WO2022000140A1 - Méthode de dépistage épidémique et appareil combinant rpa avec ai - Google Patents
Méthode de dépistage épidémique et appareil combinant rpa avec ai Download PDFInfo
- Publication number
- WO2022000140A1 WO2022000140A1 PCT/CN2020/098590 CN2020098590W WO2022000140A1 WO 2022000140 A1 WO2022000140 A1 WO 2022000140A1 CN 2020098590 W CN2020098590 W CN 2020098590W WO 2022000140 A1 WO2022000140 A1 WO 2022000140A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- epidemic
- investigation
- dialect
- rpa
- rpa system
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000012216 screening Methods 0.000 title abstract description 17
- 238000004891 communication Methods 0.000 claims abstract description 46
- 238000011835 investigation Methods 0.000 claims description 256
- 230000011218 segmentation Effects 0.000 claims description 56
- 238000003860 storage Methods 0.000 claims description 13
- 238000007689 inspection Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 11
- 239000000284 extract Substances 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 19
- 238000005457 optimization Methods 0.000 abstract description 9
- 238000013473 artificial intelligence Methods 0.000 abstract description 8
- 230000004044 response Effects 0.000 abstract description 3
- 238000013024 troubleshooting Methods 0.000 description 22
- 230000014509 gene expression Effects 0.000 description 13
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 9
- 238000003058 natural language processing Methods 0.000 description 6
- 230000036760 body temperature Effects 0.000 description 4
- 238000004801 process automation Methods 0.000 description 3
- 238000013519 translation Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 208000015181 infectious disease Diseases 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 230000002194 synthesizing effect Effects 0.000 description 2
- 230000001755 vocal effect Effects 0.000 description 2
- 241000700605 Viruses Species 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000035622 drinking Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 244000052769 pathogen Species 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
Definitions
- the present application relates to the field of artificial intelligence technology, and in particular, to an epidemic investigation method and device combining RPA and AI.
- Robotic Process Automation for short is a specific "robot software” that simulates human operations on a computer and automatically performs process tasks according to rules.
- Artificial Intelligence the English abbreviation is AI. It is a new technical science that studies and develops theories, methods, techniques and application systems for simulating, extending and expanding human intelligence. Artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, Natural language processing and expert systems, etc.
- the present application provides an epidemic investigation method and device combining RPA and AI, so as to solve the technical problems of low epidemic investigation efficiency and high risk in the prior art.
- the embodiment of the present application provides an epidemic investigation method combining RPA and AI, including: the RPA system obtains the communication directory of the person to be investigated, and generates an epidemic investigation form, wherein the epidemic investigation form includes a plurality of epidemic investigation problems and the Multiple to-be-filled items corresponding to multiple epidemic investigation problems respectively; the RPA system generates a synthetic voice for epidemic investigation problems according to the list of epidemic investigation problems; the RPA system dials the to-be-waited items in turn according to the communication directory through the outbound call system The phone number of the investigator, and voice broadcast of the synthesized voice of the epidemic investigation problem; the RPA system obtains the reply voice of the person to be inspected for the synthesized voice of the epidemic investigation problem; the RPA system recognizes the reply voice to generating a candidate identification result and a dialect type; the RPA system identifies the dialect word segmentation corresponding to the dialect type in the candidate identification result, and replaces the dialect word segmentation in the candidate identification result with a standard word segmentation to generate target recognition Results; the RPA system automatically
- the device is applied to an RPA system and includes: a first generation module, configured to obtain a communication directory of a person to be investigated, and generate an epidemic investigation table, wherein , the epidemic investigation table includes a plurality of epidemic investigation problems and a plurality of items to be filled in respectively corresponding to the plurality of epidemic investigation problems; a second generation module is used to generate a synthetic voice of epidemic investigation problems according to the list of epidemic investigation problems; The broadcasting module is used for dialing the phone numbers of the people to be checked in turn according to the communication directory through the outbound call system, and voice broadcasts the synthetic voice of the epidemic situation checking problem; the first obtaining module is used for obtaining the information about the people to be checked against the people to be checked.
- the reply speech of the synthesized speech for the above epidemic investigation problem is used to recognize the reply speech to generate the candidate recognition result and the dialect type; the replacement module is used to recognize the candidate recognition result corresponding to the dialect type dialect word segmentation, and replace the dialect word segmentation in the candidate identification results with standard word segmentation to generate target identification results; the form filling module is used to automatically fill in the target identification results into the epidemic investigation table and The items to be filled out corresponding to the epidemic investigation questions; and an investigation module, used for epidemic investigation according to the epidemic investigation form.
- Yet another embodiment of the present application provides a computer device, including a processor and a memory; wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, In order to realize the epidemic investigation method combining RPA and AI as described in the above embodiment.
- Another embodiment of the present application provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements the epidemic investigation method combining RPA and AI as described in the above embodiment .
