CN114300160A - Inquiry dialogue method and system - Google Patents
Inquiry dialogue method and system Download PDFInfo
- Publication number
- CN114300160A CN114300160A CN202111357645.7A CN202111357645A CN114300160A CN 114300160 A CN114300160 A CN 114300160A CN 202111357645 A CN202111357645 A CN 202111357645A CN 114300160 A CN114300160 A CN 114300160A
- Authority
- CN
- China
- Prior art keywords
- inquiry
- model
- information
- question
- dialogue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012549 training Methods 0.000 claims description 30
- 238000012545 processing Methods 0.000 claims description 9
- 230000004044 response Effects 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 4
- 230000035606 childbirth Effects 0.000 claims description 2
- 238000013135 deep learning Methods 0.000 claims description 2
- 230000005906 menstruation Effects 0.000 claims description 2
- 208000024891 symptom Diseases 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 206010019233 Headaches Diseases 0.000 description 3
- 231100000869 headache Toxicity 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 208000004998 Abdominal Pain Diseases 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides an inquiry dialogue method and system, belonging to the field of intelligent medical treatment. The method comprises the following steps: acquiring patient information, and generating an inquiry dialogue problem with a fixed pointing label according to the patient information and a preset model; pushing the inquiry dialogue questions to a user side; and recovering user answering information aiming at the inquiry dialogue questions, analyzing the user answering information based on the fixed pointing tags of the inquiry dialogue questions, and obtaining the demand information under the corresponding fixed pointing tags. The proposal of the invention adds a fixed pointing label to each generated inquiry dialogue problem, and provides an anthropomorphic, professional and controllable inquiry dialogue method.
Description
Technical Field
The invention relates to the field of intelligent medical treatment, in particular to an inquiry dialogue method and an inquiry dialogue system.
Background
The intelligent inquiry becomes the current common technical means for improving diagnosis and treatment efficiency, and the intelligent system collects the conditions of patients in an intelligent inquiry and answer mode, so that the workload of doctors is reduced, and the diagnosis and treatment efficiency of hospitals is improved. In the existing intelligent inquiry system, two modes mainly exist, one mode is mainly considering the spoken language of the intelligent question-answer question, namely the generated question-answer question is closer to the spoken text of a doctor. The method mainly utilized is that spoken questions are generated based on a PTM (packet transfer model) (pre-training) model, and when the PTM model is constructed, training is carried out according to the existing inquiry data or expert-labeled data, and inquiry and answer questions are generated by simulating the inquiry characteristics of doctors. When the PTM model is generated, the construction is mainly carried out based on a Transformer model, and the PTM model has obvious significance for accurate identification of semantics. The other method is a mode of embodying a serious scene, namely, the recognition result is more accurate, and the problem that the inquiry question recognition is inaccurate due to spoken semantic recognition so as to influence the condition that a doctor judges the state of an illness of a patient wrongly is avoided.
Both of the above methods are frequently used in the existing medical inquiry scenes, but both have certain disadvantages. Among them, the PTM model simulates the doctor's inquiry habit in generating an inquiry question, and thus makes the patient experience better, but it is not satisfactory in terms of conversation logic. I.e. the dialog may not be performed from a professional medical point of view, with the risk of medical imprecision and even deviation. The problem generation under the FSM rule is relatively clear based on the problem directivity, but the problem caused by the correspondence is insufficient openness. In the case of open input, intent recognition is a complex problem, FSM is difficult to cover all states in some scenarios, and the output is often template-based, with insufficient diversity. Aiming at the problems of the existing intelligent inquiry method, a new inquiry dialogue method needs to be created.
Disclosure of Invention
The invention aims to provide an inquiry dialogue method and system, which at least solve the problem that the existing intelligent inquiry method cannot be compatible with flexibility and execution.
In order to achieve the above object, a first aspect of the present invention provides an inquiry dialogue method, including: acquiring patient information, and generating an inquiry dialogue problem with a fixed pointing label according to the patient information and a preset model; pushing the inquiry dialogue questions to a user side; and recovering user answering information aiming at the inquiry dialogue questions, analyzing the user answering information based on the fixed pointing tags of the inquiry dialogue questions, and obtaining the demand information under the corresponding fixed pointing tags.
