledge data related to the to-be-answered data from a target medical knowledge base corresponding to the prediction level; according to the target knowledge data, medical response data for the to-be-responded data is generated through the selected target agent in the multiple agents, and different agents have different response modes. According to the method, the reliability and the accuracy of the target knowledge data are improved, and on the basis of ensuring the accuracy of the generated medical response data and the adaptation degree between the response mode and the user, the efficiency and the convenience of acquiring the medical related information by the user are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to the field of natural language understanding and smart medical technology, and in particular to a method, device, electronic device, storage medium and computer program product for generating response data, which can be applied in intelligent dialogue scenarios. Background Art

[0002] With the development of mobile Internet, users are increasingly relying on online resources to obtain medical and health information. However, the existing online information is not only of uneven quality, but also often cannot fully meet the needs of users in terms of relevance and accuracy. Users often need to conduct multiple searches to find satisfactory medical answers, which is not only time-consuming but also inefficient. Summary of the invention

[0003] The present disclosure provides a method, device, electronic device, storage medium and computer program product for generating response data.

[0004] According to the first aspect, a method for generating response data is provided, comprising: determining a prediction level of the uncertainty results that may be brought about by responding to data to be responded to; determining target knowledge data related to the data to be responded to from a target medical knowledge base corresponding to the prediction level; and generating medical response data for the data to be responded to according to the target knowledge data by a target intelligent agent selected from multiple intelligent agents, wherein different intelligent agents have different response modes.

[0005] According to the second aspect, a response data generating device is provided, comprising: a level determination unit, configured to determine the predicted level of the uncertainty results that may be brought about by responding to the data to be responded to; a knowledge data determination unit, configured to determine the target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the predicted level; a response data generating unit, configured to generate medical response data for the data to be responded to according to the target knowledge data through a target intelligent agent selected from multiple intelligent agents, wherein different intelligent agents have different response modes.

[0006] According to the third aspect, an intelligent agent is provided, comprising: an input module for receiving data to be responded to; a processing module for determining a medical response model for executing a response task for the data to be responded to, and obtaining medical response data for the data to be responded to by calling the medical response model to execute a method as described in any implementation method of the first aspect; and an output module for outputting the medical response data.

[0007] According to a fourth aspect, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected 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 so that the at least one processor can execute the method described in any implementation manner of the first aspect.

[0008] According to a fifth aspect, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method described in any implementation of the first aspect.

[0009] According to a sixth aspect, a computer program product is provided, comprising: a computer program, which implements the method described in any implementation manner of the first aspect when executed by a processor.

[0010] According to the technology disclosed in the present invention, a method and device for generating response data are provided, which determine the prediction level of the uncertainty results that may be brought about by responding to the data to be responded to, so as to determine the target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level, thereby improving the reliability and accuracy of the target knowledge data; through the target intelligent agent selected from multiple intelligent agents, medical response data for the data to be responded to is generated according to the target knowledge data, wherein different intelligent agents have different response modes, and on the basis of ensuring the accuracy of the generated medical response data and the adaptability between the response mode and the user, the efficiency and convenience of the user in obtaining medical related information are improved.

[0011] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0013] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present disclosure may be applied;

[0014] Figure 2 is a flowchart of an embodiment of a method for generating response data according to the present disclosure;

[0015] Figure 3 is an overall framework diagram of a method for generating response data according to the present disclosure;

[0016] Figure 4 is a schematic diagram of an application scenario of the method for generating response data according to this embodiment;

[0017] Figure 5 is a flowchart of another embodiment of a method for generating response data according to the present disclosure;

[0018] Figure 6 is a structural diagram of an embodiment of a device for generating response data according to the present disclosure;

[0019] Figure 7 It is a schematic diagram of the structure of a computer system suitable for implementing the embodiments of the present disclosure. DETAILED DESCRIPTION

[0020] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0021] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0022] Figure 1 An exemplary architecture 100 is shown to which the response data generation method and apparatus of the present disclosure can be applied.

[0023] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The communication connection between the terminal devices 101, 102, 103 constitutes a topological network, and the network 104 is used to provide a medium for the communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0024] Terminal devices 101, 102, 103 may be hardware devices or software that support network connection for data interaction and data processing. When terminal devices 101, 102, 103 are hardware, they may be various electronic devices that support network connection, information acquisition, interaction, display, processing and other functions, including but not limited to smart phones, tablet computers, e-book readers, laptop computers and desktop computers, etc. When terminal devices 101, 102, 103 are software, they may be installed in the electronic devices listed above. They may be implemented as multiple software or software modules for providing distributed services, or they may be implemented as a single software or software module. No specific limitation is made here.

[0025] The server 105 may be a server that provides various services, for example, for the data to be responded to sent by the user through the terminal devices 101, 102, and 103, the prediction level of the uncertainty result that may be brought about by responding to the data to be responded to is determined, so as to determine the target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level, and then generate a background processing server for medical response data for the data to be responded to according to the target knowledge data through the target intelligent agent. As an example, the server 105 may be a cloud server.