- the RPA system obtains the communication directory of the person to be checked, and generates an epidemic investigation form, wherein the epidemic investigation form includes multiple epidemic investigation problems and multiple to-be-filled items corresponding to the multiple epidemic investigation problems, and then generates the epidemic situation according to the list of epidemic investigation problems. Synthesize the voice for troubleshooting problems, and dial the phone numbers of the people to be checked in sequence according to the communication directory through the outbound call system, and broadcast the synthetic voice of the epidemic troubleshooting problems. Identify to generate candidate identification results, and after dialect optimization of the candidate identification results, the obtained target identification results are automatically filled in the to-be-filled items corresponding to the epidemic investigation problems in the epidemic investigation table, so as to conduct epidemic investigation according to the epidemic investigation form.
- epidemic investigation is realized, which improves the efficiency of investigation, reduces the cost and risk of investigation, and optimizes the dialect of the reply voice of the person to be inspected, which ensures the reliability of the investigation result.
- FIG. 1 is a flowchart of an epidemic investigation method combining RPA and AI according to an embodiment of the present application
- FIG. 2 is a schematic diagram of an epidemic investigation table according to an embodiment of the present application.
- FIG. 3 is a schematic diagram of a synthesis and generation interface for troubleshooting problems according to an embodiment of the present application
- FIG. 4-1 is a schematic diagram of a scenario that contains content in a synthesized speech for an epidemic investigation problem according to an embodiment of the present application
- FIG. 4-2 is a schematic diagram of a scenario that contains content in a synthesized speech for an epidemic investigation problem according to another embodiment of the present application;
- FIG. 5 is a flowchart of an epidemic investigation method combining RPA and AI according to another embodiment of the present application.
- FIG. 6 is a flowchart of an epidemic investigation method combining RPA and AI according to another embodiment of the present application.
- FIG. 7 is a schematic diagram of a network architecture of an epidemic investigation method according to an embodiment of the present application.
- FIG. 8 is a schematic structural diagram of an epidemic investigation device combining RPA and AI according to an embodiment of the present application.
- FIG. 9 is a schematic structural diagram of a computer device according to an embodiment of the present application.
- the applicant of this application provides an idea of using a robot to conduct epidemic investigation, and using artificial intelligence to conduct epidemic investigation is the main method of this application to achieve epidemic investigation.
- Robotic Process Automation refers to an automated processing technology that simulates human operations on a computer through specific "robot software” and automatically performs process tasks according to Guo Ze.
- the automatic outbound robot can work 7*24 hours without any emotion as long as it has data, it can answer in full accordance with the standard words, which can greatly reduce labor costs, improve work efficiency, and eliminate the need for manual investigations. the risk of epidemic investigation.
- this application also involves the optimization of dialects to ensure the reliability of the investigation results.
- FIG. 1 is a flowchart of an epidemic investigation method combining RPA and AI according to an embodiment of the present application. As shown in FIG. 1 , the method includes:
- Step 1 the RPA system obtains the communication directory of the person to be checked, and generates an epidemic investigation form, wherein the epidemic investigation form includes multiple epidemic investigation questions and multiple to-be-filled items corresponding to the multiple epidemic investigation questions respectively.
- the person to be checked can correspond to different sensitive groups according to the needs of the scene, for example, the sensitive group corresponding to a community, such as the sensitive group corresponding to the area above the ultimate risk, etc.
- the method of the sensitive group can be actively uploaded by the user, or can be obtained by obtaining historical location information within a preset time period of multiple people.
- Sensitive area information the sensitive area information can be specified by the relevant unit, or can be obtained from the statistics of the area where the current number of infected people is located.
- the RPA system regards people as a sensitive group.
- the corresponding person is an antibody carrier, or the corresponding person has been tested for pathogens and found that they do not carry the virus. Therefore, it is possible to obtain the current status information of the people in the sensitive group.
- the current status information may include the antibody test report of the corresponding person, etc. If the current status information of the corresponding person is judged to be a non-investigation condition, for example, it belongs to a person who carries antibodies, Then the RPA system removes the person from the sensitive group, thereby further improving the efficiency of epidemic investigation.
- the RPA system After the RPA system determines the sensitive group, it obtains the communication information (such as communication number, social account), etc. of the sensitive group determined by the RPA system to generate a communication directory with the person to be checked.
- the RPA system determines the communication information corresponding to the sensitive group. It can be uploaded actively by relevant staff, or obtained through communication with a relevant platform, for example, it can be obtained through a storage server with a communication operator.
- the epidemic investigation table corresponds to a plurality of epidemic investigation questions related to epidemic investigation and a plurality of to-be-filled items corresponding to a plurality of epidemic investigation questions respectively.
- the items to be filled are the question elements related to the epidemic investigation problem. For example, when the epidemic investigation question is "When have you been to area A recently?", the items to be filled can include the question element "Have you been to place A?” and the question element "Time to go to place A", etc. Among them, one epidemic investigation question can correspond to at least one to-be-filled item.
