CN116128546A - AI public opinion monitoring system and method for external service window in power industry - Google Patents
AI public opinion monitoring system and method for external service window in power industry Download PDFInfo
- Publication number
- CN116128546A CN116128546A CN202310019094.6A CN202310019094A CN116128546A CN 116128546 A CN116128546 A CN 116128546A CN 202310019094 A CN202310019094 A CN 202310019094A CN 116128546 A CN116128546 A CN 116128546A
- Authority
- CN
- China
- Prior art keywords
- module
- public opinion
- intelligent
- data
- sensitive
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 26
- 238000010801 machine learning Methods 0.000 claims abstract description 18
- 238000006243 chemical reaction Methods 0.000 claims description 21
- 238000004891 communication Methods 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 10
- 230000000007 visual effect Effects 0.000 claims description 8
- 238000012423 maintenance Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000005065 mining Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 3
- 238000011161 development Methods 0.000 claims description 3
- 230000018109 developmental process Effects 0.000 claims description 3
- 238000003058 natural language processing Methods 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 2
- 230000006872 improvement Effects 0.000 description 8
- 238000001514 detection method Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/338—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/355—Class or cluster creation or modification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
-
- 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
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Engineering & Computer Science (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Probability & Statistics with Applications (AREA)
- Computing Systems (AREA)
- Tourism & Hospitality (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Human Computer Interaction (AREA)
- Public Health (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Telephonic Communication Services (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses an AI public opinion monitoring system of an external service window in the power industry, which comprises an intelligent acquisition module, a high-performance edge server and a central server, wherein the intelligent acquisition module is used for acquiring audio data of the external service window and comprises a user information acquisition module and a speech prediction module; according to the invention, the BI module is deployed through the central server, so that public opinion information of external service offices can be efficiently monitored in real time, public opinion hotspots are intelligently analyzed, common people's appeal are timely known, historical data are deeply mined, recognition conditions are automatically summarized, and public opinion hotspots and trends are researched and judged; meanwhile, self-learning and self-training are realized through the machine learning module, training materials are continuously enriched, analysis accuracy is continuously improved, and data support is provided for reasonable decisions.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to an AI public opinion monitoring system and method for an external service window in the power industry.
Background
The electric power business hall is the forefront of electric power marketing business operation, can embody the business wish of common people, is the best place of grasping user satisfaction in real time, knowing the folk and the people's mind, gathering user's demand. Meanwhile, the electric power business hall is distributed in various places and is in a scattered multi-point state, and from the management perspective, the business hall is positioned at the management terminal. Too much manpower is unlikely to be input to the scene to collect common people's will uninterruptedly, and the public opinion cannot be monitored and tracked in real time through the manpower;
in addition, the public opinion monitoring has larger technical difficulty, higher requirements on real-time performance, hardware facilities and the like, is not easy to develop, can not save historical public opinion data, and is more deeply analyzed, so that the intention of common people can not be mastered, higher quality service can not be provided in a targeted manner, and data support can not be provided for more reasonable decision making.
Disclosure of Invention
Aiming at the problems, the invention aims to provide an AI public opinion monitoring system and an AI public opinion monitoring method of an external service window in the power industry, wherein a central server of the AI public opinion monitoring system of the external service window in the power industry deploys a BI module to efficiently monitor public opinion information of external service office in real time, intelligently analyze public opinion hotspots, timely know common people's appeal, deeply mine historical data, automatically summarize identification conditions and research and judge public opinion hotspots and trends; meanwhile, self-learning and self-training are realized through the machine learning module, training materials are continuously enriched, analysis accuracy is continuously improved, and data support is provided for reasonable decisions.