Optionally, the patient information includes one or more of the following information: the patient information includes one or more of the following: chief complaint information, registered department information, past history information, present history information, personal history information, family history information, and menstruation, marriage and childbirth history information.
Optionally, the preset model includes: PTM model, TAG model; wherein the PTM model is used for generating an inquiry dialogue question; the TAG model comprises: a TAG semantic model for tagging semantic types of the interrogation dialog problem; a TAG sequence model for defining a fixed orientation of the interrogation dialog problem.
Optionally, the method further includes: constructing a PTM model, comprising: acquiring historical inquiry data; wherein the historical interrogation data comprises one or more of: intelligent inquiry questions and answers, medical record sorting information and simulated inquiry and answer information; and taking the historical inquiry data as a training sample to train the PTM model.
Optionally, the method further includes: constructing a TAG semantic model, comprising: performing semantic annotation on the inquiry dialogue questions generated by the PTM model; classifying the inquiry dialogue questions according to the semantic labels, wherein each classification corresponds to a classification label; counting all the classification labels to form a label library; and training according to the corresponding relation between the inquiry dialogue question and the label library to obtain a TAG model.
Optionally, the method further includes: constructing a TAG sequence model, comprising: carrying out historical inquiry data pointing annotation according to a preset knowledge rule; and taking the historical inquiry data subjected to execution labeling as a training sample, and training based on a preset deep learning algorithm to obtain a TAG sequence model.
Optionally, the generating an inquiry dialogue question with a fixed pointing tag according to the patient information and a preset model includes: generating an inquiry dialogue question set according to the patient information and the PTM model; setting fixed pointing labels for all the inquiry dialogue questions in the inquiry dialogue question set according to the TAG semantic model; counting the number of inquiry dialogue problems under each fixed direction according to the TAG sequence model, and extracting the fixed direction with the largest number of inquiry dialogue problems; and selecting one inquiry dialogue question and the fixed pointing label as the inquiry dialogue question with the fixed pointing label under the extracted fixed pointing direction.
Optionally, the method further includes: training a PTM model, comprising: acquiring historical inquiry data; wherein the historical interrogation data comprises one or more of: intelligent inquiry questions and answers, medical record sorting information and simulated inquiry and answer information; performing semantic annotation on historical inquiry information in the historical inquiry data to respectively obtain inquiry questions in the historical inquiry information and corresponding inquiry question types; training according to each inquiry question and the corresponding inquiry question type to obtain a TAG model; and training according to the inquiry questions and the TGA model to obtain a PTM model.
Optionally, the generating an inquiry dialogue question with a fixed pointing tag according to the patient information and a preset model includes: and taking the patient information as input and the inquiry dialogue questions with the fixed pointing labels as output, and carrying out PTM model training to obtain the corresponding inquiry dialogue questions with the fixed pointing labels.
A second aspect of the invention provides an interrogation dialog system, the system comprising: the acquisition unit is used for acquiring patient information; the processing unit is used for generating an inquiry dialogue problem with a fixed pointing label according to the patient information and a preset model; the pushing unit is used for pushing the inquiry dialogue questions to the user side; the acquisition unit is further configured to: recovering user response information for the inquiry dialogue questions; the processing unit is further to: and analyzing the answering information of the user based on the fixed pointing label of the inquiry dialogue question to obtain the demand information under the corresponding fixed pointing label.
In another aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which when executed on a computer, cause the computer to perform the above-described inquiry dialogue method.
By the technical scheme, when the inquiry dialogue questions are generated, a fixed pointing label is added for each intelligent inquiry question, the inquiry purpose of the corresponding inquiry dialogue questions is specified, and the answering range of the patient is also limited. After the patient answers, the problem analysis is carried out based on the fixed direction, the logicality and the directivity of problem processing are improved, and the flexibility of problem generation is reserved.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
FIG. 1 is a flow chart illustrating the steps of an interrogation session method according to one embodiment of the present invention;
FIG. 2 is a flowchart of the steps provided by one embodiment of the present invention to generate an interrogation dialog problem with a fixed point tag;
fig. 3 is a system configuration diagram of an inquiry dialogue system according to an embodiment of the present invention.