- Step 2 the RPA system generates a synthetic voice for epidemic investigation problems according to the list of epidemic investigation problems.
- the RPA system generates a synthetic voice of epidemic investigation questions according to the list of epidemic investigation questions.
- the synthetic voice of epidemic investigation questions includes the questioning voice of the corresponding epidemic investigation question, and the synthetic voice of epidemic investigation questions replaces the traditional manual inquiries of staff, which can be repeated multiple times. Repeated, and completely in accordance with the standard vocabulary expression, improve the professionalism of the service.
- the synthesized speech of the epidemic investigation question can be generated after the questioning voice of each epidemic investigation question is generated according to natural language processing technology.
- the purpose is to make the expression of the epidemic investigation question more natural and real.
- the RPA system generates different ways of synthesizing speech for epidemic investigation problems according to the list of epidemic investigation problems. Examples are as follows:
- the question expression model is trained based on a large amount of sample data in advance, and the question expression model can obtain the expression text of the corresponding language according to the input epidemic investigation question. ”, then the verbal expression text obtained by the question expression model according to the epidemic investigation question is “Where have you been in the last week?”, so as to generate the corresponding voice according to the verbal expression text to obtain the corresponding voice of each epidemic investigation question.
- Troubleshooting voices among which, in order to further improve the sense of intelligence, when generating troubleshooting voices, multiple voices for troubleshooting problems corresponding to each epidemic investigation problem can be used in different tones, such as cartoon tones, female tones, man tones, etc. , so that in the subsequent playback of the synthesized voice for troubleshooting, the intonation may be randomly selected for playback, or the intonation may be selected for playback according to the age characteristics of the person to be checked.
- a generation interface for synthesizing speech for troubleshooting problems is constructed.
- the generation interface includes a question text input box and a tone selection box for each epidemic troubleshooting question. Enter the question text and select the tone in the tone selection box to generate the troubleshooting voice for each epidemic troubleshooting question.
- the synthesized voice of the epidemic investigation question in this embodiment includes the voice of multiple investigation questions corresponding to the multiple epidemic investigation questions corresponding to the epidemic investigation question list, in order to leave enough time for the person to be investigated to answer and ensure clarity.
- the answering voices of the people to be checked are collected in order to play the troubleshooting voices of multiple epidemic investigation problems in sequence, and pause for a certain period of time after each troubleshooting voice.
- the certain time can be the same, or different times can be set according to the troubleshooting voice.
- the difficulty value of the troubleshooting voice is marked in advance according to the experimental data. The higher the difficulty value of the troubleshooting voice, the longer the corresponding pause. Wait. That is to say, the synthesized voice for epidemic investigation problems in this embodiment includes investigation voices of multiple epidemic investigation problems, and the playing time interval between the voices for investigation problems.
- the troubleshooting voices of multiple epidemic troubleshooting problems are played in sequence, and after each troubleshooting voice is played, it is judged whether the playback switching conditions are met, for example, monitoring the pending troubleshooting
- the human voice pauses for a long time. If the voice of the person to be checked is not detected within the preset time period, it will switch to the next troubleshooting voice. That is to say, the synthesized voice for epidemic investigation problems in this embodiment includes investigation voices for multiple epidemic investigation problems, and playback switching conditions between the voices for investigation problems.
- Step 3 the RPA system dials the phone numbers of the people to be checked in sequence through the outbound call system according to the communication directory, and broadcasts the synthetic voice of the epidemic situation check problem.
- the environmental noise on the side of the person to be checked and the voiceprint characteristics of the person to be checked can also be collected when the voice broadcasts the epidemic investigation problem and synthesized speech.
- the pattern feature analyzes the age of the person to be checked, and adjusts the volume of the synthetic voice for epidemic investigation problems in real time according to environmental noise and age broadcasts. .
- Step 4 the RPA system obtains the reply voice of the person to be checked that is synthesized into the voice for the epidemic investigation problem.
- the RPA system dials the phone numbers of the people to be checked in sequence through the outbound call system according to the communication directory, and broadcasts the epidemic investigation problems to synthesize the voice, so as to collect the reply voices of the people to be checked based on the list of epidemic investigation problems.
- the answering status of the phone of the person to be checked can be monitored by the RPA system.
- the RPA system When the answering status is connected, the RPA system enables the recording function to start recording, and when the answering status is hung up, the RPA system Turn off the recording function and obtain the recording data, and extract the reply voice of the person to be checked in the recording data.
- the epidemic investigation table may also include the storage path item of the reply voice.
- the reply voice is stored in a preset location to generate a storage path, and the corresponding storage path is stored.
- the items to be filled in the epidemic investigation table may also include answering status items, answering time items, etc., so as to record the telephone dialing status of the person to be checked each time.
- Step 5 the RPA system recognizes the reply speech to generate candidate recognition results and dialect types.