In order to achieve the purpose of the invention, the invention is realized by the following technical scheme: the utility model provides an AI public opinion monitoring system and method of power industry external service window, includes intelligent data acquisition module, high performance edge server and central server, intelligent acquisition module is used for gathering external service window's audio data, intelligent acquisition module includes user information acquisition module and speech art prediction module, high performance edge server is realized based on the edge service technique, has not only improved system response speed by a wide margin, has reduced the system and has relied on network resource, and edge server simple to operate does not receive the space restriction, high performance edge server includes intelligent speech conversion module, intelligent text retrieval module and machine learning module, central server includes BI module and sensitive word stock management module.
The further improvement is that: the user information acquisition module is used for acquiring user basic information of communication and history communication records of the user according to an external service dialogue window, wherein the user basic information comprises information such as gender, age, cultural level, history occupation and the like, and language style and character bias of the user are judged by the basic information.
The further improvement is that: the conversation prediction module comprises a conversation analysis module and a solution proposal module, wherein the conversation analysis module is used for analyzing voice data of a user communication calendar and carrying out conversation prediction, and the solution proposal module is used for automatically prompting dialogue reply proposal according to the communication records of the same type.
The further improvement is that: the intelligent voice conversion module is used for converting collected voice data into characters, and the intelligent character retrieval module is used for identifying sensitive words in the converted characters based on an artificial intelligent natural language processing technology, so that the words can be efficiently segmented and accurately retrieved, sensitive words preset by a system can be accurately retrieved, and the detection omission and false detection are avoided to the greatest extent; the machine learning module automatically expands the sensitive word stock in the process of identifying the sensitive words based on the artificial intelligent machine self-learning model, and can simultaneously use the identified (sensitive) alarm data as training data in the operation process, so that the training materials are continuously expanded, the self-training and the self-improvement can be realized, and the accuracy is continuously improved.
The further improvement is that: the BI module is used for comprehensively analyzing the identified sensitive words and results, can comprehensively and accurately count and analyze the historical public opinion data, presents the historical public opinion data to a user in visual expression forms such as a statistics table and a chart, predicts future trends according to the historical statistics data, gathers and classifies hot spots of public opinion, focus problems of common people concern and the like, and provides data support for management decision making of a manager; the sensitive word stock management module is used for maintaining a business handling process and providing sensitive words required by public opinion monitoring.
The further improvement is that: the sensitive word stock management module further comprises a sensitive word stock maintenance module, an intelligent recognition adding module and a sensitive word classification module, wherein the sensitive word stock maintenance module is used for increasing or decreasing sensitive words and adjusting the sensitive words, the intelligent recognition adding module is used for automatically checking the change of the sensitive word stock and dynamically adding the recognized sensitive words into a recognition algorithm, and the sensitive word classification module is used for classifying the sensitive words in multiple fields.
The further improvement is that: the BI module further comprises an intelligent prediction module and a data conversion display module, wherein the intelligent prediction module is used for intelligently analyzing and predicting public opinion trends based on the occurrence frequency of public opinion sensitive words, and the data conversion display module is used for converting predicted and analyzed public opinion trend results into a concise and visual chart form for display.
An AI public opinion monitoring method for an external service window in the power industry comprises the following steps:
step one, data acquisition, an intelligent data acquisition module acquires audio data of an external service site in real time and transmits the audio data to a high-performance edge server;
step two, conversion analysis, namely converting audio data into text data by an intelligent voice conversion module in the high-performance edge server, performing word segmentation and monitoring by utilizing an intelligent text retrieval module, and simultaneously calling data of a sensitive word stock for recognition analysis;
thirdly, training is established by the model, a machine learning model is established, history data is used for pre-training, the machine learning model is put into use after pre-training, and monitoring data are used as training materials for self-training while monitoring, analysis and recognition are carried out;
step four, actively prompting, namely when the model of the high-performance edge server identifies the sensitive word, sending out a warning prompt at the first time, and processing by a manager according to the warning prompt;
and fifthly, public opinion analysis output, namely mining historical data through a BI module of the central server, analyzing the identified sensitive words, and converting an analysis result into a visual chart form to display through a data conversion display module.