Description of the reference numerals
10-an acquisition unit; 20-a processing unit; 30-pushing unit.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
Fig. 3 is a system configuration diagram of an inquiry dialogue system according to an embodiment of the present invention. As shown in fig. 3, an embodiment of the present invention provides an inquiry dialogue system, which includes: an acquisition unit 10 for acquiring patient information; a processing unit 20, configured to generate an inquiry dialogue question with a fixed orientation according to the patient information and a preset model; the pushing unit 30 is used for pushing the affiliated inquiry dialogue questions with fixed directions to the user side; the acquisition unit 10 is also used for recovering user response information; the processing unit 20 is further configured to analyze the user response information based on the fixed direction, and obtain demand information corresponding to the direction.
Fig. 1 is a flowchart of a method of an inquiry session in accordance with an embodiment of the present invention. As shown in fig. 1, an embodiment of the present invention provides an inquiry dialogue method, which includes:
step S10: acquiring patient information, and generating an inquiry dialogue problem with a fixed pointing label according to the patient information and a preset model.
Specifically, in the existing intelligent inquiry system, there are two main ways of generating intelligent questions, the first is mainly considering spoken language of the intelligent question-answer questions, i.e. the generated question-answer questions are closer to the spoken text of the doctor, which has the advantages that when the patient makes intelligent question-answer, the patient has the substitution feeling of dialogue with the doctor, and the participation feeling of the patient is greatly helped. The method mainly utilized is that spoken questions are generated based on a PTM (packet transfer model) (pre-training) model, and when the PTM model is constructed, training is carried out according to the existing inquiry data or expert-labeled data, and inquiry and answer questions are generated by simulating the inquiry characteristics of doctors. When the PTM model is generated, the construction is mainly carried out based on a Transformer model, and the PTM model has obvious significance for accurate identification of semantics. The other method is a mode of embodying a serious scene, namely, the recognition result is more accurate, and the problem that the inquiry question recognition is inaccurate due to spoken semantic recognition so as to influence the condition that a doctor judges the state of an illness of a patient wrongly is avoided.
Both of the above methods are frequently used in the existing medical inquiry scenes, but both have certain disadvantages. Among them, the PTM model simulates the doctor's inquiry habit in generating an inquiry question, and thus makes the patient experience better, but it is not satisfactory in terms of conversation logic. That is, the patient easily solves the question, and when answering, the patient inputs the content irrelevant to the question, and when performing semantic recognition, the final recognition content does not match the purpose. For example, the questions of the inquiry are: how long this condition has occurred? The patient answers: beginning two months ago, abdominal pain develops between them, perhaps two weeks ago, or more. When semantic identification is carried out, because of no logical association, the symptom time inquired before can not be accurately identified between two months, two weeks and longer, and the condition of a patient can not be accurately judged by a doctor. The problem generation under the FSM rule is relatively clear based on the problem directivity, but the problem caused by the correspondence is insufficient openness. In the case of open input, intent recognition is a complex problem, FSM is difficult to cover all states in some scenarios, and the output is often template-based, with insufficient diversity. Based on the respective advantages and disadvantages of the two methods, the method combines the two methods, performs advantage complementation, gets rid of the respective disadvantages, can keep spoken language expression of doctors under the condition of ensuring better directivity of inquiry problems in dialogue inquiry, and improves the participation degree of patients on the premise of ensuring accurate acquisition of the illness states of the patients.
Based on this, we need to obtain a problem generation model with specified performance, and in the generated model, the application provides two possible implementation methods, the first is that in the generation of the PTM model, the directivity of the problem is directly specified, that is, in the generation of the problem, the output result directly carries a fixed directional label. And the other method is to generate problems according to a conventional PTM model, analyze and count the generated problems, add labels and push the problems. Specifically, as shown in fig. 2, the method includes the following steps:
step S101: and constructing a preset model.