- the RPA system in this application recognizes the reply speech to generate a candidate recognition result, it does not directly use the candidate recognition result as the investigation result, but obtains the dialect type, where , the dialect type is determined according to the currently divided dialect area, including Shanghai dialect, Cantonese, Tianjin dialect, North China dialect, Northwest dialect, etc.
- step 5 the implementation of step 5 is different, the examples are as follows
- step 5 includes:
- Step 51 The RPA system performs text recognition on the reply speech to generate a candidate recognition result.
- text conversion can be performed based on Automatic Speech Recognition (ASR) technology to obtain candidate recognition results.
- ASR Automatic Speech Recognition
- the RPA system performs semantic recognition on the text results to generate candidate recognition results, so that the text results are processed into semantic expression results, and the noise interference in language expression is removed. For example, when the text recognition result is "I just Beijing is back", the candidate recognition result after speech recognition is "Visit Beijing on June 20, 2020".
- Step 52 The RPA system extracts the attribution keyword in the candidate identification result.
- the attribution keyword in the candidate identification result is extracted, and the attribution keyword can be understood as the attribution key number for the user's current phone number, and can be understood as the user's residence keyword.
- Step 53 The RPA system determines the dialect type according to the attribution keyword.
- the dialect type can be determined according to the attribution keyword.
- the same dialect type can The dialect type with the largest number of attribution keywords is used as the last dialect type.
- step 5 includes:
- Step 54 the RPA system performs text recognition on the reply speech to generate a candidate recognition result.
- text conversion can be performed based on Automatic Speech Recognition (ASR) technology to obtain candidate recognition results.
- ASR Automatic Speech Recognition
- the RPA system performs semantic recognition on the text results to generate candidate recognition results, so that the text results are processed into semantic expression results, and the noise interference in language expression is removed. For example, when the text recognition result is "I just Beijing is back", the candidate recognition result after speech recognition is "Visit Beijing on June 20, 2020".
- Step 55 the RPA system extracts intonation features in the reply speech.
- Step 56 the RPA system determines the dialect type according to the intonation feature.
- the intonation characteristics of different dialect types are different, the intonation characteristics may include intonation changes in the reply speech, the composition of the flat tongue, etc. Therefore, in this embodiment, the dialect type is determined according to the intonation characteristics.
- Step 6 the RPA system identifies the dialect word segmentation corresponding to the dialect type in the candidate identification result, and replaces the dialect word segmentation in the candidate identification result with the standard word segmentation to generate the target identification result.
- the RPA system identifies the dialect word segmentation corresponding to the dialect type in the candidate identification result, and replaces the dialect word segmentation in the candidate identification result with the standard word segmentation to generate the target identification result. For example, for the scenario proposed in the above embodiment, "harmony" in the candidate identification result is replaced with "Hexi".
- the RPA system performs word segmentation processing on the candidate recognition results to generate a plurality of first word segmentations
- the RPA system queries the correspondence table corresponding to the dialect type, and determines that the plurality of first word segmentations belong to the correspondence table
- the dialect participle of the dialect type in which the relationship table corresponding to the dialect type stores the correspondence between the dialect participle and the standard participle under the dialect type. For example, the standard participle: "Hexi" and the wrong dialect participle under the corresponding dialect type are stored. "Hexi” and “harmony” are the corresponding articles. Therefore, the RPA system queries the correspondence table, determines the standard word segmentation corresponding to the dialect word segmentation, and replaces the dialect word segmentation in the candidate recognition result with the standard word segmentation to generate the target recognition result.
- a dialect dictionary corresponding to each epidemic investigation question in the candidate results is preliminarily targeted, and the dialect dictionary corresponds to the possible translation results (including correct results and incorrect results) corresponding to each epidemic investigation question. result) and the corresponding correct result.
- the corresponding dialect dictionary includes ⁇ "Peace”:["Peace”,”Drinking a bottle”],”Hedong”:["Hedong”,”Answer”],”Hexi”:["Hexi”,”Hexi”,” Harmony”],”Nankai”:["Nankai”,”That piece”],”Jizhou”:["Jizhou”,”Jeju”,”Izhou”] ⁇ , thus, according to the way of dictionary traversal to determine the target Identify the wrong word segmentation in the candidate results corresponding to the current epidemic investigation problem, and replace the corresponding word segmentation with the correct identification result.
- Step 7 The RPA system automatically fills in the target identification results into the items to be filled in the epidemic investigation table corresponding to the epidemic investigation problems.
- the investigation result is actually obtained.
- the RPA system recognizes the reply voice to generate candidate recognition results, and after dialect optimization of the candidate recognition results, will generate
- the target identification results of the epidemic situation are automatically filled into the to-be-filled items corresponding to the epidemic investigation problems in the epidemic investigation form, so as to complete the sorting of the investigation results and facilitate the epidemic investigation. Since the target identification results are counted in the form of the epidemic investigation form in this application, It is convenient for the sorting of subsequent epidemic results.
