CN112214654A - Universal intelligent question-answering automatic operation and maintenance system and method - Google Patents
Universal intelligent question-answering automatic operation and maintenance system and method Download PDFInfo
- Publication number
- CN112214654A CN112214654A CN202011117491.XA CN202011117491A CN112214654A CN 112214654 A CN112214654 A CN 112214654A CN 202011117491 A CN202011117491 A CN 202011117491A CN 112214654 A CN112214654 A CN 112214654A
- Authority
- CN
- China
- Prior art keywords
- answering
- intelligent question
- question
- intelligent
- data
- 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.)
- Pending
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 title claims description 20
- 230000005540 biological transmission Effects 0.000 claims abstract description 15
- 238000013523 data management Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims description 45
- 238000012545 processing Methods 0.000 claims description 39
- 238000012360 testing method Methods 0.000 claims description 27
- 230000000694 effects Effects 0.000 claims description 23
- 238000011156 evaluation Methods 0.000 claims description 22
- 238000000605 extraction Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 6
- 230000003321 amplification Effects 0.000 claims description 4
- 238000013144 data compression Methods 0.000 claims description 4
- 230000006837 decompression Effects 0.000 claims description 4
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 4
- 238000004806 packaging method and process Methods 0.000 claims description 4
- 230000001419 dependent effect Effects 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 238000005538 encapsulation Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 abstract description 53
- 238000012800 visualization Methods 0.000 abstract description 3
- 238000013135 deep learning Methods 0.000 abstract description 2
- 238000010801 machine learning Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 238000007726 management method Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000003058 natural language processing Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000001667 episodic effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
- G06F16/90332—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Physics (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a universal intelligent question-answering automatic operation and maintenance system, which comprises an intelligent question-answering function client unit, an intelligent question-answering function server unit, an intelligent question-answering data transmission service portal and an intelligent question-answering data management unit; the intelligent question-answering data transmission service portal is respectively connected with the intelligent question-answering function client unit, the intelligent question-answering function server unit and the intelligent question-answering data management unit; the intelligent question-answering function server unit is connected with the intelligent question-answering data management unit. The invention can effectively improve the automatic operation and maintenance efficiency of intelligent question answering and realize the visualization of the machine learning deep learning right task.
Description
Technical Field
The invention relates to the field of artificial intelligent question answering and automatic operation and maintenance, in particular to a universal intelligent question answering automatic operation and maintenance system and a universal intelligent question answering automatic operation and maintenance method.
Background
Under the background of big data era, due to the coming of a new wave of artificial intelligence wave in this year, the related technology of intelligent question and answer is applied to various industries and fields, and along with the development of natural language processing technology, the research of an intelligent question and answer system combining the natural language processing technology has been widely developed. Various research and product forms are various, and the research and product forms can be mainly divided into voice assistants, chat robots, machine partners and the like. The voice assistant has been exploded for some time, but then gradually cools over a period of one or two years because the technology is not mature and the freshness decreases after some time of use by the user, naturally breaking away from the use viscosity. At present, chat robot products are basically mature and have respective characteristics, but problems with simple answers can be freely answered basically, the performance of the chat robot is not good for slightly complex problems, and particularly the use experience of the chat robot is not satisfactory for professional problems. Whether it is a voice assistant, a chat robot or a machine partner, there is currently a dilemma.
Disclosure of Invention
In view of this, the present invention provides a general intelligent question-answering automation operation and maintenance system and method, which can effectively reduce the time cost and labor cost of intelligent question-answering.
In order to achieve the purpose, the invention adopts the following technical scheme:
a universal intelligent question-answering automatic operation and maintenance system comprises an intelligent question-answering function client unit, an intelligent question-answering function server unit, an intelligent question-answering data transmission service portal and an intelligent question-answering data management unit; the intelligent question-answering data transmission service portal is respectively connected with the intelligent question-answering function client unit, the intelligent question-answering function server unit and the intelligent question-answering data management unit; the intelligent question-answering function server unit is connected with the intelligent question-answering data management unit.
Furthermore, the intelligent question-answering function client unit is used for packaging related resources of the user intelligent question-answering service process and submitting the resources to the intelligent question-answering function server unit for processing;
the intelligent question-answering function server unit is used for receiving the request and data of the client, performing one-to-one question-answering service, issuing the intelligent question-answering service of the user version aiming at the user data, and providing the one-to-one question-answering service of the intelligent question-answering service of the user version and the system version;
the intelligent question-answer data transmission service portal is used for receiving client requests in a centralized mode, processing the requests by using appropriate service components according to the types of the client requests and responding to the requests.