Specifically, the first embodiment:
before PTM model training, a large number of training samples are required to be collected, the training samples contain a large number of spoken inquiry rules, the features are extracted, and spoken question generation simulation can be carried out. Therefore, the historical data comprises past question and answer information, the question and answer process information is stored in a past medical record, a past intelligent question and answer system or the active compiling content of experts, the historical data comprises patient information, a question and a question, in brief, the PTM model is used for generating spoken question and question based on the patient information and the question, and based on the spoken question and question, the PTM model is obtained by taking the patient information and the question as input and taking the question as output to perform model training. And when the historical data is sorted, a part of the historical data is reserved as verification data, after the PTM model is recorded, the model test is carried out according to the verification data, and the accuracy of the model is judged.
Example two:
problem execution is directly specified when PTM model training is performed. For example, the final question is "[ ask time to occur ] how long did the headache? ", wherein the content in parentheses: asking how long the directional label is fixed, and how long the back headache is, the question content simulating the linguistics of the doctor. In the above embodiment, only how to generate a model simulating a spoken problem of a doctor is constructed, which does not limit the execution of the problem. In the embodiment of the present application, it is desirable to complete the label addition of the corresponding question in the process of constructing the PTM model. During model training, the classification of the interrogation executive type is made based on the purpose of the interrogation, e.g., the splitting includes: ask about the time of occurrence, ask about the trigger and ask about the symptoms. Then if the question is: recently did there be something special about eating? It will be classified into question incentive types. Based on the relationship, firstly, historical data labeling is carried out, namely, the type labeling is carried out on the inquiry questions. The marking process can be carried out based on the PTM model, after the data marking is finished, the PTM model training is carried out, the TAG model is directly added in the training process, and the PTM model containing the TAG model is generated according to the type marking and the spoken language problem. So that in the subsequent question generation process, the output question directly contains the inquiry dialogue question with the fixed pointing tag.
Step S102: patient information is acquired.
Specifically, the PTM model generates a corresponding question-and-call dialogue problem based on patient information, so that input data of the model is patient information, and the patient information includes chief complaint information, registered department information, past history information, and basic identity information. This information provides a basis for generating problems, and when patient information is acquired, it is preferably acquired by patient input information and retention information. For example, the past history information of the patient in the corresponding hospital or medical system can be directly used for problem generation according to the patient's needs. Preferably, when the patient is a follow-up patient, the patient can directly select the follow-up requirement, and the remaining information of the patient is directly read, so that the problem generation is carried out based on the remaining information. If the patient is a first-visit patient, whether the patient has other medical history remaining information in the system is judged, and when the problem is generated, the remaining information is used for reference, and the associated problem is generated according to the medical knowledge map, so that the intelligence of the system is improved.
Step S103: and generating an inquiry dialogue problem with a fixed pointing label according to the patient information and a preset model.
Specifically, two model generation methods are provided in step S101, and the generation problem is the same based on different PTM models. In the first embodiment, for the PTM model construction method, the corresponding method for generating an inquiry dialogue question with a fixed direction tag includes:
firstly, patient information is standardized, namely patient data is processed into corresponding input data according to the universality requirement of the PTM model, and then the input data is imported into the constructed PTM model. Based on the characteristics of the PTM model, because no label limitation exists, the output problem is multiple, and in order to accurately identify the problem needing to be pushed, all the generated problems are combined to form an inquiry dialogue problem set. And then the TAG model sets labels of all the inquiry dialogue questions in the inquiry dialogue question set, extracts the inquiry purpose of each inquiry question in the setting process, then performs label type matching based on the inquiry purpose, and performs corresponding question label adding according to the matched labels. The same type of question was formed as to how long the headache was. And counting the number of the problems under each label type, and then comparing to screen out the problem label with the largest number, wherein the problem label implies the problem inquiry logic, and the larger the number of the problems, the higher the weight of the corresponding problem. Because questions on the same label are intended identically, the timing presentation is different, for example, under the "ask symptom" label type, "where uncomfortable? "place with or without discomfort" all means the same, so in the type of the tag in which the number of questions is the largest, one question is selected as an inquiry dialogue question with a fixed direction, and the question is pushed. Firstly, setting inquiry dialogue problem labels through a TAG semantic model, then carrying out pointing statistics based on a TAG sequence model, and carrying out problem screening based on a statistical result.