- step 7 is different, and the examples are as follows:
- the RPA system performs the segmentation process on the target recognition result to generate multiple second word segments, and the RPA system performs attribute analysis on each second word segment in the multiple recognition results, and determines the attribute information of each second word segment,
- the attribute information distinguishes the attributes in the prior art, and corresponds to the attributes of the items to be filled, such as whether the participle is a place noun, a time noun, or a body temperature noun, etc.
- each to-be-filled item is set with corresponding attributes, and the RPA system is based on the attributes. The information determines the to-be-filled item corresponding to each second participle, and fills in the corresponding to-be-filled item for each second participle.
- the neural network model is pre-trained, and after the target recognition result is processed, the word segmentation corresponding to the target recognition result is obtained, and multiple word segmentations are input into the neural network model, and the number of words associated with each word segmentation and each to-be-filled item is obtained.
- Matching value according to the matching value, determine the participle that matches the item to be filled, and fill in the corresponding item to be filled.
- Step 8 the RPA system conducts epidemic investigation according to the epidemic investigation table.
- the epidemic investigation table clearly reflects the investigation results of the personnel to be investigated. Therefore, the RPA system conducts epidemic investigation according to the epidemic investigation form to avoid the problem of low efficiency caused by manual investigation.
- the inspection requirements are set for the content of the item to be filled in advance, and the RPA system determines whether the content in the to-be-filled item meets the inspection requirements. requirements, the person to be checked is determined to be a low-risk person.
- the RPA system obtains the activity location information in the items to be filled, and the RPA system determines whether the activity location information belongs to a preset risk area. .
- the RPA system obtains the body temperature information in the items to be filled, and the RPA system determines whether the body temperature information is greater than or equal to the preset temperature. If it is greater than the preset temperature, the content of the items to be filled meets the inspection requirements.
- the epidemic investigation method combining RPA and AI in the embodiment of the present application is applied to the robot.
- the robot uses RPA and automatic outbound call technology, and the RPA is responsible for obtaining outbound people to be investigated, sorting out the epidemic investigation table, and creating outbound calling tasks. , upload the epidemic investigation table, start the outbound call task, download the reply voice of the outbound call, generate the report of the epidemic investigation table and send it to the relevant personnel, which greatly improves the efficiency of the epidemic investigation.
- 300 people to be checked, each call is about 5 minutes (including information recording time), and it takes 4 hours a day to record and count the physical conditions of the people to be checked, time is tight and work intensity is high.
- an outbound call form can only be automatically generated by judging intentions, and only some questions whose answers are affirmative or negative can be asked. If you ask questions such as age, education, body temperature, area, etc., it is impossible to determine the intention, and it is impossible to automatically generate a form. Therefore, the outbound call system usually does not ask open-ended questions, which greatly limits the application scenarios of the outbound call system.
- the method for implementing epidemic investigation includes the business system side, the RPA side, and the outbound call platform, wherein the business system side includes client software, mailboxes, and web pages. , database, EXCEL form, etc., use the RPA data collection capability to automatically collect communication information (such as phone numbers, etc.) from the customer's business system, such as reading communication information based on the database, and generate a communication directory according to the requirements of the outbound call system.
- communication information such as phone numbers, etc.
- the predetermined rules automatically log in to the outbound call system to create outbound dialects and tasks, import outbound call lists, and start outbound tasks; after the outbound call is completed, the RPA robot extracts the list of people to be checked about the epidemic situation investigation problem to generate a synthetic voice for epidemic investigation problems ;
- the RPA system uses the outbound call system according to the reply voice of the communication, and then automatically generates the epidemic investigation table after dialect optimization processing.
- the RPA can push the outbound call report to the relevant personnel on the client side through email, IM, etc. It is also possible to write back the contents of the epidemic checklist to the customer's business system.
- the RPA side After the RPA side captures the relevant data from the business system, it stores the relevant data in the memory. It needs to perform data cleaning and data screening, extract the communication directory required by the outbound call platform, and write the communication directory into the epidemic investigation table. And according to certain data cleaning rules and screening conditions, the RPA process is formulated, and the RPA automatically completes the data processing in the background. RPA needs to complete actions such as logging in to the outbound call platform, creating an outbound call task, uploading a user form, and opening an outbound call task.
- RPA automatically opens the webpage, enters the user name, password, verification code, triggers the login button to log in, and then generates the epidemic investigation table according to the received communication directory and current business needs, and constructs the call tasks and call rules for the people to be checked, and then Trigger the outbound call platform to make outbound calls.
- Relevant services on the RPA side can be based on artificial intelligence-based processing strategies, based on technology, and through specific "robot software" to simulate human operations on the computer, and automatically execute the above process tasks according to rules.
- the outbound call platform in this application may include an artificial intelligence module, and the artificial intelligence module includes an automatic speech recognition technology (Automatic Speech Recognition, ASR) module, natural language Processing (Natural Language Processing, NLP) module, from text to speech (TextToSpeech, TTS) module, also includes IP multimedia subsystem, used to coordinate functions such as speech recognition and speech synthesis.
- ASR Automatic Speech Recognition
- NLP Natural Language Processing
- TTS text to speech
- IP multimedia subsystem used to coordinate functions such as speech recognition and speech synthesis.
- the outbound call platform uses automatic outbound call technology to realize batch automatic outbound calls, uses TTS technology to simulate human voices to broadcast epidemic investigation problems to the called party to synthesize voice, and uses NLP engine to realize multiple rounds of human-machine dialogue, and the entire call process is recorded. After the call is completed, the outbound call platform uses ASR technology to translate the recording to the dialogue text.
- the robot in this embodiment can save a lot of time of the community personnel.
- the community personnel only need to send the original communication directory and the like to the designated device, and then wait for receiving the investigation result. If there are 300 people to be checked in a community who need to make outbound calls, it only takes 15 minutes to use the robot to complete all operations, extract the question and answer responses of the people to be checked, and generate an epidemic checklist based on the question and answer recovery and send it to the community personnel to achieve end-to-end automation.
- Use RPA robots to automate the solid, cumbersome and repetitive business to achieve the goal of optimizing the central part of the workflow, reducing costs, improving work efficiency, and reducing the rate of operational errors.
- the RPA system obtains the communication directory of the person to be investigated, and generates an epidemic investigation form, wherein the epidemic investigation form includes a plurality of epidemic investigation problems and a plurality of epidemic investigation problems.
- the epidemic investigation form includes a plurality of epidemic investigation problems and a plurality of epidemic investigation problems.
- the reply voice is recognized to generate a candidate recognition result, and after dialect optimization of the candidate recognition result, the obtained target recognition result is automatically filled in the epidemic investigation form and the epidemic investigation
- the to-be-filled items corresponding to the questions are used for epidemic investigation according to the epidemic investigation form.
- epidemic investigation is realized, which improves the efficiency of investigation, reduces the cost and risk of investigation, and optimizes the dialect of the reply voice of the person to be inspected, which ensures the reliability of the investigation result.
- FIG. 8 is a schematic structural diagram of an epidemic investigation apparatus combining RPA and AI according to an embodiment of the present application.
- the epidemic investigation apparatus combining RPA and AI includes: a first generation module 10 and a second generation module 20 , the broadcast module 30, the first acquisition module 40, the identification module 50, the replacement module 60, the form filling module 70 and the investigation module 80, wherein,
- the first generating module 10 is used to obtain the communication directory of the person to be checked, and generate an epidemic investigation table, wherein the epidemic investigation table includes a plurality of epidemic investigation problems and a plurality of items to be filled corresponding to the plurality of epidemic investigation problems respectively;
- the second generation module 20 is configured to generate a synthetic voice of the epidemic investigation problem according to the list of epidemic investigation problems;
- the broadcasting module 30 is used for dialing the telephone numbers of the persons to be checked in sequence according to the communication catalogue through the outbound calling system, and broadcasting the epidemic situation checking problems to synthesize the voice;
- the first obtaining module 40 is used for obtaining the reply voice synthesized by the person to be checked for the epidemic investigation problem
- the recognition module 50 is used for recognizing the reply speech to generate the candidate recognition result and dialect type
- the replacement module 60 is used to identify the dialect word segmentation corresponding to the dialect type in the candidate identification result, and replace the dialect word segmentation in the candidate identification result with the standard word segmentation to generate the target identification result;
- the form filling module 70 is used to automatically fill in the target identification result into the to-be-filled item corresponding to the epidemic investigation problem in the epidemic investigation form;
- the investigation module 80 is used to conduct epidemic investigation according to the epidemic investigation table.
- the recognition module 50 is specifically configured to: perform text recognition on the reply speech, and generate a candidate recognition result
- the dialect type is determined according to the attribution keywords.
- the recognition module 50 is specifically configured to: perform text recognition on the reply speech, and generate a candidate recognition result
- Dialect types are determined based on intonation characteristics.
- the replacement module 60 is specifically used for:
- the candidate recognition result is word-segmented to generate multiple first word segmentations
- the form filling module 70 is specifically used for:
- the word segmentation process of the target recognition result generates a plurality of second word segmentations
- the to-be-filled item corresponding to each second participle is determined according to the attribute information, and each second participle is filled into the corresponding to-be-filled item.
- the troubleshooting module 70 is specifically used for:
- the person to be inspected is determined to be a low-risk person.
- the RPA system obtains the communication directory of the person to be investigated, and generates an epidemic investigation table, wherein the epidemic investigation table includes a plurality of epidemic investigation problems and a plurality of epidemic investigation problems. After corresponding multiple items to be filled in, generate a synthetic voice for epidemic investigation problems according to the list of epidemic investigation problems, and dial the numbers of the people to be checked in sequence according to the communication directory through the outbound call system, and broadcast the synthetic voice of the epidemic investigation problems, and obtain the synthetic voice of the epidemic investigation problems.
- the reply voice is recognized to generate a candidate recognition result, and after dialect optimization of the candidate recognition result, the obtained target recognition result is automatically filled in the epidemic investigation form and the epidemic investigation
- the to-be-filled items corresponding to the questions are used for epidemic investigation according to the epidemic investigation table.
- FIG. 9 is a schematic structural diagram of a computer device according to an embodiment of the present application.
- a memory 21 As shown in FIG. 9 , a memory 21 , a processor 22 , and a computer program stored on the memory 21 and executable on the processor 22 .
- the computer equipment also includes:
- the communication interface 23 is used for communication between the memory 21 and the processor 22 .
- the memory 21 is used to store computer programs that can be executed on the processor 22 .
- the memory 21 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.
- the processor 22 is configured to implement the epidemic investigation method combining RPA and AI described in the above embodiment when executing the program.
- the bus may be an Industry Standard Architecture (referred to as ISA) bus, a Peripheral Component (referred to as PCI) bus or an Extended Industry Standard Architecture (referred to as EISA) bus Wait.
- ISA Industry Standard Architecture
- PCI Peripheral Component
- EISA Extended Industry Standard Architecture
- the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.
- the memory 21, the processor 22 and the communication interface 23 are integrated on one chip, the memory 21, the processor 22 and the communication interface 23 can communicate with each other through the internal interface.
- the processor 22 may be a central processing unit (Central Processing Unit, referred to as CPU), or a specific integrated circuit (Application Specific Integrated Circuit, referred to as ASIC), or is configured to implement one or more of the embodiments of the present application integrated circuit.
- CPU Central Processing Unit
- ASIC Application Specific Integrated Circuit
- the present application also proposes a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by the processor, so that the combination of RPA and AI as described in the above-mentioned embodiments can be executed. method of epidemic investigation.
- the present application also proposes a computer program product, when the instruction processor in the computer program product executes, executes the epidemic investigation method combining RPA and AI as described in the above embodiments.
- first and second are only used for descriptive purposes, and should not be construed as indicating or implying relative importance or implying the number of indicated technical features. Thus, a feature delimited with “first”, “second” may expressly or implicitly include at least one of that feature.
- plurality means at least two, such as two, three, etc., unless expressly and specifically defined otherwise.
- a "computer-readable medium” can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with an instruction execution system, apparatus, or apparatus.
- computer readable media include the following: electrical connections with one or more wiring (electronic devices), portable computer disk cartridges (magnetic devices), random access memory (RAM), Read Only Memory (ROM), Erasable Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM).
- the computer readable medium may even be paper or other suitable medium on which the program may be printed, as the paper or other medium may be optically scanned, for example, followed by editing, interpretation, or other suitable medium as necessary process to obtain the program electronically and then store it in computer memory.
- each functional unit in each embodiment of the present application may be integrated into one processing module, or each unit may exist physically alone, or two or more units may be integrated into one module.
- the above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.
- the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, and the like.
Landscapes
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Machine Translation (AREA)
Abstract
La présente invention concerne une méthode et un appareil de dépistage épidémique combinant RPA et AI, se rapportant au domaine technique de l'intelligence artificielle. La solution technique décrite spécifiquement comprend : un système RPA acquiert un catalogue de communication de personnes à dépister, et génère un tableau de dépistage épidémique ; le système RPA contacte successivement lesdites personnes selon le catalogue de communication et au moyen d'un système d'appel sortant, et effectue une diffusion vocale par parole synthétique de questions de dépistage épidémique ; le système RPA acquiert la réponse desdites personnes en réponse à l'énonciation par parole synthétique des questions de dépistage épidémique ; le système RPA effectue une reconnaissance sur les réponses pour générer des résultats de reconnaissance candidats, effectue une optimisation dialectale des résultats candidats, puis remplit automatiquement les résultats de reconnaissance cibles dans les éléments à remplir, qui correspondent aux questions de dépistage épidémique, dans le tableau de dépistage épidémique ; et le système RPA effectue le dépistage épidémique en fonction du tableau de dépistage épidémique. Ainsi, un dépistage épidémique est réalisé sur la base de RPA et d'une technologie d'appel sortant, l'efficacité de dépistage est améliorée, et les coûts et les risques de dépistage sont réduits ; et une optimisation dialectale est effectuée sur les réponses des personnes à dépister, ce qui permet d'assurer la fiabilité des résultats de dépistage.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2020/098590 WO2022000140A1 (fr) | 2020-06-28 | 2020-06-28 | Méthode de dépistage épidémique et appareil combinant rpa avec ai |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2020/098590 WO2022000140A1 (fr) | 2020-06-28 | 2020-06-28 | Méthode de dépistage épidémique et appareil combinant rpa avec ai |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022000140A1 true WO2022000140A1 (fr) | 2022-01-06 |
Family
ID=79317743
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/098590 WO2022000140A1 (fr) | 2020-06-28 | 2020-06-28 | Méthode de dépistage épidémique et appareil combinant rpa avec ai |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2022000140A1 (fr) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6577713B1 (en) * | 1999-10-08 | 2003-06-10 | Iquest Technologies, Inc. | Method of creating a telephone data capturing system |
US20160379230A1 (en) * | 2015-06-25 | 2016-12-29 | Alibaba Group Holding Limited | System, device, and method for making automatic calls |
CN110489534A (zh) * | 2019-07-26 | 2019-11-22 | 美的置业集团有限公司 | 一种语音在线报事报单管理方法、装置、介质及终端设备 |
CN110909134A (zh) * | 2018-09-18 | 2020-03-24 | 奇酷互联网络科技(深圳)有限公司 | 语音转换的方法、移动终端和可读存储介质 |
CN110955770A (zh) * | 2019-12-18 | 2020-04-03 | 圆通速递有限公司 | 一种智能对话系统 |
CN111128173A (zh) * | 2019-12-31 | 2020-05-08 | 重庆和贯科技有限公司 | 基于语音识别技术的智能家居语音交互方法 |
CN111178807A (zh) * | 2019-12-30 | 2020-05-19 | 苏宁云计算有限公司 | 一种基于订单拨打电话的方法和系统 |
-
2020
- 2020-06-28 WO PCT/CN2020/098590 patent/WO2022000140A1/fr active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6577713B1 (en) * | 1999-10-08 | 2003-06-10 | Iquest Technologies, Inc. | Method of creating a telephone data capturing system |
US20160379230A1 (en) * | 2015-06-25 | 2016-12-29 | Alibaba Group Holding Limited | System, device, and method for making automatic calls |
CN110909134A (zh) * | 2018-09-18 | 2020-03-24 | 奇酷互联网络科技(深圳)有限公司 | 语音转换的方法、移动终端和可读存储介质 |
CN110489534A (zh) * | 2019-07-26 | 2019-11-22 | 美的置业集团有限公司 | 一种语音在线报事报单管理方法、装置、介质及终端设备 |
CN110955770A (zh) * | 2019-12-18 | 2020-04-03 | 圆通速递有限公司 | 一种智能对话系统 |
CN111178807A (zh) * | 2019-12-30 | 2020-05-19 | 苏宁云计算有限公司 | 一种基于订单拨打电话的方法和系统 |
CN111128173A (zh) * | 2019-12-31 | 2020-05-08 | 重庆和贯科技有限公司 | 基于语音识别技术的智能家居语音交互方法 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109710772B (zh) | 一种基于深度学习的问答库知识管理系统及其实现方法 | |
CN106295807B (zh) | 一种信息处理的方法及装置 | |
CN107329967B (zh) | 基于深度学习的问答系统以及方法 | |
CN111182162B (zh) | 基于人工智能的电话质检方法、装置、设备和存储介质 | |
CN111724908A (zh) | 基于机器人流程自动化rpa的疫情排查方法和装置 | |
US10157619B2 (en) | Method and device for searching according to speech based on artificial intelligence | |
CN104903954A (zh) | 使用基于人工神经网络的亚语音单位区分的说话人验证及识别 | |
CN109360550A (zh) | 语音交互系统的测试方法、装置、设备和存储介质 | |
CN111193834B (zh) | 基于用户声音特征分析的人机交互方法、装置和电子设备 | |
CN108932945A (zh) | 一种语音指令的处理方法及装置 | |
CN111724909A (zh) | 一种结合rpa和ai的疫情排查方法和装置 | |
CN110135879A (zh) | 基于自然语言处理的客服质量自动评分方法 | |
CN111694940A (zh) | 一种用户报告的生成方法及终端设备 | |
CN113297365B (zh) | 一种用户意向判定方法、装置、设备及存储介质 | |
CN111180025B (zh) | 表示病历文本向量的方法、装置及问诊系统 | |
CN106776832A (zh) | 用于问答交互日志的处理方法、装置及系统 | |
CN111581074A (zh) | 通话场景覆盖测试方法、装置、计算机设备及存储介质 | |
Wagner et al. | Applying cooperative machine learning to speed up the annotation of social signals in large multi-modal corpora | |
WO2022000141A1 (fr) | Procédé d'investigation épidémique utilisant une automatisation de processus robotique (arp), et dispositif | |
WO2022000140A1 (fr) | Méthode de dépistage épidémique et appareil combinant rpa avec ai | |
CN112434953A (zh) | 一种基于计算机数据处理的客服人员考核方法和装置 | |
CN117251581A (zh) | 一种基于文本分析的设备故障信息诊断方法 | |
Glass et al. | Data collection and performance evaluation of spoken dialogue systems: the MIT experience. | |
CN114186048A (zh) | 基于人工智能的问答回复方法、装置、计算机设备及介质 | |
CN114462376A (zh) | 基于rpa和ai的庭审笔录生成方法、装置、设备及介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20943247 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 10/05/2023) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20943247 Country of ref document: EP Kind code of ref document: A1 |