[0026] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, software or software modules used to provide distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0027] It should also be noted that the method for generating response data provided in the embodiments of the present disclosure is generally executed by a server, but the possibility of being executed by a terminal device, or being executed by the server and the terminal device in cooperation with each other is not excluded. Accordingly, the various parts (such as various units) included in the device for generating response data can be all set in the server, or all set in the terminal device, or can be set in the server and the terminal device respectively.

[0028] It should be understood that Figure 1 The number of terminal devices, networks, and servers in the system architecture is only illustrative. Depending on the implementation requirements, there may be any number of terminal devices, networks, and servers. When the electronic device on which the method for generating response data runs does not need to perform data transmission with other electronic devices, the system architecture may only include the electronic device (e.g., terminal device or server) on which the method for generating response data runs.

[0029] Please refer to Figure 2 and Figure 3 , Figure 2 A flowchart of a method for generating response data provided by an embodiment of the present disclosure, Figure 3 The overall framework diagram of a method for generating response data provided by an embodiment of the present disclosure. In the process 200, the following steps are included:

[0030] Step 201, determining the prediction level of the uncertainty result that may be brought about by responding to the data to be responded.

[0031] In this embodiment, the execution subject of the method for generating response data (for example, Figure 1The server in the communication network can obtain the data to be responded to remotely or locally through a wired network connection or a wireless network connection, and determine the prediction level of the uncertainty result that may be caused by responding to the data to be responded to.

[0032] The data to be answered are generally questions related to medicine (or medical treatment) generated by users, and the users are, for example, patients or doctors.

[0033] The uncertain results that may result from responding to pending data mainly refer to the potential hazards or adverse consequences that may be caused when providing medical advice or information.

[0034] Taking the user's medication problem as an example, these risks may include the following aspects:

[0035] 1. Medication Errors:

[0036] Dosage Error: Providing an incorrect dose of medication, which may result in overdose or underdose.

[0037] Frequency Error: Providing an incorrect medication frequency may result in less effective medication or increased side effects.

[0038] Incorrect usage: Incorrect medication methods provided, such as oral, topical, injection, etc., may result in the drug not being able to exert its intended effect.

[0039] 2. Drug interactions:

[0040] Interaction with other drugs: Failure to take into account other drugs that the user is taking may lead to drug interactions and adverse reactions.

[0041] Interactions with food: Failure to consider that certain foods may affect the absorption and metabolism of the drug, resulting in changes in drug efficacy.

[0042] 3. Allergic reaction:

[0043] Known Allergies: Failure to ask or consider the user's allergy history may result in the user experiencing a severe allergic reaction.

[0044] Unknown allergic reaction: The user may be allergic to a drug ingredient, but this has not been discovered before, and the agent fails to adequately indicate this possibility.

[0045] 4. Contraindications:

[0046] Disease contraindications: Certain medications are contraindicated for patients with specific diseases, and the agent fails to identify and alert the user.

[0047] Contraindications for special groups: Pregnant women, breastfeeding women, children, the elderly and other special groups may not be suitable for the use of certain drugs, and the intelligent body fails to fully remind them.

[0048] 5. Misleading information:

[0049] Inaccurate information: Inaccurate or outdated information about drugs provided may lead to user misunderstanding or misuse.

[0050] Incomplete information: Incomplete information about a drug is provided and important precautions or warnings are omitted.

[0051] In this embodiment, the classification standard and the number of classifications of the prediction levels can be flexibly determined according to actual conditions, for example, including a low prediction level, a medium prediction level, and a high prediction level.

[0052] As an example, the above-mentioned execution entity can determine the correspondence between each pending response data and the prediction level based on the big data analysis of the user's pending response data, and establish a correspondence table; further, based on the correspondence table, the prediction level of the uncertain results that may be brought about by responding to the pending response data can be determined.

[0053] As another example, the above-mentioned execution entity may first perform natural language understanding on the data to be responded to or the conversation data up to the data to be responded to through a large language model to obtain a semantic understanding result; then, based on the semantic understanding result, determine the prediction level of the uncertainty result that may be brought about by responding to the data to be responded to.

[0054] Step 202: Determine target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level.

[0055] In this embodiment, the execution subject can determine the target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level. Different medical knowledge bases are pre-set for different prediction levels.

[0056] As an example, the execution subject determines the correlation or similarity between the data to be answered and each part of the knowledge data in the target medical knowledge base, and then recalls a preset number of target knowledge data with a greater correlation or similarity to the data to be answered from the target medical knowledge base. The preset number can be specifically set according to actual conditions and is not limited here.

[0057] As another example, the execution subject determines keywords in the data to be responded to, and then determines partial knowledge data including at least one keyword from the target knowledge data, and then integrates the partial knowledge data corresponding to each keyword to obtain the target knowledge data.

[0058] In some optional implementations of this embodiment, the execution subject may perform step 202 in the following manner:

[0059] The first step is to adjust the data to be responded to according to the conversation data before the data to be responded to, so as to obtain the adjusted data to be responded to.

[0060] In the process of multiple rounds of dialogue, there are cases where users briefly express the data to be answered based on the context. If the answer is only based on the current data to be answered, the generated medical answer data may be inaccurate.