For the PTM model construction rule in the second embodiment, because the TAG model is already fused when the PTM model is constructed, the problem classification is directly performed in the generation problem, and the problem generated is actually limited by the TAG rule, so the type of the generated TAG is unique, although there may be a plurality of expression ways, the increase of the screening process is avoided, only one problem with a fixed pointing TAG is generated when the problem generation is performed, and then the problem push is performed based on the problem. In the generation of the question, after the user data is input, the question is prespecified based on the TAG sequence model, i.e. its orientation is specified, and then in the subsequent PTM model, the question is generated based on the specified orientation. Since no restriction is specified in the real-time example one, the problem generation is still performed according to the knowledge graph, so that the directions and the number of output results are various, and in the example two, the PTM model training is performed. The actual input data is the execution of the patient information and the TAG sequence model specification, so the range of limitation is increased, model training is carried out based on the limitation, and the problem with a fixed pointing label is output.
Step S20: and pushing the inquiry dialogue questions to a user side.
Specifically, no matter which PTM model is, only one final inquiry dialogue question with fixed direction is available, the question has a fixed direction label and specific question content, and the push unit pushes the question to the device side of the user. And opening the corresponding question recovery box to provide an open answer range for the user. Because of having a fixed directional label, the patient has a smaller range of thinking when answering, and close to the self. After the user finishes answering the question, the user triggers the completion button, and the acquisition unit correspondingly recovers the answering information of the patient.
Step S30: and recovering user answering information aiming at the inquiry dialogue questions, and analyzing the user answering information based on the fixed pointing label to obtain the demand information under the corresponding label.
Specifically, after the user response information is recovered, the response information can be analyzed according to the pointing label carried by the user response information. For example, if the fixed directional label is "ask symptom", the answer information analysis is performed by performing semantic analysis according to the indicating map, screening out words related to the symptom, and directly filtering the rest unrelated words, so that the amount of data processing is reduced, the uniqueness of the analysis direction is ensured, and the interference of interference information on the final semantic recognition result is avoided. The identification result is closer to the requirement, and the information acquisition logic of the patient is more definite.
In the embodiment of the invention, the generation of spoken language of the inquiry dialogue problem is realized by combining the PTM model and the TAG model, and good directivity and logicality are kept at the same time. By combining the advantages of PTM model semantic understanding, dialogue fluency and FSM controllability, an anthropomorphic, professional and controllable inquiry dialogue system is obtained in the serious professional field of medical treatment.
Embodiments of the present invention also provide a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, cause the computer to perform the above-mentioned inquiry dialogue method.
Those skilled in the art will appreciate that all or part of the steps in the method for implementing the above embodiments may be implemented by a program, which is stored in a storage medium and includes several instructions to enable a single chip, a chip, or a processor (processor) to execute all or part of the steps in the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
While the embodiments of the present invention have been described in detail with reference to the accompanying drawings, the embodiments of the present invention are not limited to the details of the above embodiments, and various simple modifications can be made to the technical solution of the embodiments of the present invention within the technical idea of the embodiments of the present invention, and the simple modifications are within the scope of the embodiments of the present invention. It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, the embodiments of the present invention will not be described separately for the various possible combinations.
In addition, any combination of the various embodiments of the present invention is also possible, and the same should be considered as disclosed in the embodiments of the present invention as long as it does not depart from the spirit of the embodiments of the present invention.
Claims (11)
1. An interrogation dialog method, the method comprising:
acquiring patient information, and generating an inquiry dialogue problem with a fixed pointing label according to the patient information and a preset model;
pushing the inquiry dialogue questions to a user side;
and recovering user answering information aiming at the inquiry dialogue questions, analyzing the user answering information based on the fixed pointing tags of the inquiry dialogue questions, and obtaining the demand information under the corresponding fixed pointing tags.