The further improvement is that: and step two, the sensitive word stock is based on a sensitive word stock management module of the central server, and the sensitive word stock is updated and expanded at fixed time according to the development of the service.
The further improvement is that: in the fourth step, corresponding processing scheme suggestions are given through the speaking prediction module when the sensitive word is detected, and reasonable language reply suggestions are given after speaking audios are recognized through the speaking prediction module when the sensitive word is not detected.
The beneficial effects of the invention are as follows: the BI module is deployed in the central server, so that public opinion information of external service offices can be efficiently monitored in real time, public opinion hotspots are intelligently analyzed, common people's appeal are timely known, historical data are deeply mined, recognition conditions are automatically summarized, and public opinion hotspots and trends are researched and judged; meanwhile, self-learning and self-training are realized through the machine learning module, training materials are continuously enriched, analysis accuracy is continuously improved, the problems existing in the prior art are effectively solved, and data support is provided for reasonable decisions.
Drawings
FIG. 1 is a system architecture diagram of an embodiment 1 of the present invention.
FIG. 2 is a flow chart of the method of embodiment 2 of the present invention.
Detailed Description
The present invention will be further described in detail with reference to the following examples, which are only for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
Example 1
According to the embodiment shown in fig. 1, an AI public opinion monitoring system of an external service window in the power industry is provided, which comprises an intelligent data acquisition module, a high-performance edge server and a central server, wherein the intelligent acquisition module is installed in the external service window in the power industry and is used for acquiring audio data of the external service window and sending the acquired data to the high-performance edge server in real time, and the intelligent acquisition module comprises a user information acquisition module and a speech prediction module;
the high-performance edge server is realized based on an edge service technology, has the characteristics of high speed, small volume, easiness in installation and the like, is internally provided with a linux system, has high safety, is installed in an external business hall in the power industry, is connected with an intelligent data acquisition device, and is connected with a central server to bear the calculation task of on-site public opinion monitoring and identification; the high-performance edge server comprises an intelligent voice conversion module, an intelligent text retrieval module and a machine learning module;
the edge computing technology is adopted, computing power is deployed at the forefront end (an electric power business hall window) of system detection, data field acquisition and field analysis are carried out, so that system calculation is more efficient, all original data are not required to be sent to a central server, only identification results are required to be sent, network blocking caused by data transmission is greatly reduced, the pressure of the system on the whole network is reduced, in addition, all computing power is distributed to the edge server, the dependence of the system on the central server is greatly reduced, the pressure of the central server is reduced, and the system is more stable, efficient and robust.
The central server comprises a BI module and a sensitive word stock management module, the central server is arranged in a central machine room of a power grid management unit and used for bearing the calculation work of the BI module of the system, the BI module is deployed, the intelligent data acquisition module, the high-performance edge server and the central server are connected through an electric power intranet, and the whole system operates in the electric power intranet; by automatically collecting real-time voice information of a business hall, the natural language analysis technology is used for analyzing and identifying voice contents, automatically identifying keywords appearing in voice and accurately analyzing public opinion of a business handling window.
The user information acquisition module is used for acquiring user basic information of communication and history communication records of the user according to the external service dialogue window, wherein the user basic information comprises information such as gender, age, cultural level, history occupation and the like, and language style and character bias of the user are judged by the basic information.
The conversation prediction module comprises a conversation analysis module and a solution proposal module, wherein the conversation analysis module is used for analyzing the voice data of the user communication calendar and performing conversation prediction, and the solution proposal module is used for automatically prompting dialogue reply proposal according to the history same type communication record.
The intelligent voice conversion module is used for converting the collected voice data into characters; the intelligent text retrieval module recognizes sensitive (word) words in the converted text based on an artificial intelligent natural language processing technology, is deployed on a high-performance edge server, can access a system sensitive word library management module, reads sensitive (word) words set by a system, compares and retrieves the sensitive (word) words with data sent by the intelligent data acquisition module, and recognizes the sensitive (word) words in real time;
by adopting an artificial intelligent natural language technology, the system word segmentation supports an accurate mode, a full mode and a search engine mode, in addition, the system supports the PaddlePaddle mode by utilizing a PaddlePaddle deep learning framework, the system sets reasonable word segmentation specifications, the problems of ambiguity segmentation, unregistered word recognition and the like in the word segmentation process are effectively solved, the system marks and recognizes from multiple angles such as words, word frequencies, parts of speech and the like, efficient word segmentation and accurate search are achieved, the preset sensitive words of the system can be accurately searched, and the detection omission and false detection are avoided to the greatest extent.
The machine learning module expands the sensitive word library automatically in the process of recognizing the sensitive word based on the artificial intelligent machine self-learning model, is deployed on a high-performance edge server, and adopts the artificial intelligent machine learning technology to enable the system to have self-training and self-improving capabilities and continuously improve the recognition accuracy of the system.
The BI module is used for comprehensively analyzing the identified sensitive (word) words and results, belongs to a central server, is used for storing and reusing data generated in the public opinion monitoring process, and is used for mining the value of the data based on the data; the sensitive word library management module is used for maintaining a business handling process, providing sensitive (word) words required by public opinion monitoring, deploying the sensitive (word) words in an edge server, and maintaining the sensitive (word) words required by the business handling process and the public opinion monitoring, wherein the functions comprise adding, deleting, modifying, searching and the like of the sensitive words.
The sensitive word stock management module also comprises a sensitive word stock maintenance module, an intelligent recognition adding module and a sensitive word classification module, wherein the sensitive word stock maintenance module is used for increasing or reducing sensitive (word) words and adjusting the sensitive (word) words, the intelligent recognition adding module is used for automatically checking the change of the sensitive (word) word stock and dynamically adding the recognized sensitive (word) words into a recognition algorithm, and the sensitive word classification module is used for classifying the sensitive (word) words in multiple fields; the user can dynamically increase or decrease or modify the sensitive words, the sensitive word library of the system is tightly combined with the actual business of the power industry, the sensitive words frequently appearing in the business handling process of the service window are precisely summarized, the business pertinence is high, the boundaries of the business range are provided, and the flexible strain strategy is provided, so that the system is a common sensitive dictionary specially applied to the power industry.
The BI module also comprises an intelligent prediction module and a data conversion display module, wherein the intelligent prediction module is used for intelligently analyzing and predicting public opinion trends based on the occurrence frequency of public opinion sensitive (word) words, and the data conversion display module is used for converting the predicted and analyzed public opinion trend results into a concise and visual chart form for display;
the BI technology is introduced, history monitoring data are fully applied, the past public opinion information is deeply mined, the public opinion hotspots, common surrogates and the like are comprehensively analyzed according to the word frequency appearing in the public opinion, intelligent prediction is carried out on public opinion trends, a simple and visual chart is output by the system, all analysis results are displayed, and the management personnel can easily understand the analysis results to provide data support for making corresponding management decisions.
By the deployment of the system, an edge computing technology is adopted, all original data are not required to be completely transmitted to a central server, and only identification results are required to be transmitted, so that network blocking caused by data transmission is greatly reduced, the pressure of the system on the whole network is reduced, and meanwhile, the pressure of the central server is reduced; the system effectively solves the problems of ambiguity segmentation, unregistered word recognition and the like in the word segmentation process, can accurately search the preset sensitive words of the system, can intelligently predict and output public opinion trends, and converts and outputs a concise and visual chart so that management staff can understand the graphs more easily.
Example 2
According to the embodiment shown in fig. 2, the AI public opinion monitoring method for the external service window in the power industry includes the following steps:
step one, data acquisition, an intelligent data acquisition module acquires audio data of an external service site in real time and transmits the audio data to a high-performance edge server.