[0061] In this implementation, the agent has understood the conversation data before the data to be responded to based on the natural language understanding technology during the multiple rounds of conversations with the user. According to the understanding results of the conversation data before the data to be responded to, the description of the data to be responded to is adjusted or the brief content of the data to be responded to is supplemented to obtain the adjusted data to be responded to.

[0062] The second step is to determine target knowledge data related to the adjusted data to be responded to from the target medical knowledge base.

[0063] In this implementation, the target knowledge data related to the adjusted data to be responded to can be determined from the target medical knowledge base in a similar manner as described above, and the target knowledge data related to the data to be responded to can be determined from the target medical knowledge base, which will not be repeated here.

[0064] In this implementation, based on the contextual understanding of multiple rounds of dialogues, the data to be responded to is adjusted to obtain complete and accurate adjusted data to be responded to, which helps to improve the accuracy of the target knowledge data obtained based on the adjusted data to be responded to.

[0065] In some optional implementations of this embodiment, before executing the second step, the execution subject may also perform the following operation: determining a target medical knowledge base corresponding to the prediction level from multiple medical knowledge bases corresponding to the target intelligent agent.

[0066] In this implementation, each of the multiple intelligent agents is provided with multiple medical knowledge bases, and the multiple medical knowledge bases correspond one-to-one to the multiple prediction levels.

[0067] In this implementation method, first, the data to be answered is adjusted according to the conversation data before the data to be answered, so as to obtain the adjusted data to be answered; then, the target medical knowledge base corresponding to the prediction level is determined from the multiple medical knowledge bases corresponding to the target intelligent agent; finally, the target knowledge data related to the adjusted data to be answered is determined from the target medical knowledge base.

[0068] In this implementation, each intelligent agent is provided with multiple medical knowledge bases corresponding to multiple prediction levels, so as to respond to dialogue scenarios of different prediction levels in a targeted manner through different medical knowledge bases, which helps to reduce the risks that may be brought about by responding to pending data and improve the security of multi-round dialogues.

[0069] In some optional implementations of this embodiment, the execution subject may perform the second step in the following manner:

[0070] First, initial target knowledge data related to the adjusted data to be responded to is determined from the target medical knowledge base.

[0071] As an example, for each type of data source corresponding to the target medical knowledge base, a preset number of knowledge data related to the adjusted data to be responded to is determined from the medical knowledge data under this type of data source in the target medical knowledge base, and the determined knowledge data under different types of data sources are combined to obtain the initial knowledge data.

[0072] Then, according to the correlation between the initial target knowledge data and the data to be answered, as well as the accuracy of the initial target knowledge data, the initial target data is sorted to obtain the target knowledge data.

[0073] In this implementation, the accuracy of the knowledge data can be determined by the type of data source to which the knowledge data belongs. For example, the accuracy of the knowledge data in medical canons is generally higher than the accuracy of the concise content compiled by doctors in the field of medicine.

[0074] It can be understood that although the amount of recalled knowledge data is large, the knowledge data based on which the intelligent agent generates medical response data may only be a few of the top-ranked and more important ones. Therefore, the order of the knowledge data in the initial target knowledge data is more important.

[0075] As an example, for a few knowledge data ranked at the top in the sorting method based only on relevance from large to small, their sorting can be adjusted through the following table, referring to the relevance between the knowledge data and the data to be answered, as well as the accuracy of the knowledge data.

[0076]

[0077] Take "Yidian>=0.5top2rank1" as an example to explain as follows:

[0078] When a piece of knowledge data recalled from the medical dictionary has a correlation score greater than or equal to 0.5 with the data to be answered, and is ranked second (top 2) in the ranking based on the correlation score, it is adjusted to the first place (rank 1).

[0079] In this implementation, the correlation between the initial target knowledge data and the data to be responded to, as well as the accuracy of the initial target knowledge data, are taken into account to sort the initial target data, especially the few knowledge data that are ranked first, which improves the rationality and accuracy of the sorting of the target knowledge data, and helps to improve the information acquisition efficiency and generation efficiency in the process of the intelligent agent generating medical response data based on the target knowledge data.

[0080] Step 203 , generating medical response data for the data to be responded to according to the target knowledge data by means of a target intelligent agent selected from the multiple intelligent agents.

[0081] In this embodiment, the execution subject can generate medical response data for the data to be responded to according to the target knowledge data through the target agent selected from the multiple agents. Different agents have different response modes. The multiple agents can be network models with the same network model structure or different network model structures, such as ERNIE-speed (lightweight large model).

[0082] As an example, multiple agents with different response modes include:

[0083] 1. Digital doctor agent: As a virtual digital person of a doctor, it is good at answering questions from professional departments.

[0084] Specifically, the response mode of the digital doctor agent is:

[0085] Professional and Rigorous: Use medical terminology and provide detailed scientific explanations to ensure the accuracy and authority of information.

[0086] In-depth analysis: In-depth exploration of medical concepts and treatment methods, suitable for users who need a deeper understanding.

[0087] 2. Health Assistant Intelligent Body: Authoritative and serious popular science, and hopes to enhance the reading experience through actual cases and pictures.

[0088] Specifically, the response mode of the health assistant agent is:

[0089] Authoritative science popularization: Provide verified health information to ensure the accuracy and reliability of the content.

[0090] Vivid and easy to understand: Explain complex medical concepts in easy-to-understand language, and improve users' understanding and interest through real cases and pictures.

[0091] 3. Health Assistant (Senior Edition) Intelligent Agent: It can generate colloquial and anthropomorphic content through the voice interaction mode of making phone calls, highlighting the emotional interaction with users.

[0092] Specifically, the response mode of the health assistant (senior version) agent is:

[0093] Friendly speaking: Use everyday spoken language and avoid complex vocabulary and sentence structures to make it easy for the elderly to understand.

[0094] Warm interaction: Answer questions in the tone of an authoritative and professional doctor, being concise, warm and positive, increasing interactivity in the conversation and narrowing the distance with users.

[0095] As an example, the above-mentioned execution entity can display multiple agents to the user, and under the user's selection instruction (for example, a click operation on a virtual button corresponding to a certain agent), use the agent selected by the user as the target agent; further, use the target knowledge data as the prompt of the agent to generate medical response data for the data to be responded to.

[0096] As another example, the above-mentioned execution entity can receive the user's prompt in text or voice form such as "Please act as an agent with XX response style (for example, a health assistant agent)" through the large language model, and the large language model recalls the corresponding target agent based on the prompt.

[0097] Among them, the intelligent agent can be trained in the following ways:

[0098] First, obtain the training samples corresponding to the agent's desired mode.

[0099] For example, by collecting conversation data from real online users and performing processes such as manual verification, response mode adjustment, and labeling, we can obtain training samples corresponding to the agent.

[0100] Then, a machine learning algorithm is used, with the training data in the training samples as input and the labels in the training samples as expected output, to train the intelligent agent.

[0101] During the training process, the agent's iteration directions include:

[0102] a) Enhanced understanding ability: Adding multiple rounds of historical dialogue information to improve context understanding ability;

[0103] b) Answer structure optimization: highlight the core answers and improve the rationality of medical answer data segmentation;

[0104] c) Answer strategy improvement: Instead of the agent completely refusing to answer, we can use the enhanced model capabilities to provide indirect answers and expand the sources of knowledge evidence.

[0105] In addition to the above training methods, sample-side optimization and model-side optimization can also be performed.

[0106] The main purpose of sample optimization is to achieve a balance between sample content and style, as well as diverse styles of speech. For agents with different response modes, based on manual adjustments to the collected historical dialogue data, training samples with corresponding content and style are set for the agents, and the samples are guaranteed to include diverse styles of speech, so as to improve the diversity and flexibility of the model's answers.

[0107] For model-side optimization, it can be achieved through vertical data enhancement, DPO (Direct Preference Optimization) preference training, data flywheel, etc.

[0108] Data augmentation is an effective way to solve the overfitting problem. It enhances the information extracted from the original data set so that the enhanced data set can more comprehensively represent the data set, thereby narrowing the gap between the training set and the validation set. Vertical data augmentation refers to enhancing data for specific fields (such as medical conversations) or tasks to better meet the needs of model training. For example, in natural language processing tasks, text can be enhanced through synonym replacement, sentence transformation, and so on.

[0109] DPO is a direct preference optimization method, which is mainly used in the post-training stage to achieve precise control of the output of large models by directly optimizing the language model. DPO can accurately judge and learn the user's preferences without reinforcement learning, and the effect is more prominent. Its working principle is to increase the log probability of preference samples and reduce the log probability of non-preferential sample responses, thereby optimizing the model output. DPO relies on theoretical preference models (such as the Bradley-Terry model) to measure the degree of alignment between the reward function and the empirical preference data, and directly defines the preference loss based on the strategy, without explicitly learning the reward function or sampling from the strategy during training. This method makes DPO more direct and efficient when optimizing the model.

[0110] In the context of model-side optimization, the data flywheel can be seen as a process of continuous iteration and optimization. By continuously collecting, analyzing, feeding back, and optimizing data, a self-reinforcing positive feedback loop is formed, thereby continuously improving the performance, user experience, and business value of the model.

[0111] In some optional implementations of this embodiment, the execution subject may perform step 203 in the following manner:

[0112] Through the target intelligent agent, according to the target knowledge data and the preset response data organization method, medical response data for the data to be responded is generated.

[0113] The response data organization method includes at least one of the following:

[0114] 1. The classification response method for various types of answers included in the medical response data can also be described as a "classification aggregation method".

[0115] 2. The way of giving an example response to one type of answer among multiple types can also be described as "illustrative method".

[0116] 3. The complete response method for medical response data including one type of answer can also be described as a "complete aggregation method".

[0117] The following example illustrates the above response data organization method:

[0118] When users ask: How to take ibuprofen?

[0119] It is understandable that there are many particle sizes related to ibuprofen, such as ibuprofen tablets, ibuprofen sustained-release capsules, etc., and the tablet specifications of ibuprofen tablets are also different, so we cannot simply answer "one tablet at a time." Therefore, we use different wordings to meet safety requirements and also consider the practicality of users.

[0120] Classification aggregation method:

[0121] Q: How to use and how much ibuprofen should I take?

[0122] A: The usage and dosage of ibuprofen vary depending on the dosage form and specifications. The common dosage forms are as follows:

[0123] tablet:……

[0124] Injection: ...

[0125] You can also choose the specific dosage form of the drug so that I can give you a more accurate answer. Please note that the usage and dosage of the drug needs to follow the advice of professionals~

[0126] Example

[0127] Q: How to use and how much ibuprofen should I take?

[0128] A: The usage and dosage of ibuprofen vary depending on the dosage form and specifications. Take 0.5g of Merlin Ibuprofen Tablets as an example:

[0129] …

[0130] You can also choose the specific dosage form of the drug so that I can give you a more accurate answer. Please note that the usage and dosage of the drug needs to follow the advice of professionals~

[0131] Complete aggregation method: The barcode contents of the entities listed by the user are the same: Generally speaking, drugs such as targeted drugs that correspond to unique barcodes, or Chinese patent medicines, are simple and consistent in usage

[0132] Q: What is the dosage of gefitinib?

[0133] A: The dosage of gefitinib is usually:

[0134] Take 1 tablet at a time, once a day.

[0135] Please note that your question involves the usage and dosage of drugs. It is recommended that you seek medical advice offline in a timely manner or get help through online consultation.

[0136] In this implementation, the target intelligent agent can select the corresponding response data organization method according to the recalled target knowledge data under its specific response mode, and generate medical response data for the data to be responded to according to the response mode and the response data organization method, thereby further improving the security and accuracy of the response data on the basis of flexible and changeable answer styles.

[0137] Continue to see Figure 4 , Figure 4 400 is a schematic diagram of an application scenario of the method for generating response data according to the present embodiment. The server 401 displays a selection interface among multiple agents to the user through the terminal device 402, wherein different agents have different response modes, including a professional and rigorous digital doctor agent with in-depth analysis, an authoritative popular science agent with vivid and easy-to-understand explanations, and a health assistant agent (senior version) with friendly spoken language and warm voice interaction. The user 403 operates the terminal device 402 to select a target agent from multiple agents.

[0138] During each round of dialogue between the target intelligent agent and the user, first, the target intelligent agent obtains the user's data to be responded to and determines the prediction level of the uncertainty results that may be brought about by responding to the data to be responded to; then, the target knowledge data related to the data to be responded to is determined from the target medical knowledge base corresponding to the prediction level; finally, the target intelligent agent generates medical response data for the data to be responded to based on the target knowledge data.

[0139] In the present embodiment, a method for generating response data is provided to determine the prediction level of the uncertainty results that may be brought about by responding to the data to be responded to, so as to determine the target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level, thereby improving the reliability and accuracy of the target knowledge data; through the target intelligent agent selected from multiple intelligent agents, medical response data for the data to be responded to is generated according to the target knowledge data, wherein different intelligent agents have different response modes, and on the basis of ensuring the accuracy of the generated medical response data, the efficiency and convenience of users in obtaining medical related information are improved.

[0140] In some optional implementations of this embodiment, the above-mentioned execution subject may generate the target medical knowledge base in the following manner:

[0141] First, medical knowledge data corresponding to the prediction level is obtained from multiple types of data sources corresponding to the target intelligent agent.

[0142] Continuing with the example of the intelligent agents including the digital doctor agent, the health assistant agent and the health assistant (senior version) agent, the multiple types of data sources corresponding to the digital doctor agent include, for example, authoritative content written by doctors, books, academic papers, etc.; the multiple types of data sources corresponding to the health assistant agent and the health assistant (senior version) agent include, for example, medical classics, medical-related concise content, medical books, medical guidelines, etc.

[0143] When the prediction level is high, it is necessary to focus on the accuracy of the medical knowledge data in the data source.

[0144] Then, the data parts in the medical knowledge data are combined according to their themes to obtain multiple groups of grouped data.

[0145] As an example, a topic can be represented by a first-level title. The above execution subject can determine the intention of each first-level title in the medical knowledge data; then cluster part of the data under each first-level title according to the intention, and obtain multiple intentions. Figure 1 One corresponding to multiple groups of grouped data.

[0146] Then, according to the content expressed by the grouped data at different levels, the grouped data is divided level by level to obtain a tree structure.

[0147] In this implementation, for each grouped data, the intention of the first-level title corresponding to the grouped data is used as the root node, and the data of different levels under the root node are divided to obtain intermediate nodes until a leaf node is obtained.

[0148] When the length of the knowledge data under a leaf node is moderate, the title of the leaf node is supplemented according to the subject or intention of the span data under the leaf node, and is independently cut into a data block; when the length of the knowledge data under a leaf node is too long, it needs to be further split, and the number of characters in the data block is not greater than the preset threshold max_token; when the length of the knowledge data under a leaf node is too short, the knowledge data under it needs to be supplemented, and it needs to be supplemented by the knowledge data under other nodes, and the priority of supplementation is self>ancestor>peer>parent_peer.

[0149] Finally, the target medical knowledge base is created based on the node name data, partial medical knowledge data and sample pending response data corresponding to the leaf nodes in the tree structure.

[0150] The node name data corresponding to the leaf node can be obtained by connecting the node path from the root node to the leaf node through “_”. Part of the medical knowledge data is the medical knowledge under the leaf node, which can be used as the data basis for the response sample to be responded to.

[0151] In this embodiment, a 256-dimensional feature vector data embedding may be calculated based on the query (sample data to be answered), title (node ​​name data), and content (partial medical knowledge data) corresponding to the leaf node.

[0152] In this implementation, a method for creating a target medical knowledge base is provided. The target medical knowledge base includes node name data corresponding to leaf nodes, partial medical knowledge data and sample pending response data, which helps to improve the efficiency of determining target knowledge data from the target medical knowledge base.

[0153] In some optional implementations of this embodiment, the above-mentioned execution entity can perform the above-mentioned second step in the following manner: determine the target knowledge data from the target medical knowledge base based on the correlation between the adjusted data to be responded to and the sample data to be responded to in the target medical knowledge base.

[0154] As an example, the above-mentioned execution entity can perform feature extraction on the data to be responded to, determine its feature data embedding, and then determine the distance between the embedding of the data to be responded to and the embedding in the target medical knowledge base (more specifically, it can be the partial embedding corresponding to the sample data to be responded to) to determine the correlation between the two, thereby determining the target knowledge data related to the data to be responded to from the target medical knowledge base.

[0155] In this implementation, based on the data structure of node name data, partial medical knowledge data and sample pending response data in the target medical knowledge base, the difficulty of calculating the correlation is reduced, which helps to further improve the efficiency of determining the target knowledge data.

[0156] In some optional implementations of this embodiment, the above-mentioned execution subject may also perform the following operations: in response to the target medical knowledge base not including the target knowledge data, generating fallback response data for the data to be responded to through the fallback model.

[0157] In the case where the target knowledge data is not included in the target medical knowledge base, or the target knowledge data is not recalled from the target medical knowledge base, the above-mentioned execution subject can use a fallback model with stronger computing and natural language understanding capabilities to generate fallback response data for the data to be responded to based on specific words, for example:

[0158] A: I'm sorry, I can't find any content related to this question. The use of medicines varies from person to person, please follow the advice of professional doctors.

[0159] You can also continue to ask me other questions about medicines, I will be happy to answer them for you~

[0160] In this implementation, the bottom-line operation based on the bottom-line model can improve the response capability for the data to be responded to.

[0161] Continue to refer Figure 5 , shows a schematic process 500 of another embodiment of a method for generating response data according to the present disclosure. In process 500, the following steps are included:

[0162] Step 501, determining the prediction level of the uncertainty result that may be brought about by responding to the data to be responded.

[0163] Step 502: adjusting the data to be responded to according to the conversation data before the data to be responded to, to obtain adjusted data to be responded to.

[0164] Step 503, determining a target medical knowledge base corresponding to the prediction level from multiple medical knowledge bases corresponding to the target intelligent agent.

[0165] Step 504: determine initial target knowledge data related to the adjusted data to be responded to from the target medical knowledge base.

[0166] Step 505 , sorting the initial target data according to the correlation between the initial target knowledge data and the data to be responded to, and the accuracy of the initial target knowledge data, to obtain the target knowledge data.

[0167] Step 506: Generate medical response data for the data to be responded to based on the target knowledge data by using the target intelligent agent selected from the multiple intelligent agents.

[0168] Among them, different agents have different response modes.

[0169] It can be seen from this embodiment that Figure 2 Compared with the corresponding embodiments, process 500 of the method for generating response data in this embodiment specifically illustrates the recall process of target knowledge data, further improves the accuracy and security of the target knowledge data, and helps to improve the security and accuracy of the multi-round dialogue process.

[0170] Continue to refer Figure 6 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a device for generating response data. The system embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the system can be specifically applied to various electronic devices.

[0171] like Figure 6 As shown, the response data generating device 600 includes: a level determination unit 601, configured to determine the prediction level of the uncertainty result that may be brought about by responding to the data to be responded; a knowledge data determination unit 602, configured to determine the target knowledge data related to the data to be responded from the target medical knowledge base corresponding to the prediction level; a response data generating unit 603, configured to generate medical response data for the data to be responded according to the target knowledge data through a target intelligent agent selected from multiple intelligent agents, wherein different intelligent agents have different response modes.

[0172] In some optional implementations of this embodiment, the knowledge data determination unit 602 is further configured to: adjust the data to be responded to according to the conversation data before the data to be responded to obtain the adjusted data to be responded to; and determine the target knowledge data related to the adjusted data to be responded to from the target medical knowledge base.

[0173] In some optional implementations of this embodiment, the above-mentioned device also includes: a knowledge base determination unit (not shown in the figure), which is configured to: determine the target medical knowledge base corresponding to the prediction level from multiple medical knowledge bases corresponding to the target intelligent body.

[0174] In some optional implementations of this embodiment, the knowledge data determination unit 602 is further configured to: determine initial target knowledge data related to the adjusted data to be responded to from the target medical knowledge base; sort the initial target data according to the correlation between the initial target knowledge data and the data to be responded to, and the accuracy of the initial target knowledge data, to obtain the target knowledge data.

[0175] In some optional implementations of the present embodiment, the above-mentioned device also includes: a knowledge base creation unit (not shown in the figure), which is configured to: obtain medical knowledge data corresponding to the prediction level from multiple types of data sources corresponding to the target intelligent agent respectively; combine each part of the data in the medical knowledge data according to the theme of each part of the data to obtain multiple groups of grouped data; divide the grouped data level by level according to the content expressed by the grouped data at different levels to obtain a tree structure; create a target medical knowledge base according to the node name data, partial medical knowledge data and sample pending response data corresponding to the leaf nodes in the tree structure.

[0176] In some optional implementations of this embodiment, the knowledge data determination unit 602 is further configured to: determine the target knowledge data from the target medical knowledge base according to the correlation between the adjusted data to be responded to and the sample data to be responded to in the target medical knowledge base.

[0177] In some optional implementations of the present embodiment, the response data generation unit 603 is further configured to: generate medical response data for the data to be responded to according to the target knowledge data and in accordance with a preset response data organization method through the target intelligent agent, wherein the response data organization method includes at least one of the following: a classified response method for multiple types of answers included in the medical response data; an example response method for one type of answer among multiple types; a complete response method for one type of answer included in the medical response data.

[0178] In some optional implementations of this embodiment, the above-mentioned device also includes: a fallback unit (not shown in the figure), which is configured to: in response to the target medical knowledge base not including the target knowledge data, generate fallback response data for the data to be responded to through a fallback model.

[0179] In the present embodiment, a device for generating response data is provided to determine the prediction level of the uncertainty results that may be brought about by responding to the data to be responded to, so as to determine the target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level, thereby improving the reliability and accuracy of the target knowledge data; through the target intelligent agent selected from multiple intelligent agents, medical response data for the data to be responded to is generated according to the target knowledge data, wherein different intelligent agents have different response modes, and on the basis of ensuring the accuracy of the generated medical response data, the efficiency and convenience of users in obtaining medical related information are improved.

[0180] According to an embodiment of the present disclosure, the present disclosure also provides an intelligent agent, including: an input module for receiving data to be responded to; a processing module for determining a medical response model for executing a response task for the data to be responded to, and obtaining medical response data for the data to be responded to by calling the medical response model to execute the response data generation method described in any of the above embodiments; and an output module for outputting the medical response data.

[0181] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, which includes: at least one processor; and a memory communicatively connected 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 so that the at least one processor can implement the method for generating response data described in any of the above embodiments when executing.

[0182] According to an embodiment of the present disclosure, the present disclosure further provides a readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to implement the method for generating response data described in any of the above embodiments when executed.

[0183] An embodiment of the present disclosure provides a computer program product. When the computer program is executed by a processor, the method for generating response data described in any of the above embodiments can be implemented.

[0184] Figure 7 A schematic block diagram of an example electronic device 700 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0185] like Figure 7 As shown, the device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0186] A number of components in the device 700 are connected to the I / O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0187] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 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 running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 701 performs the various methods and processes described above, such as a method for generating response data. For example, in some embodiments, the method for generating response data may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 700 via ROM 702 and / or a communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of the method for generating response data described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform a method for generating response data in any other appropriate manner (e.g., by means of firmware).

[0188] Various implementations of the systems and techniques described above 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), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can 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.

[0189] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable response data generating device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine, or completely on a remote machine or server.

[0190] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. 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, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0191] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0192] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communications network). Examples of communications networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0193] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and virtual private servers (VPS) services; it may also be a server for a distributed system, or a server combined with a blockchain.

[0194] According to the technical solution of the embodiment of the present disclosure, a method and device for generating response data are provided, which determine the prediction level of the uncertainty results that may be brought about by responding to the data to be responded to, so as to determine the target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level, thereby improving the reliability and accuracy of the target knowledge data; through the target intelligent agent selected from multiple intelligent agents, medical response data for the data to be responded to is generated according to the target knowledge data, wherein different intelligent agents have different response modes, and on the basis of ensuring the accuracy of the generated medical response data, the efficiency and convenience of users in obtaining medical related information are improved.

[0195] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution provided by this disclosure can be achieved, and this document does not limit this.

[0196] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for generating response data, comprising: Determine the prediction level of the uncertainty results that may be brought about by responding to the pending data; Determining target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level; A target intelligent agent selected from a plurality of intelligent agents generates medical response data for the data to be responded to according to the target knowledge data, wherein different intelligent agents have different response modes.

2. The method according to claim 1, wherein: The step of determining target knowledge data related to the data to be responded to from the target medical knowledge base corresponding to the prediction level includes: Adjusting the data to be responded to according to the conversation data before the data to be responded to, to obtain adjusted data to be responded to; Target knowledge data related to the adjusted data to be responded to is determined from the target medical knowledge base.