2. The method of claim 1, wherein the patient information includes one or more of the following:
chief complaint information, registered department information, past history information, present history information, personal history information, family history information, and menstruation, marriage and childbirth history information.
3. The method of claim 1, wherein the pre-set model comprises:
PTM model, TAG model; wherein,
the PTM model is used for generating an inquiry dialogue question;
the TAG model comprises:
a TAG semantic model for tagging semantic types of the interrogation dialog problem;
a TAG sequence model for defining a fixed orientation of the interrogation dialog problem.
4. The method of claim 3, further comprising:
constructing a PTM model, comprising:
acquiring historical inquiry data; wherein the historical interrogation data comprises one or more of: intelligent inquiry questions and answers, medical record sorting information and simulated inquiry and answer information;
and taking the historical inquiry data as a training sample to train the PTM model.
5. The method of claim 4, further comprising:
constructing a TAG semantic model, comprising:
performing semantic annotation on the inquiry dialogue questions generated by the PTM model;
classifying the inquiry dialogue questions according to the semantic labels, wherein each classification corresponds to a classification label;
counting all the classification labels to form a label library;
and training according to the corresponding relation between the inquiry dialogue question and the label library to obtain a TAG semantic model.
6. The method of claim 5, further comprising:
constructing a TAG sequence model, comprising:
carrying out historical inquiry data pointing annotation according to a preset knowledge rule;
and taking the historical inquiry data subjected to execution labeling as a training sample, and training based on a preset deep learning algorithm to obtain a TAG sequence model.
7. The method of claim 6, wherein generating an interrogation dialog question with a fixed point tag based on the patient information and a preset model comprises:
generating an inquiry dialogue question set according to the patient information and the PTM model;
setting fixed pointing labels for all the inquiry dialogue questions in the inquiry dialogue question set according to the TAG semantic model;
counting the number of inquiry dialogue problems under each fixed direction according to the TAG sequence model, and extracting the fixed direction with the largest number of inquiry dialogue problems;
and selecting one inquiry dialogue question and the fixed pointing label as the inquiry dialogue question with the fixed pointing label under the extracted fixed pointing direction.
8. The method of claim 3, further comprising:
training a PTM model, comprising:
acquiring historical inquiry data; wherein the historical interrogation data comprises one or more of: intelligent inquiry questions and answers, medical record sorting information and simulated inquiry and answer information;
performing semantic annotation on historical inquiry information in the historical inquiry data to respectively obtain inquiry questions in the historical inquiry information and corresponding inquiry question types;
training according to each inquiry question and the corresponding inquiry question type to obtain a TAG model;
and training to obtain a PTM model according to the inquiry question and the TAG model.
9. The method of claim 8, wherein generating an interrogation dialog question with a fixed point tag based on the patient information and a preset model comprises:
and taking the patient information as input and the inquiry dialogue questions with the fixed pointing labels as output, and carrying out PTM model training to obtain the corresponding inquiry dialogue questions with the fixed pointing labels.
10. An interrogation dialog system, the system comprising:
the acquisition unit is used for acquiring patient information;
the processing unit is used for generating an inquiry dialogue problem with a fixed pointing label according to the patient information and a preset model;
the pushing unit is used for pushing the inquiry dialogue questions to the user side;
the acquisition unit is further configured to: recovering user response information for the inquiry dialogue questions;
the processing unit is further to: and analyzing the answering information of the user based on the fixed pointing label of the inquiry dialogue question to obtain the demand information under the corresponding fixed pointing label.