Step two, conversion analysis, namely converting audio data into text data by an intelligent voice conversion module in the high-performance edge server, performing word segmentation and monitoring by utilizing an intelligent text retrieval module, and simultaneously calling data of a sensitive (word) word stock for recognition analysis;
the sensitive word stock is based on a sensitive word stock management module of a central server, and is updated and expanded at fixed time according to the development of the service.
Thirdly, training is established by the model, a machine learning model is established, history data is used for pre-training, the machine learning model is put into use after pre-training, and monitoring data are used as training materials for self-training while monitoring, analysis and recognition are carried out;
the artificial intelligent deep learning technology is adopted, a machine learning model is firstly built, and the model is trained in a large data amount, so that the model identification reaches higher accuracy, the data identified during the system operation can be automatically saved and used as new training data, the trained sample data can be continuously expanded, the system can self-train and self-improve through the continuously-growing sample data, the performance of the model is continuously perfected, and the accuracy of the system identification is improved.
Step four, actively prompting, namely when the model of the high-performance edge server identifies a sensitive word, sending out a warning prompt at the first time, and processing by a manager according to the warning prompt;
and when the sensitive word is detected, a corresponding processing scheme suggestion is given through a speaking prediction module, and when the sensitive word is not detected, a reasonable language reply suggestion is given after speaking audio is identified through the speaking prediction module.
And fifthly, public opinion analysis output, namely mining historical data through a BI module of the central server, analyzing the identified sensitive (word) words, and converting the analysis result into an intuitive chart form to display through a data conversion display module.
The method can realize the on-site data acquisition and on-site analysis, so that the system calculation is more efficient, simultaneously, a machine learning model is adopted to train by using a large amount of data, and the model is self-trained in the running process, so that the performance of the model can be continuously improved, and the accuracy of system identification is improved; meanwhile, the BI module is adopted to fully apply historical monitoring data, deep mining is carried out on the prior public opinion information, intelligent prediction is carried out on public opinion trend, and a prediction result is converted into a chart to be displayed intuitively, so that management staff can understand the chart conveniently, measures can be taken more quickly, and data support is provided for making corresponding management decisions.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. An AI public opinion monitoring system of the external service window of electric power trade, its characterized in that: the intelligent data acquisition system comprises an intelligent data acquisition module, a high-performance edge server and a central server, wherein the intelligent acquisition module is used for acquiring audio data of an external service window, the intelligent acquisition module comprises a user information acquisition module and a speech prediction module, the high-performance edge server is realized based on an edge service technology, the high-performance edge server comprises an intelligent voice conversion module, an intelligent text retrieval module and a machine learning module, and the central server comprises a BI module and a sensitive word stock management module.
2. The AI public opinion monitoring system of claim 1 for an external service window of a power industry, wherein: the user information acquisition module is used for acquiring user basic information of communication and history communication records of the user according to an external service dialogue window, wherein the user basic information comprises gender, age, cultural level and history occupation.
3. The AI public opinion monitoring system of claim 1 for an external service window of a power industry, wherein: the conversation prediction module comprises a conversation analysis module and a solution proposal module, wherein the conversation analysis module is used for analyzing voice data of a user communication calendar and carrying out conversation prediction, and the solution proposal module is used for automatically prompting dialogue reply proposal according to the communication records of the same type.
4. The AI public opinion monitoring system of claim 1 for an external service window of a power industry, wherein: the intelligent voice conversion module is used for converting collected voice data into characters, the intelligent character retrieval module is used for identifying sensitive words in the converted characters based on an artificial intelligent natural language processing technology, and the machine learning module is used for automatically expanding a sensitive word stock in the process of identifying the sensitive words based on an artificial intelligent machine self-learning model.
5. The AI public opinion monitoring system of claim 1 for an external service window of a power industry, wherein: the BI module is used for comprehensively analyzing the identified sensitive words and results, and the sensitive word bank management module is used for maintaining a business handling process and providing sensitive words required by public opinion monitoring.