The intelligent question-answering data management unit is used for processing related resources of the intelligent question-answering service process of the user in a centralized mode, and the processing mode includes but is not limited to original data type identification, structured document question-answering pair extraction, unstructured document question-answering pair extraction, data duplication removal, data amplification, data distributed storage and reading and writing, data decompression and data compression.
Further, the related resources include user information, data, algorithms, parameters, and rules.
Furthermore, the intelligent question-answering function client unit comprises a user request packaging module, an intelligent question-answering function issuing resource module and an intelligent question-answering single question-answering processing module.
Furthermore, the intelligent question-answering data transmission service portal is accessed by adopting an hdfs protocol, supports three data serialization protocols of XML, JSON and ProtoBuf, supports two modes of synchronous processing and asynchronous processing, automatically converts the request which needs to be processed for a long time by the intelligent question-answering function server unit into asynchronous processing, and queries the processing result through state updating or other operations by the user.
Furthermore, the intelligent question-answering function server unit comprises an intelligent question-answering model training component, an intelligent question-answering mode recommending component, an intelligent question-answering model automatic optimizing component, an intelligent question-answering effect evaluating component, an intelligent question-answering data automatic managing component, an intelligent question-answering candidate set automatic updating component and an intelligent question-answering control panel.
Further, the intelligent question-answering effect automatic evaluation component: generating one or more models through training, using a default test data set according to an intelligent question-answering mode if the data transmitted by the client does not contain the test data set according to the test data set transmitted by the client, supporting single-model effect evaluation and multi-model effect evaluation, using the default test data set for the test data set transmitted by the client and each intelligent question-answering mode by the evaluation data set, wherein the evaluation modes include but are not limited to loss, auc, accuracy, recall rate and F1 value; evaluation situations include, but are not limited to, textual displays, tabular displays, line graphs, graphical graphs, bar graphs, and other graphical displays.
Further, the intelligent question-answering model automatic optimization component comprises a model optimization module, a model application module and a question answering service publishing module.
Furthermore, the intelligent question-answering service module supports the loading and prediction of three question-answering models, namely a retrieval type question-answering model, a map question-answering model and a generation type question-answering model; and predicting the input question sentence, and returning the result to the client again.
An operation and maintenance method of a universal intelligent question-answering automatic operation and maintenance system comprises the following steps:
step S1: a user encapsulates related intelligent question-answering resources through an intelligent question-answering function client unit and sends a request to an intelligent question-answering function server unit service portal;
step S2: the intelligent question-answering function server unit service portal receives the corresponding request, starts an asynchronous processing mechanism, forwards the request to the intelligent question-answering task assignment component for scheduling processing, stores the request into a database if the request has data, and simultaneously returns a serial number and state information for scheduling processing of the client;
step S3: according to the assignment result of the intelligent question-answering task assignment component, determining to perform intelligent question-answering mode recommendation or intelligent question-answering service, wherein the intelligent question-answering mode recommendation is performed in the step 4, and the intelligent question-answering service is performed in the step 8;
step S4: automatically judging an optimal intelligent question-answering mode according to a data set sent by the intelligent question-answering function client unit, and adopting a specified intelligent question-answering mode if the intelligent question-answering mode is specified by the data of the intelligent question-answering function client unit;
step S5: reading training data in a database under a selected intelligent question-answering mode, segmenting the data, starting to train an intelligent question-answering model according to a training parameter set in intelligent question-answering function client unit data, storing the training model after the training is finished, and generating question-answering dependent data;
step S6: performing question answering effect evaluation on the generated model through a test set in a test parameter set database in the intelligent question answering function client unit data to further obtain an optimal intelligent question answering model;
step S7: loading the selected optimal intelligent question-answering model to an intelligent question-answering service for calling when the service carries out intelligent question-answering;
step S8: the intelligent question-answering service obtains answers for the question data in the intelligent question-answering function client unit data through the prediction of the intelligent question-answering model, and returns the answers to the intelligent question-answering function client unit;
step S9: the intelligent question-answering function client unit supports intelligent question-answering effect feedback, and the intelligent question-answering candidate set automatic updating component can automatically evaluate whether to update the candidate set or not according to a feedback result.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention can effectively reduce the time cost and the labor cost of intelligent question answering;
2. the invention can effectively improve the automatic operation and maintenance efficiency of intelligent question answering and realize the visualization of the machine learning deep learning right task.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
Referring to fig. 1, the present invention provides a general intelligent question-answering automated operation and maintenance system, which includes an intelligent question-answering function client unit, an intelligent question-answering function server unit, an intelligent question-answering data transmission service portal, and an intelligent question-answering data management unit; the intelligent question-answering data transmission service portal is respectively connected with the intelligent question-answering function client unit, the intelligent question-answering function server unit and the intelligent question-answering data management unit; the intelligent question-answering function server unit is connected with the intelligent question-answering data management unit.
The intelligent question-answering function client unit is used for packaging relevant resources of the user intelligent question-answering service process, including user information, data, algorithms, parameters and rules, and submitting the resources to the intelligent question-answering function server unit for processing;
the intelligent question-answering function server unit is used for receiving the request and data of the client, performing one-to-one question-answering service, issuing the intelligent question-answering service of the user version aiming at the user data, and providing the one-to-one question-answering service of the intelligent question-answering service of the user version and the system version;
the intelligent question-answer data transmission service portal is used for receiving client requests in a centralized mode, processing the requests by using appropriate service components according to the types of the client requests and responding to the requests.
The intelligent question-answering data management unit is used for processing related resources of the intelligent question-answering service process of the user in a centralized mode, and the processing mode includes but is not limited to original data type identification, structured document question-answering pair extraction, unstructured document question-answering pair extraction, data duplication removal, data amplification, data distributed storage and reading and writing, data decompression and data compression.
Preferably, in this embodiment, the smart question-and-answer function client unit includes a user request encapsulation module, a smart question-and-answer function publishing resource module, and a smart question-and-answer single-question-and-answer processing module.
The user packages the request to be processed, and taking a newly-built intelligent question-answer as an example, the user needs to package three parts of information: the first is the meta information of the question and answer user, including unique ID, name, description, creator, creation time, authority, etc.; secondly, the resources related to the intelligent question answering function release comprise: training data set, test data set, question-answer mode type (optional, ask question-answer, generate question-answer), training parameter set (round number, maximum sentence length, etc.), prediction parameter set; thirdly, the intelligent question answering single question answering relates to data including meta-questions and answer effect feedback. Besides the intelligent question-answering data information, the client also supports the sending of requests such as training state updating, training parameter recommendation, customized candidate library addition, test evaluation result acquisition and the like.
Preferably, in this embodiment, the intelligent question and answer data transmission service portal uses hdfs protocol for access, supports three data serialization protocols of XML, JSON, and ProtoBuf, and supports two modes of synchronous processing and asynchronous processing, and for a request that needs to be processed for a long time by the intelligent question and answer function server unit, the portal automatically changes to asynchronous processing, and a user queries a processing result through state update or other operations.
Preferably, in this embodiment, the intelligent question-answering task dispatching component: the method comprises two assignment schemes, wherein one scheme is a scheme issued by model training and is used for transmitting data sent by a client to an intelligent question-answering mode recommendation component; one is a scheme for directly answering questions asked by a client, and question data sent by the client is transmitted to an intelligent question-answering service.
Preferably, in this embodiment, the intelligent question-answering mode recommending component: except for recommending an optimal default question-answer mode according to the question-answer data type transmitted by the client, the optimal default question-answer mode comprises a retrieval type question-answer mode, a map question-answer mode, a generation type question-answer mode and the like, for example: the question and answer data uploaded by the user are text files, table files, json files and the like formed by question and answer pairs, and a preferred retrieval type question and answer mode is recommended; the question and answer data uploaded by the user are table files, json files and the like formed by table statistical information, and preferably map question and answer is recommended; but also includes, but is not limited to, data format recognition, data format conversion, data grouping (training set, validation set, test set), and the like.
Preferably, in this embodiment, the intelligent question-answering model training component is configured to train a parameter set according to the analyzed model, and if the parameter set does not include a question-answering mode, recommend the component to select more question-answering modes according to the intelligent question-answering mode, and implement automation of the training process through automatic training control, where the obtained trained model supports download output, database storage, database invocation, and the like.
The training is automatically controlled: including but not limited to training progress feedback, training effect feedback, training early-stop, episodic model output, and the like.
Preferably, the training data set and the testing data set both support transmission in a plurality of data formats, including but not limited to text files (txt), compressed files (rar, zip, etc.), table files (csv, tsv, xlsx, etc.), json files, etc. The parameter types of the training parameter set and the prediction parameter set include, but are not limited to: floating point, integer, boolean.
In this embodiment, the intelligent question-answering function server unit includes an intelligent question-answering model training component, an intelligent question-answering mode recommending component, an intelligent question-answering model automatic optimizing component, an intelligent question-answering effect evaluating component, an intelligent question-answering data automatic managing component, an intelligent question-answering candidate set automatic updating component, and an intelligent question-answering control panel. The intelligent question-answering function server unit further comprises training parameter analysis, distributed cluster management, distributed training, task scheduling and the like.
In this embodiment, preferably, the intelligent question-answering task dispatching component enters different task modes according to metadata requested by a client;
the intelligent question-answering model training component selects a model training mode according to the type of a client request, performs model training, and supports custom processing on the model training according to the training parameters of the client request;
the intelligent question-answer mode recommending component is used for recommending an optimal default question-answer mode according to the type of question-answer data transmitted by a client, and the optimal default question-answer mode comprises a search-type question-answer, a map question-answer, a generating-type question-answer and the like;
the intelligent question-answering model automatic application component is used for optimizing a model generated by training;
the intelligent question-answering effect evaluation component is used for testing and evaluating the effect of a model generated by training;
the intelligent question-answer data automatic management component is used for automatically preprocessing data uploaded by a client and data generated by intelligent question-answer, and reading, writing and storing the data in a database form;
the intelligent question-answer candidate set automatic updating component is used for automatically determining whether to update a candidate set required in a search-type intelligent question-answer according to a request of a client;
the intelligent question-answering control panel manages information including client request conditions, parameter simulation evaluation comparison visualization, simulation learning conditions and the like.
In this embodiment, preferably, the intelligent automatic question-answering effect evaluation component: generating one or more models through training, using a default test data set according to an intelligent question-answering mode if the data transmitted by the client does not contain the test data set according to the test data set transmitted by the client, supporting single-model effect evaluation and multi-model effect evaluation, using the default test data set for the test data set transmitted by the client and each intelligent question-answering mode by the evaluation data set, wherein the evaluation modes include but are not limited to loss, auc, accuracy, recall rate and F1 value; assessment situations include, but are not limited to, text displays, tabular displays, line graphs, and bar graph displays.
In this embodiment, preferably, the intelligent question-answering model automatic application component: the method is specifically divided into model optimization, model application and question answering service release (model prediction output only): according to the intelligent question-answering effect automatic evaluation component, an optimal model in a plurality of models is generated through optimal training, self-selection and default options of ranking indexes are supported, answers of input questions are predicted through the models, colleagues support the prediction function of the models to be changed into a service mode, the models can be issued to be independent, and the secondary function supports intelligent question-answering service calling.
The intelligent question-answering service comprises the following steps: the method supports the loading and prediction of three question-answer models, namely a retrieval type question-answer model, a map question-answer model and a generation type question-answer model; and predicting the input question sentence, and returning the result to the client again.
In this embodiment, preferably, the intelligent question-answering data automatic management component: the method comprises the steps of automatically processing training data and test data (optional) uploaded by a client into a form required by model training, wherein the processing mode comprises but is not limited to original data type identification, structured document question-answer pair extraction, unstructured document question-answer pair extraction, data duplication removal (optional), data amplification (optional), data distributed storage and reading and writing, data decompression and data compression.
The intelligent question-answering model automatic application component comprises a model optimization module, a model application module and a question answering service issuing module.
In this embodiment, preferably, the intelligent question-answer candidate set automatic updating component: the method comprises the steps of automatically updating a question candidate set which is depended on by a query question and answer, wherein the updating operation comprises but is not limited to adding question and answer pairs, deleting question and answer pairs and replacing question and answer pairs (question sentences and answer sentences), when the prediction effect of a query question and answer model is not ideal, a system collects user feedback, and the intelligent question and answer candidate set is automatically started to be updated.
Preferably, in this embodiment, an operation and maintenance method of a general intelligent question and answer automation operation and maintenance system is further provided, which includes the following steps:
step S1: a user encapsulates related intelligent question-answering resources through an intelligent question-answering function client unit and sends a request to an intelligent question-answering function server unit service portal;
step S2: the intelligent question-answering function server unit service portal receives the corresponding request, starts an asynchronous processing mechanism, forwards the request to the intelligent question-answering task assignment component for scheduling processing, stores the request into a database if the request has data, and simultaneously returns a serial number and state information for scheduling processing of the client;
step S3: according to the assignment result of the intelligent question-answering task assignment component, determining to perform intelligent question-answering mode recommendation or intelligent question-answering service, wherein the intelligent question-answering mode recommendation is performed in the step 4, and the intelligent question-answering service is performed in the step 8;
step S4: automatically judging an optimal intelligent question-answering mode according to a data set sent by the intelligent question-answering function client unit, and adopting a specified intelligent question-answering mode if the intelligent question-answering mode is specified by the data of the intelligent question-answering function client unit;
step S5: reading training data in a database under a selected intelligent question-answering mode, segmenting the data, starting to train an intelligent question-answering model according to a training parameter set in intelligent question-answering function client unit data, storing the training model after the training is finished, and generating question-answering dependent data;
step S6: performing question answering effect evaluation on the generated model through a test set in a test parameter set database in the intelligent question answering function client unit data to further obtain an optimal intelligent question answering model;
step S7: loading the selected optimal intelligent question-answering model to an intelligent question-answering service for calling when the service carries out intelligent question-answering;
step S8: the intelligent question-answering service obtains answers for the question data in the intelligent question-answering function client unit data through the prediction of the intelligent question-answering model, and returns the answers to the intelligent question-answering function client unit;
step S9: the intelligent question-answering function client unit supports intelligent question-answering effect feedback, and the intelligent question-answering candidate set automatic updating component can automatically evaluate whether to update the candidate set or not according to a feedback result.
The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (10)
1. A universal intelligent question-answering automatic operation and maintenance system is characterized by comprising an intelligent question-answering function client unit, an intelligent question-answering function server unit, an intelligent question-answering data transmission service portal and an intelligent question-answering data management unit; the intelligent question-answering data transmission service portal is respectively connected with the intelligent question-answering function client unit, the intelligent question-answering function server unit and the intelligent question-answering data management unit; the intelligent question-answering function server unit is connected with the intelligent question-answering data management unit.
2. The universal intelligent question-answering automation operation and maintenance system according to claim 1,
the intelligent question-answering function client unit is used for packaging related resources of the user intelligent question-answering service process and submitting the related resources to the intelligent question-answering function server unit for processing;
the intelligent question-answering function server unit is used for receiving the request and data of the client, performing one-to-one question-answering service, issuing the intelligent question-answering service of the user version aiming at the user data, and providing the one-to-one question-answering service of the intelligent question-answering service of the user version and the system version;
the intelligent question-answer data transmission service portal is used for intensively receiving a client request, processing the request by using a proper service component according to the type of the client request and responding to the request;
the intelligent question-answering data management unit is used for processing related resources of the intelligent question-answering service process of the user in a centralized mode, and the processing mode includes but is not limited to original data type identification, structured document question-answering pair extraction, unstructured document question-answering pair extraction, data duplication removal, data amplification, data distributed storage and reading and writing, data decompression and data compression.
3. The system of claim 1, wherein the related resources comprise user information, data, algorithms, parameters, and rules.
4. The system of claim 1, wherein the intelligent Q & A function client unit comprises a user request encapsulation module, an intelligent Q & A function publishing resource module, and an intelligent Q & A single Q & A processing module.
5. The system according to claim 1, wherein the intelligent question-answering data transmission service portal is accessed by using hdfs protocol, supports three data serialization protocols of XML, JSON and ProtoBuf, and supports two modes of synchronous processing and asynchronous processing, and the portal automatically converts to asynchronous processing for a request that the intelligent question-answering function server unit needs to process for a long time, and a user queries a processing result through state updating or other operations.
6. The system of claim 1, wherein the intelligent question-answering function server unit comprises an intelligent question-answering model training component, an intelligent question-answering mode recommending component, an intelligent question-answering model automatic optimizing component, an intelligent question-answering effect evaluating component, an intelligent question-answering data automatic managing component, an intelligent question-answering candidate set automatic updating component and an intelligent question-answering control panel.
7. The system of claim 6, wherein the intelligent automated question-answering operation and maintenance component: generating one or more models through training, using a default test data set according to an intelligent question-answering mode if the data transmitted by the client does not contain the test data set according to the test data set transmitted by the client, supporting single-model effect evaluation and multi-model effect evaluation, using the default test data set for the test data set transmitted by the client and each intelligent question-answering mode by the evaluation data set, wherein the evaluation modes include but are not limited to loss, auc, accuracy, recall rate and F1 value; assessment situations include, but are not limited to, text displays, tabular displays, line graphs, and bar graph displays.
8. The general intelligent question-answering automation operation and maintenance system as claimed in claim 6, wherein the intelligent question-answering model automatic optimization component comprises a model optimization module, a model application module and a question answering service issuing module.
9. The general intelligent question-answer automated operation and maintenance system according to claim 8, wherein the intelligent question-answer service module supports the loading and prediction of three question-answer models, namely a search-type question-answer model, a map question-answer model and a generation-type question-answer model; and predicting the input question sentence, and returning the result to the client again.
10. An operation and maintenance method of a universal intelligent question-answering automatic operation and maintenance system is characterized by comprising the following steps:
step S1: a user encapsulates related intelligent question-answering resources through an intelligent question-answering function client unit and sends a request to an intelligent question-answering function server unit service portal;
step S2: the intelligent question-answering function server unit service portal receives the corresponding request, starts an asynchronous processing mechanism, forwards the request to the intelligent question-answering task assignment component for scheduling processing, stores the request into a database if the request has data, and simultaneously returns a serial number and state information for scheduling processing of the client;
step S3: according to the assignment result of the intelligent question-answering task assignment component, determining to perform intelligent question-answering mode recommendation or intelligent question-answering service, wherein the intelligent question-answering mode recommendation is performed in the step 4, and the intelligent question-answering service is performed in the step 8;
step S4: automatically judging an optimal intelligent question-answering mode according to a data set sent by the intelligent question-answering function client unit, and adopting a specified intelligent question-answering mode if the intelligent question-answering mode is specified by the data of the intelligent question-answering function client unit;
step S5: reading training data in a database under a selected intelligent question-answering mode, segmenting the data, starting to train an intelligent question-answering model according to a training parameter set in intelligent question-answering function client unit data, storing the training model after the training is finished, and generating question-answering dependent data;
step S6: performing question answering effect evaluation on the generated model through a test set in a test parameter set database in the intelligent question answering function client unit data to further obtain an optimal intelligent question answering model;
step S7: loading the selected optimal intelligent question-answering model to an intelligent question-answering service for calling when the service carries out intelligent question-answering;
step S8: the intelligent question-answering service obtains answers for the question data in the intelligent question-answering function client unit data through the prediction of the intelligent question-answering model, and returns the answers to the intelligent question-answering function client unit;
step S9: the intelligent question-answering function client unit supports intelligent question-answering effect feedback, and the intelligent question-answering candidate set automatic updating component can automatically evaluate whether to update the candidate set or not according to a feedback result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011117491.XA CN112214654A (en) | 2020-10-19 | 2020-10-19 | Universal intelligent question-answering automatic operation and maintenance system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011117491.XA CN112214654A (en) | 2020-10-19 | 2020-10-19 | Universal intelligent question-answering automatic operation and maintenance system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112214654A true CN112214654A (en) | 2021-01-12 |
Family
ID=74055769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011117491.XA Pending CN112214654A (en) | 2020-10-19 | 2020-10-19 | Universal intelligent question-answering automatic operation and maintenance system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112214654A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090125598A1 (en) * | 2007-11-08 | 2009-05-14 | Huawei Technologies Co., Ltd. | Method, client and server for implementing question and answer services |
| CN105630938A (en) * | 2015-12-23 | 2016-06-01 | 深圳市智客网络科技有限公司 | Intelligent question-answering system |
| CN107301213A (en) * | 2017-06-09 | 2017-10-27 | 腾讯科技（深圳）有限公司 | Intelligent answer method and device |
| CN109241258A (en) * | 2018-08-23 | 2019-01-18 | 江苏索迩软件技术有限公司 | A kind of deep learning intelligent Answer System using tax field |
| CN109446387A (en) * | 2018-10-09 | 2019-03-08 | 众蚁(上海)信息技术有限公司 | A kind of Owners Committee's intelligent Answer System based on artificial intelligence |
| CN110321420A (en) * | 2019-07-04 | 2019-10-11 | 河海大学常州校区 | The intelligent Answer System generated based on question sentence |
| CN111241237A (en) * | 2019-12-31 | 2020-06-05 | 中国建设银行股份有限公司 | Intelligent question and answer data processing method and device based on operation and maintenance service |
| CN111367633A (en) * | 2020-02-27 | 2020-07-03 | 深圳市腾讯信息技术有限公司 | Model service management method and device in question-answering system and computer equipment |
| CN111414461A (en) * | 2020-01-20 | 2020-07-14 | 福州大学 | Intelligent question-answering method and system fusing knowledge base and user modeling |
-
2020
- 2020-10-19 CN CN202011117491.XA patent/CN112214654A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090125598A1 (en) * | 2007-11-08 | 2009-05-14 | Huawei Technologies Co., Ltd. | Method, client and server for implementing question and answer services |
| CN105630938A (en) * | 2015-12-23 | 2016-06-01 | 深圳市智客网络科技有限公司 | Intelligent question-answering system |
| CN107301213A (en) * | 2017-06-09 | 2017-10-27 | 腾讯科技（深圳）有限公司 | Intelligent answer method and device |
| CN109241258A (en) * | 2018-08-23 | 2019-01-18 | 江苏索迩软件技术有限公司 | A kind of deep learning intelligent Answer System using tax field |
| CN109446387A (en) * | 2018-10-09 | 2019-03-08 | 众蚁(上海)信息技术有限公司 | A kind of Owners Committee's intelligent Answer System based on artificial intelligence |
| CN110321420A (en) * | 2019-07-04 | 2019-10-11 | 河海大学常州校区 | The intelligent Answer System generated based on question sentence |
| CN111241237A (en) * | 2019-12-31 | 2020-06-05 | 中国建设银行股份有限公司 | Intelligent question and answer data processing method and device based on operation and maintenance service |
| CN111414461A (en) * | 2020-01-20 | 2020-07-14 | 福州大学 | Intelligent question-answering method and system fusing knowledge base and user modeling |
| CN111367633A (en) * | 2020-02-27 | 2020-07-03 | 深圳市腾讯信息技术有限公司 | Model service management method and device in question-answering system and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210232761A1 (en) | Methods and systems for improving machine learning performance | |
| US11030230B2 (en) | System and method for providing technology assisted data review with optimizing features | |
| RU2694001C2 (en) | Method and system for creating a parameter of quality forecast for a forecasting model performed in a machine learning algorithm | |
| CN110362667B (en) | Intelligent customer service method, device, equipment and readable storage medium | |
| CN116737129B (en) | Supply chain control tower generation type large language model and construction method thereof | |
| CN115130065B (en) | Method, device and equipment for processing characteristic information of supply terminal and computer readable medium | |
| CN110069573A (en) | Product data integration method, apparatus, computer equipment and storage medium | |
| CN112035325A (en) | Automatic monitoring method and device for text robot | |
| US20220083580A1 (en) | Information processing apparatus and information processing method | |
| CN111353290B (en) | Method and system for automatically responding to user inquiry | |
| CN114175007A (en) | Active learning for data matching | |
| CN116757270A (en) | Data processing method and server based on man-machine interaction model or large model | |
| CN109819019B (en) | Monitoring and statistical analysis method and system for large-scale network data acquisition | |
| CN118035425A (en) | Interaction method and device based on natural language model, electronic equipment and medium | |
| CN116521653A (en) | Food material question-answering method and system based on knowledge graph | |
| CN107832342B (en) | Robot chatting method and system | |
| CN112214654A (en) | Universal intelligent question-answering automatic operation and maintenance system and method | |
| Kaviya et al. | Artificial intelligence based farmer assistant chatbot | |
| CA3081825A1 (en) | Interactive guidance system for selecting thermodynamics methods in process simulations | |
| WO2020106950A1 (en) | User-experience development system | |
| CN111538822B (en) | Method and system for generating training data of intelligent customer service robot | |
| CN116976294B (en) | Method and system for realizing automatic filling of complex electronic forms | |
| JP7493195B1 (en) | Program, method, information processing device, and system | |
| CN118332097B (en) | Information interaction method and device | |
| CN115827171B (en) | Cloud parameter adjusting system, parameter adjusting method and parameter adjusting system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210112 |
|
| RJ01 | Rejection of invention patent application after publication |