11. A computer-readable storage medium having instructions stored thereon, which when executed on a computer, cause the computer to perform the interrogation session method of any of claims 1-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111357645.7A CN114300160B (en) | 2021-11-16 | 2021-11-16 | Inquiry dialogue method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111357645.7A CN114300160B (en) | 2021-11-16 | 2021-11-16 | Inquiry dialogue method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114300160A true CN114300160A (en) | 2022-04-08 |
CN114300160B CN114300160B (en) | 2022-10-18 |
Family
ID=80964321
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111357645.7A Active CN114300160B (en) | 2021-11-16 | 2021-11-16 | Inquiry dialogue method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114300160B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115982335A (en) * | 2023-02-14 | 2023-04-18 | 智慧眼科技股份有限公司 | Active AI medical question-answering system, method, equipment and storage medium |
Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107247868A (en) * | 2017-05-18 | 2017-10-13 | 深思考人工智能机器人科技(北京)有限公司 | A kind of artificial intelligence aids in interrogation system |
CN109036588A (en) * | 2018-09-10 | 2018-12-18 | 百度在线网络技术(北京)有限公司 | The method, apparatus, equipment and computer-readable medium of interrogation on line |
CN109524108A (en) * | 2018-11-21 | 2019-03-26 | 上海依智医疗技术有限公司 | A kind of method for building up and device of inquiry model |
US20190207911A1 (en) * | 2017-12-28 | 2019-07-04 | Ethicon Llc | Interactive surgical systems with encrypted communication capabilities |
US20200151516A1 (en) * | 2018-11-13 | 2020-05-14 | CurieAI, Inc. | Methods for Explainability of Deep-Learning Models |
CN111221947A (en) * | 2019-12-26 | 2020-06-02 | 北京邮电大学 | Multi-round conversation realization method and device of ophthalmologic pre-inquiry device |
CN111415759A (en) * | 2020-03-03 | 2020-07-14 | 北京中锐福宁控股集团有限公司 | Human-computer interaction method and system of traditional Chinese medicine pre-diagnosis robot based on inquiry |
CN111599488A (en) * | 2020-05-19 | 2020-08-28 | 万达信息股份有限公司 | Intelligent inquiry implementing method, system and storage medium |
CN111899884A (en) * | 2020-06-23 | 2020-11-06 | 北京左医科技有限公司 | Intelligent auxiliary inquiry method, device and storage medium |
CN112035636A (en) * | 2020-08-28 | 2020-12-04 | 康键信息技术(深圳)有限公司 | Question-answer management method, device, equipment and storage medium of medical inquiry system |
CN112164477A (en) * | 2020-10-07 | 2021-01-01 | 北京大学 | Intelligent question-answering system for hypertension patient based on knowledge graph and establishing method thereof |
CN112700832A (en) * | 2021-01-06 | 2021-04-23 | 北京左医科技有限公司 | Personalized electronic case generation method and system |
CN112768091A (en) * | 2021-01-26 | 2021-05-07 | 北京搜狗科技发展有限公司 | Method, device and medium for processing inquiry information |
CN113409907A (en) * | 2021-07-19 | 2021-09-17 | 广州方舟信息科技有限公司 | Intelligent pre-inquiry method and system based on Internet hospital |
CN113611434A (en) * | 2021-08-26 | 2021-11-05 | 中国医学科学院阜外医院 | Auxiliary inquiry system and method |
-
2021
- 2021-11-16 CN CN202111357645.7A patent/CN114300160B/en active Active
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107247868A (en) * | 2017-05-18 | 2017-10-13 | 深思考人工智能机器人科技(北京)有限公司 | A kind of artificial intelligence aids in interrogation system |
US20190207911A1 (en) * | 2017-12-28 | 2019-07-04 | Ethicon Llc | Interactive surgical systems with encrypted communication capabilities |
CN109036588A (en) * | 2018-09-10 | 2018-12-18 | 百度在线网络技术(北京)有限公司 | The method, apparatus, equipment and computer-readable medium of interrogation on line |
US20200151516A1 (en) * | 2018-11-13 | 2020-05-14 | CurieAI, Inc. | Methods for Explainability of Deep-Learning Models |
CN109524108A (en) * | 2018-11-21 | 2019-03-26 | 上海依智医疗技术有限公司 | A kind of method for building up and device of inquiry model |
CN111221947A (en) * | 2019-12-26 | 2020-06-02 | 北京邮电大学 | Multi-round conversation realization method and device of ophthalmologic pre-inquiry device |
CN111415759A (en) * | 2020-03-03 | 2020-07-14 | 北京中锐福宁控股集团有限公司 | Human-computer interaction method and system of traditional Chinese medicine pre-diagnosis robot based on inquiry |