6. The AI public opinion monitoring system of claim 1 for an external service window of a power industry, wherein: the sensitive word stock management module further comprises a sensitive word stock maintenance module, an intelligent recognition adding module and a sensitive word classification module, wherein the sensitive word stock maintenance module is used for increasing or decreasing sensitive words and adjusting the sensitive words, the intelligent recognition adding module is used for automatically checking the change of the sensitive word stock and dynamically adding the recognized sensitive words into a recognition algorithm, and the sensitive word classification module is used for classifying the sensitive words in multiple fields.
7. The AI public opinion monitoring system of claim 1 for an external service window of a power industry, wherein: the BI module further comprises an intelligent prediction module and a data conversion display module, wherein the intelligent prediction module is used for intelligently analyzing and predicting public opinion trends based on the occurrence frequency of public opinion sensitive words, and the data conversion display module is used for converting predicted and analyzed public opinion trend results into a concise and visual chart form for display.
8. The AI public opinion monitoring method for the external service window in the power industry is characterized by comprising the following steps:
step one, data acquisition, an intelligent data acquisition module acquires audio data of an external service site in real time and transmits the audio data to a high-performance edge server;
step two, conversion analysis, namely converting audio data into text data by an intelligent voice conversion module in the high-performance edge server, performing word segmentation and monitoring by utilizing an intelligent text retrieval module, and simultaneously calling data of a sensitive word stock for recognition analysis;
thirdly, training is established by the model, a machine learning model is established, history data is used for pre-training, the machine learning model is put into use after pre-training, and monitoring data are used as training materials for self-training while monitoring, analysis and recognition are carried out;
step four, actively prompting, namely when the model of the high-performance edge server identifies the sensitive word, sending out a warning prompt at the first time, and processing by a manager according to the warning prompt;
and fifthly, public opinion analysis output, namely mining historical data through a BI module of the central server, analyzing the identified sensitive words, and converting an analysis result into a visual chart form to display through a data conversion display module.
9. The AI public opinion monitoring method for the external service window of the power industry of claim 8, wherein: and step two, the sensitive word stock is based on a sensitive word stock management module of the central server, and the sensitive word stock is updated and expanded at fixed time according to the development of the service.
10. The AI public opinion monitoring method for the external service window of the power industry of claim 8, wherein: in the fourth step, corresponding processing scheme suggestions are given through the speaking prediction module when the sensitive word is detected, and reasonable language reply suggestions are given after speaking audios are recognized through the speaking prediction module when the sensitive word is not detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310019094.6A CN116128546A (en) | 2023-01-06 | 2023-01-06 | AI public opinion monitoring system and method for external service window in power industry |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310019094.6A CN116128546A (en) | 2023-01-06 | 2023-01-06 | AI public opinion monitoring system and method for external service window in power industry |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116128546A true CN116128546A (en) | 2023-05-16 |
Family
ID=86296833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310019094.6A Pending CN116128546A (en) | 2023-01-06 | 2023-01-06 | AI public opinion monitoring system and method for external service window in power industry |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116128546A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117972621A (en) * | 2024-04-02 | 2024-05-03 | 河北科迪新能源科技有限公司 | Artificial intelligence-based anti-electricity-stealing intelligent early warning method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103268350A (en) * | 2013-05-29 | 2013-08-28 | 安徽雷越网络科技有限公司 | Internet public opinion information monitoring system and monitoring method |
| CN104933093A (en) * | 2015-05-19 | 2015-09-23 | 武汉泰迪智慧科技有限公司 | Regional public opinion monitoring and decision-making auxiliary system and method based on big data |
| CN107644269A (en) * | 2017-09-11 | 2018-01-30 | 国网江西省电力公司南昌供电分公司 | A kind of electric power public opinion prediction method and device for supporting risk assessment |
| CN108319666A (en) * | 2018-01-19 | 2018-07-24 | 国网浙江省电力有限公司电力科学研究院 | A kind of electric service appraisal procedure based on multi-modal the analysis of public opinion |
| US20220270594A1 (en) * | 2021-02-24 | 2022-08-25 | Conversenowai | Adaptively Modifying Dialog Output by an Artificial Intelligence Engine During a Conversation with a Customer |
-
2023
- 2023-01-06 CN CN202310019094.6A patent/CN116128546A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103268350A (en) * | 2013-05-29 | 2013-08-28 | 安徽雷越网络科技有限公司 | Internet public opinion information monitoring system and monitoring method |
| CN104933093A (en) * | 2015-05-19 | 2015-09-23 | 武汉泰迪智慧科技有限公司 | Regional public opinion monitoring and decision-making auxiliary system and method based on big data |
| CN107644269A (en) * | 2017-09-11 | 2018-01-30 | 国网江西省电力公司南昌供电分公司 | A kind of electric power public opinion prediction method and device for supporting risk assessment |
| CN108319666A (en) * | 2018-01-19 | 2018-07-24 | 国网浙江省电力有限公司电力科学研究院 | A kind of electric service appraisal procedure based on multi-modal the analysis of public opinion |
| US20220270594A1 (en) * | 2021-02-24 | 2022-08-25 | Conversenowai | Adaptively Modifying Dialog Output by an Artificial Intelligence Engine During a Conversation with a Customer |
Non-Patent Citations (2)
| Title |
|---|
| 王亨桂: "电力行业舆情监测***的设计与实现", 信息技术, pages 1 - 5 * |
| 王鹏举 等: "基于云服务的政府舆情监测平台架构的设计与实现", 电子设计工程, vol. 23, no. 6, pages 1 - 4 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117972621A (en) * | 2024-04-02 | 2024-05-03 | 河北科迪新能源科技有限公司 | Artificial intelligence-based anti-electricity-stealing intelligent early warning method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7912714B2 (en) | Method for segmenting communication transcripts using unsupervised and semi-supervised techniques | |
| US11950020B2 (en) | Methods and apparatus for displaying, compressing and/or indexing information relating to a meeting | |
| KR20160060243A (en) | Apparatus and method for customer interaction service | |
| KR102353545B1 (en) | Method and Apparatus for Recommending Disaster Response | |
| CN111294468A (en) | Tone quality detection and analysis system for customer service center calling | |
| CN112579789A (en) | Equipment fault diagnosis method and device and equipment | |
| US20220301547A1 (en) | Method for processing audio signal, method for training model, device and medium | |
| CN116128546A (en) | AI public opinion monitoring system and method for external service window in power industry | |
| CN116663664A (en) | Customer marketing scene data analysis system and method based on NLP algorithm | |
| CN112685215A (en) | Cloud platform abnormal log analysis method | |
| CN114491034B (en) | Text classification method and intelligent device | |
| CN112860868A (en) | Customer service telephone analysis method, system, equipment and storage medium | |
| US20220318681A1 (en) | System and method for scalable, interactive, collaborative topic identification and tracking | |
| CN116756688A (en) | Public opinion risk discovery method based on multi-mode fusion algorithm | |
| CN111968628B (en) | Signal accuracy adjusting system and method for voice instruction capture | |
| CN115019411A (en) | Routing inspection system and method based on voice interaction | |
| CN114974229A (en) | Method and system for extracting abnormal behaviors based on audio data of power field operation | |
| CN114186041A (en) | Answer output method | |
| CN116828109A (en) | Intelligent evaluation method and system for telephone customer service quality | |
| CN116542800A (en) | Intelligent financial statement analysis system based on cloud AI technology | |
| CN111427996A (en) | Method and device for extracting date and time from human-computer interaction text | |
| CN116804691A (en) | Fault monitoring method for dispatching automation equipment of power system | |
| CN111352820A (en) | Method, equipment and device for predicting and monitoring running state of high-performance application | |
| Sharma et al. | Speech Emotion Recognition System using SVD algorithm with HMM Model | |
| Zhang et al. | A multiple instance learning algorithm using graph convolutional network for speech content classification |
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 |