3. The method according to claim 2, wherein: Before determining the target knowledge data related to the adjusted data to be responded to from the target medical knowledge base, the method further includes: From the multiple medical knowledge bases corresponding to the target intelligent agent, determine the target medical knowledge base corresponding to the prediction level.

4. The method according to claim 2 or 3, wherein: The step of determining target knowledge data related to the adjusted data to be responded to from the target medical knowledge base includes: Determining initial target knowledge data related to the adjusted data to be responded to from the target medical knowledge base; According to the correlation between the initial target knowledge data and the data to be responded, and the accuracy of the initial target knowledge data, the initial target data is sorted to obtain the target knowledge data.

5. The method according to claim 3, wherein: Before determining the target medical knowledge base corresponding to the prediction level from the multiple medical knowledge bases corresponding to the target intelligent agent, the method further includes: Acquire medical knowledge data corresponding to the prediction level from multiple types of data sources corresponding to the target intelligent agent respectively; Combining the data parts according to the subject of each part of the medical knowledge data to obtain a plurality of grouped data; According to the contents expressed by the grouped data at different levels, the grouped data are divided level by level to obtain a tree structure; The target medical knowledge base is created according to the node name data, partial medical knowledge data and sample pending response data corresponding to the leaf nodes in the tree structure.

6. The method according to claim 5, wherein: The step of determining target knowledge data related to the adjusted data to be responded to from the target medical knowledge base includes: According to the correlation between the adjusted data to be responded to and the sample data to be responded to in the target medical knowledge base, the target knowledge data is determined from the target medical knowledge base.

7. The method according to claim 1, wherein: The target intelligent agent selected from the plurality of intelligent agents generates medical response data for the data to be responded to according to the target knowledge data, including: Generate medical response data for the data to be responded to by the target agent according to the target knowledge data and in accordance with a preset response data organization method, wherein the response data organization method includes at least one of the following: A classification response method for multiple types of answers included in the medical response data; An example response to an answer of one of the plurality of types; A complete response mode is provided for the medical response data including one type of answer.

8. The method according to claim 1, wherein: Also includes: In response to the target medical knowledge base not including the target knowledge data, a fallback response data for the data to be responded to is generated through a fallback model.

9. A device for generating response data, comprising: A level determination unit, configured to determine a predicted level of uncertainty results that may be brought about by responding to the data to be responded to; a knowledge data determination unit configured to determine target knowledge data related to the data to be responded to from a target medical knowledge base corresponding to the prediction level; The response data generating unit is configured to generate medical response data for the data to be responded to according to the target knowledge data through a target intelligent agent selected from a plurality of intelligent agents, wherein different intelligent agents have different response modes.

10. The device according to claim 9, wherein: The knowledge data determination unit is further configured to: The data to be responded to are adjusted according to the conversation data before the data to be responded to, so as to obtain the adjusted data to be responded to; and the target knowledge data related to the adjusted data to be responded to are determined from the target medical knowledge base.

11. The device according to claim 10, wherein: Also includes: The knowledge base determination unit is configured to determine the target medical knowledge base corresponding to the prediction level from the multiple medical knowledge bases corresponding to the target intelligent agent.

12. The device according to claim 10 or 11, wherein: The knowledge data determination unit is further configured to: Determine initial target knowledge data related to the adjusted data to be responded from the target medical knowledge base; sort the initial target data according to the correlation between the initial target knowledge data and the data to be responded, and the accuracy of the initial target knowledge data to obtain the target knowledge data.

13. The device according to claim 11, wherein: Also includes: The knowledge base creation unit is configured as follows: Respectively obtaining medical knowledge data corresponding to the prediction level from multiple types of data sources corresponding to the target intelligent agent; combining each part of the data in the medical knowledge data according to the theme of each part of the data to obtain multiple groups of grouped data; According to the contents expressed by the grouped data at different levels, the grouped data are divided level by level to obtain a tree structure; The target medical knowledge base is created according to the node name data, partial medical knowledge data and sample pending response data corresponding to the leaf nodes in the tree structure.

14. The device according to claim 13, wherein: The knowledge data determination unit is further configured to: According to the correlation between the adjusted data to be responded to and the sample data to be responded to in the target medical knowledge base, the target knowledge data is determined from the target medical knowledge base.

15. The device according to claim 9, wherein: The response data generating unit is further configured to: Generate medical response data for the data to be responded to by the target agent according to the target knowledge data and in accordance with a preset response data organization method, wherein the response data organization method includes at least one of the following: A classification response method for multiple types of answers included in the medical response data; An example response to an answer of one of the plurality of types; A complete response mode is provided for the medical response data including one type of answer.

16. The device according to claim 9, wherein: Also includes: The fallback unit is configured to: in response to the target medical knowledge base not including the target knowledge data, generate fallback response data for the data to be responded to through a fallback model.

17. An intelligent agent, comprising: An input module, used for receiving data to be responded; a processing module, configured to determine a medical response model for executing a response task for the data to be responded to, and to obtain medical response data for the data to be responded to by calling the medical response model to execute the method of any one of claims 1 to 8; An output module is used to output the medical response data.

18. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

19. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 8.

20. A computer program product comprising: A computer program which, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images