CN111599488A (en) * | 2020-05-19 | 2020-08-28 | 万达信息股份有限公司 | Intelligent inquiry implementing method, system and storage medium |
CN111899884A (en) * | 2020-06-23 | 2020-11-06 | 北京左医科技有限公司 | Intelligent auxiliary inquiry method, device and storage medium |
CN112035636A (en) * | 2020-08-28 | 2020-12-04 | 康键信息技术(深圳)有限公司 | Question-answer management method, device, equipment and storage medium of medical inquiry system |
CN112164477A (en) * | 2020-10-07 | 2021-01-01 | 北京大学 | Intelligent question-answering system for hypertension patient based on knowledge graph and establishing method thereof |
CN112700832A (en) * | 2021-01-06 | 2021-04-23 | 北京左医科技有限公司 | Personalized electronic case generation method and system |
CN112768091A (en) * | 2021-01-26 | 2021-05-07 | 北京搜狗科技发展有限公司 | Method, device and medium for processing inquiry information |
CN113409907A (en) * | 2021-07-19 | 2021-09-17 | 广州方舟信息科技有限公司 | Intelligent pre-inquiry method and system based on Internet hospital |
CN113611434A (en) * | 2021-08-26 | 2021-11-05 | 中国医学科学院阜外医院 | Auxiliary inquiry system and method |
Non-Patent Citations (3)
Title |
---|
EMB/11073/EMBS_WG WORKING GROUP: "Draft Standard for Health informatics - Point-of-care medical device communication - Nomenclature - Implantable device, cardiac", 《IEEE》 * |
李舟军等: "基于Web的问答系统综述", 《计算机科学》 * |
田迎等: "基于知识图谱的抑郁症自动问答系统研究", 《湖北大学学报(自然科学版)》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115982335A (en) * | 2023-02-14 | 2023-04-18 | 智慧眼科技股份有限公司 | Active AI medical question-answering system, method, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN114300160B (en) | 2022-10-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111128394B (en) | Medical text semantic recognition method and device, electronic equipment and readable storage medium | |
CN108986908A (en) | Interrogation data processing method, device, computer equipment and storage medium | |
US7685082B1 (en) | System and method for identifying, prioritizing and encapsulating errors in accounting data | |
US20210233191A1 (en) | Method, apparatus and computer program for operating a machine learning framework with active learning technique | |
Camburu | Explaining deep neural networks | |
CN110931128B (en) | Method, system and device for automatically identifying unsupervised symptoms of unstructured medical texts | |
JPH07295989A (en) | Device that forms interpreter to analyze data | |
CN111428448A (en) | Text generation method and device, computer equipment and readable storage medium | |
US20190377996A1 (en) | Method, device and computer program for analyzing data | |
CN109902301A (en) | Relation inference method, device and equipment based on deep neural network | |
CN114547274B (en) | Multi-turn question and answer method, device and equipment | |
CN114300160B (en) | Inquiry dialogue method and system | |
CN112613322A (en) | Text processing method, device, equipment and storage medium | |
KR102275658B1 (en) | System for automatically collecting and learning data using machine learning model | |
CN112685550A (en) | Intelligent question answering method, device, server and computer readable storage medium | |
Fedele et al. | Explaining siamese networks in few-shot learning for audio data | |
KR20190049627A (en) | Method, apparatus and computer program for interpreting analysis results of machine learning framework | |
CN110909174B (en) | Knowledge graph-based method for improving entity link in simple question answering | |
CN107704580A (en) | Question and answer method for pushing, device, server and storage medium based on user's period | |
Wang et al. | Coad: Automatic diagnosis through symptom and disease collaborative generation | |
CN116186223A (en) | Financial text processing method, device, equipment and storage medium | |
CN112949305B (en) | Negative feedback information acquisition method, device, equipment and storage medium | |
CN115145928A (en) | Model training method and device and structured abstract acquisition method and device | |
CN113761149A (en) | Dialogue information processing method, device, computer equipment and storage medium | |
Jayasuriya et al. | Ontology based software design documentation for design reasoning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |