CN110929007A - Electric power marketing knowledge system platform and application method - Google Patents
Electric power marketing knowledge system platform and application method Download PDFInfo
- Publication number
- CN110929007A CN110929007A CN201911085091.2A CN201911085091A CN110929007A CN 110929007 A CN110929007 A CN 110929007A CN 201911085091 A CN201911085091 A CN 201911085091A CN 110929007 A CN110929007 A CN 110929007A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- power
- library
- question
- electric power
- 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
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012545 processing Methods 0.000 claims abstract description 39
- 239000000463 material Substances 0.000 claims abstract description 32
- 238000003860 storage Methods 0.000 claims abstract description 13
- 238000005065 mining Methods 0.000 claims abstract description 7
- 238000012795 verification Methods 0.000 claims abstract description 7
- 238000004458 analytical method Methods 0.000 claims description 20
- 230000011218 segmentation Effects 0.000 claims description 18
- 230000000875 corresponding effect Effects 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000013515 script Methods 0.000 claims description 6
- 230000002596 correlated effect Effects 0.000 claims description 5
- 230000004048 modification Effects 0.000 claims description 5
- 238000012986 modification Methods 0.000 claims description 5
- 230000008439 repair process Effects 0.000 claims description 3
- 230000005612 types of electricity Effects 0.000 claims description 2
- 230000005611 electricity Effects 0.000 claims 2
- 238000004220 aggregation Methods 0.000 claims 1
- 230000002776 aggregation Effects 0.000 claims 1
- 238000012423 maintenance Methods 0.000 abstract description 2
- 238000010276 construction Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 238000012550 audit Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 230000001915 proofreading effect Effects 0.000 description 1
- 239000013049 sediment Substances 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
- G06F16/30—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
- G06F16/35—Clustering; Classification
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Public Health (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a power marketing knowledge system platform and an application method, wherein the platform comprises: the data layer mainly comprises power files, 95598 work order data and 95598 knowledge bases; the processing layer is used for extracting power knowledge from two angles of automatic machine mining and self-building by a user; the examination and verification layer is used for examining, publishing and warehousing aiming at the knowledge requirements of machine mining and user self-construction; the storage layer mainly comprises an electric document material library, an electric entry library and an electric question and answer experience library; and the consumption layer applies the stored knowledge to different consumption scenes. The knowledge system platform constructed by the invention effectively integrates multisource, heterogeneous and fragmented data into knowledge which can be called by the system, improves the data processing capability of the power marketing knowledge, optimizes the user experience and reduces the manual maintenance cost.
Description
Technical Field
The invention relates to the technical field of electric power marketing application, in particular to an electric power marketing knowledge system platform and an application method.
Background
Along with the rapid development of power marketing business, the importance of the accumulation, transmission, analysis and application of marketing knowledge is increasingly highlighted, and a unified knowledge management platform is preliminarily established by combining business requirements of a national network customer service center, a national network Jiangsu company, a national network Shandong company and the like, so that the knowledge service and the technical support are provided for power marketing personnel to process customer appeal. With the application of the internet, especially the fragmented application scenario of the mobile internet, the knowledge user puts higher and more updated requirements on knowledge management along with the development of the internet, and the following problems are mainly solved in urgent need: the method comprises the following steps of 'difficult use of a classification system', 'search results cannot be accurate to atomized knowledge', 'too coarse granularity of document splitting' and 'lack of a knowledge automatic updating mechanism'.
Disclosure of Invention
The invention aims to provide an electric power marketing knowledge system platform, which realizes electric power marketing information sharing of whole company, whole specialty, whole link and whole process.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the embodiment of the invention provides an electric power marketing knowledge system platform, which comprises:
the data layer is used for acquiring original power knowledge data;
the processing layer is used for processing the original power knowledge data of the data layer to generate knowledge points;
the examination and check layer is used for examining and checking the knowledge points to generate a knowledge subject and issuing the knowledge subject;
and a consumption layer for applying the generated knowledge topic.
Further, the data layer is specifically configured to obtain:
legal and legal policy documents refer to documents which are issued at the national level, the national network company level and the province company level and have guiding function on work;
95598 knowledge base, including power knowledge points of various service types of electricity charge and business service;
95598 work order data refers to 95598 work orders of business consultation and fault repair accepted by the national network customer service center through multiple channels;
an intelligent question and answer system (FAQ) library refers to a customer frequently-asked question and answer library in an intelligent customer service robot;
and the client label library is generated from the user dimension, the knowledge dimension and the answer dimension and comprises the service type, the time, the professional, the applicable unit and whether the label content is opened for the client.
Further, the process layer is particularly useful for,
adopting a predefined processing mode, and cleaning data by using different preprocessing scripts aiming at different texts;
performing semantic analysis on the cleaned data, including Chinese word segmentation, Chinese pinyin fault tolerance, Chinese part-of-speech tagging and Chinese syntax dependence analysis;
semantic elements are extracted from the content after semantic analysis, and the extracted semantic elements comprise names of people, names of places, names of organizations, time, prices and numerical words;
and automatically generating potential knowledge points according to the extracted semantic elements and a predefined template.
Further, the processing layer is further used for autonomously creating the knowledge points according to a knowledge point modification requirement or a newly added knowledge point collection requirement provided by a user.
Furthermore, the auditing layer is specifically used for,
selecting a matched knowledge template according to the knowledge content of the knowledge points, and checking and modifying the knowledge content according to the requirement of knowledge template specification; the knowledge template is a document template, a lexical template and a question and answer template;
adding knowledge labels and attributes to the checked knowledge points, filling in the association relation between the knowledge points and the professional words, and then storing the knowledge points in the form of the atomized knowledge;
filling corresponding attributes into the same classified atomized knowledge point content according to a preset entry knowledge format to generate a knowledge theme.
And further returning if the knowledge points submitted by the processing layer do not meet the specification requirements of the knowledge template.
Further, the storing of the knowledge points in the form of the atomized knowledge means that the knowledge contents of the knowledge points are divided according to the essential attributes thereof into the first-level classification and the second-level classification, the region of the knowledge points and the profession of the knowledge points.
The storage layer is specifically used for storing an electric document material library, an electric entry library and an electric question-answering experience library;
the power document material library is used for storing power document knowledge and related labels generated by the auditing layer according to the document template;
the electric power entry library is used for storing electric power entry knowledge and related labels generated by the examination layer according to the entry template examination;
the electric power question-answering experience base is used for storing electric power question-answering and related labels generated by the examination and verification layer according to the question-answering template examination.
Furthermore, the electric document material library, the electric entry library and the electric question-answering experience library have an association relationship:
the electric power entry library refers to files in the electric power document material library, hot experience of the electric power question and answer experience library is correlated, and meanwhile, manual correlation and intelligent mining of relations among entries are adopted inside the electric power entry library;
professional words appearing in the power document material library are linked to corresponding entries in the power entry library, hot experience of the power question-answering experience library is correlated, and meanwhile, manual correlation and intelligent mining of relations among documents are adopted inside the power document material library;
the electric power question-answering experience library refers to files in the electric power document material library, professional words appearing in the electric power question-answering experience library can be linked to corresponding entries in the electric power entry library, and meanwhile, similar question-answering experiences are manually related and intelligently mined in the electric power question-answering experience library.
Further, the consumer layer is particularly useful for,
providing power document retrieval and power document keyword query applications aiming at a power document material library;
providing electric entry retrieval and hot entry query application aiming at the electric entry library;
and providing power question-answering experience retrieval and hot question-answering experience query application aiming at the power question-answering experience library.
The embodiment of the invention also provides an application method of the power marketing knowledge, which comprises the following steps:
acquiring original power knowledge data;
processing the acquired original power knowledge data to generate knowledge points;
auditing the knowledge points to generate a knowledge subject and publishing the knowledge subject;
applying the generated knowledge topic.
Further, the processing the acquired raw power knowledge data to generate knowledge points includes:
adopting a predefined processing mode, and cleaning data by using different preprocessing scripts aiming at different texts;
performing semantic analysis on the cleaned data, including Chinese word segmentation, Chinese pinyin fault tolerance, Chinese part-of-speech tagging and Chinese syntax dependence analysis;
semantic elements are extracted from the content after semantic analysis, and the extracted semantic elements comprise names of people, names of places, names of organizations, time, prices and numerical words;
and automatically generating potential knowledge points according to the extracted semantic elements and a predefined template.
Further, auditing the knowledge points to generate a knowledge topic and publishing the knowledge topic, including:
selecting a matched knowledge template according to the knowledge content of the knowledge points, and checking and modifying the knowledge content according to the requirement of knowledge template specification; the knowledge template is a document template, a lexical template and a question and answer template;
adding knowledge labels and attributes to the checked knowledge points, filling in the association relation between the knowledge points and the professional words, and then storing the knowledge points in the form of the atomized knowledge;
filling corresponding attributes into the same classified atomized knowledge point content according to a preset entry knowledge format to generate a knowledge theme.
Further, the method also comprises the following steps:
storing the power document knowledge and the related labels generated according to the document template verification in a power document material library;
storing the electric power entry knowledge and the related labels generated according to the examination of the entry template in an electric power entry library;
and storing the electric power question and answer generated according to the question and answer template examination and the related tags in an electric power question and answer experience library.
Further, the generated knowledge topic is applied, and the method comprises the following steps:
carrying out power document retrieval and power document keyword query application on a power document material library;
carrying out electric entry retrieval and hot entry query application on the electric entry library;
and carrying out electric power question-answering experience retrieval and hot question-answering experience query application on the electric power question-answering experience base.
The invention has the beneficial effects that: the platform realizes the sharing of the electric power marketing information of the whole company, the whole specialty, the whole link and the whole process, further improves the professional business quality of marketing personnel, and meanwhile, actively provides value-added advanced services for clients by passively solving the client appeal, and continuously improves the whole service level of the clients in the whole province. The invention takes the customer service work upgrading and efficiency improvement as the overall target, refines an integrated, interactive and intelligent knowledge system building method covering the whole business field of the power company, supports the effective precipitation and sharing of the knowledge system, supports the continuous improvement of the quality of talent teams, supports the continuous improvement of the customer service level, simultaneously enables the knowledge to be spread more widely, and helps enterprises to realize the innovation of the knowledge.
Drawings
FIG. 1 is a block diagram of a power marketing knowledge platform of the present invention;
FIG. 2 is a block diagram of the correlation of the audit layer of knowledge points audit processing according to the present invention;
FIG. 3 is a flow chart illustrating a data processing process performed by the processing layer according to the present invention.
Detailed Description
The invention is further described below. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby. .
As shown in fig. 1, in view of the life cycle dimension of knowledge, the embodiment of the present invention provides an electric power marketing knowledge system platform, which mainly includes a data layer, a processing layer, an examination layer, a storage layer and a consumption layer from the viewpoint of the overall architecture. The knowledge storage dimension can be divided into a power document material library, a power entry library and a power question and answer experience library.
The power knowledge data sources at the data layer comprise:
(1) laws and regulations, policy documents: mainly refers to the files which are issued at the national level, the national network company level and the province company level and have guiding function to the work;
(2)95598 knowledge base: the electric power knowledge points comprise 11 service types such as electric charges, business services and the like, and powerful support is provided for customer service personnel to answer customer demands;
(3)95598 work order data: the customer service center of the national network accepts 95598 work orders of service types such as service consultation, fault repair and the like through multiple channels;
(4) an intelligent question-answering system FAQ library: the system refers to a client common question and answer library in an intelligent customer service robot, and comprises tens of thousands of standard question and answer libraries;
(5) a customer label library: the method is generated from a user dimension, a knowledge dimension and an answer dimension, and mainly comprises a service type, time, a professional, an applicable unit, whether to be opened to a client or not and the like.
Wherein, the 95598 knowledge base, the 95598 work order data and the FAQ base of the intelligent question answering system are known technical terms in the field of electric power.
The processing layer is used for processing the data of the data layer, and as shown in fig. 3, the processing layer mainly includes the following steps:
step 1: and (6) data cleaning. And performing data cleaning by using different preprocessing scripts aiming at different texts by using a predefined processing mode. The system can be used for cleaning the work order data, the dialogue data, the document data and the log data in a customized manner and collecting the standard data.
Step 2: and (6) semantic parsing. The system performs semantic analysis on the data cleaned in the step 1 by using a Chinese natural language processing technology. The method mainly comprises a Chinese word segmentation technology, a Chinese pinyin fault-tolerant technology, a Chinese part-of-speech tagging technology and a Chinese syntax dependence analysis technology, and belongs to the prior art means.
The Chinese word segmentation technology adopts an algorithm based on character labeling and adopts the idea of constructing words by characters. The method of forming words by characters is a method in the prior art, namely, Chinese word segmentation is converted into a sequence labeling problem by taking the characteristic that a plurality of characters in Chinese only appear at specific positions of words and phrases as a word segmentation basis and judging which part of the words and phrases a character belongs to. Mainly, models such as HMM (hidden markov model), CRF (conditional random field), and RNN, LSTM in deep learning that has emerged in recent years are used. The word segmentation algorithm can well process the problem of unknown words while the word segmentation is accurate, and the comprehensive effect is optimal. But word segmentation speed is also slower.
TABLE 1 example of word formation from words
Electric power | Can be used for | Watch (A) | Display device | Number of | A | Straight bar | In that | Increase | Adding |
B | M | E | B | E | B | E | S | B | E |
The expression of the word position in the method of forming words by characters is as follows: s-single word; b-start of multi-word; m-the middle of the multi-word; e-multiword endings.
The method can achieve the effect of recognizing words by enabling a computer to simulate the understanding of a sentence by a person, and various language knowledge is difficult to effectively help due to the complex Chinese semantics; on the other hand, Chinese word segmentation is the basis of understanding. The full understanding has limited improvement in word performance. Word segmentation as the first step in semantic understanding necessarily requires more accurate results. Therefore, the robot system adopts a word segmentation method based on statistics, and in order to take efficiency and accuracy into consideration, the method based on full segmentation and the method based on character marking are combined in practice, so that a good effect is achieved. In addition, a plurality of processing steps based on the word formation method are added into the word segmentation tool, and the overall effect is improved.
And step 3: and extracting semantic elements. Aiming at the contents processed by the Chinese natural language processing technology in the step 2, namely the data containing the results of word segmentation, pinyin, part of speech tagging and syntax dependence analysis, semantic elements such as knowledge conditions, states, entities and the like are extracted, and named entity identification and customized extraction contents based on a semantic template are added. And finishing the extraction work of the semantic elements in a manual proofreading mode. Semantic element extraction includes name of person, place name, organization name, time, price, number words, etc.
And 4, step 4: and generating potential knowledge points. And combining the generation results of the step 2 and the step 3, automatically generating potential knowledge points according to a template which is manually defined in advance according to the knowledge type, and waiting for the user to check.
In addition, knowledge can be created in a user self-building mode on a processing layer, the knowledge is mainly initiated from a superior level, a subordinate level and a unit, and business personnel can actively put forward a knowledge point modification requirement or a newly added knowledge point collection requirement when finding missing, inaccurate or incomplete knowledge points in a system in the process of serving customers or carrying out business work, and independently create knowledge content.
The examination and verification layer is used for examining and publishing knowledge requirements proposed by users and potential knowledge points generated and further examining and verifying publicly-created knowledge, and comprises a documentary template, a lexical template and a question-answering template, and a knowledge manager can modify and improve knowledge contents by referring to the template contents. The detailed processing procedure is shown in fig. 2, which specifically includes the following steps:
step 1: and the knowledge manager checks and processes the knowledge requirements proposed by the user and the generated potential knowledge points one by one. And when the knowledge administrator examines the knowledge, the knowledge type and the knowledge template which are preset manually are selected to be matched according to the knowledge content, the knowledge content is examined and modified according to the template specification requirement, edited and sorted and converted into the knowledge required by the client or the internal staff, and if the knowledge points submitted by the processing layer do not meet the specification requirement, the knowledge is returned to be collected again.
Step 2: adding knowledge labels and attributes to the knowledge points checked in the step 1, filling the association relation between the knowledge points and the professional words, and then storing the knowledge points in the form of the atomized knowledge. Namely, the knowledge content is divided according to the necessary attributes, such as the first class, the second class and the region, and the specific filling content is shown in table 2 below:
TABLE 2 atomized knowledge morphology
And step 3: and (3) aiming at the contents of the two or more classified atomized knowledge points in the step (2), filling corresponding attributes into the contents according to a manually preset format of the vocabulary entry knowledge to form a knowledge subject, and creating the vocabulary entry knowledge. The term knowledge mainly comprises basic concepts, basic contents, related backgrounds, specific contents, common client questions and answers and the like.
And the storage layer is used for storing an electric document material library, an electric entry library and an electric question-answering experience library. The storage layer stores the data corrected by revision of the audit layer and the data previously input by the technician.
Electric power document material library: the method is mainly used for storing relevant label and template information of power regulation, law and regulation and the like.
Electric power entry library: the method is mainly used for storing the electric power vocabulary entry type knowledge, the related labels of the vocabulary entries and the template information. Mainly comes from the vocabulary entry knowledge data created by the examination layer.
Electric power question-answering experience base: the method is mainly used for storing relevant label information of power question and answer and power experience type knowledge.
The method comprises the following steps that an incidence relation exists among a storage layer, an electric power document material library, an electric power entry library and an electric power question-answering experience library in a knowledge base system, and the incidence relation specifically comprises the following steps:
(1) the electric power entry library refers to related files of the electric power document material library, hot experience of the electric power question and answer experience library is correlated, and meanwhile, manual correlation and intelligent mining are adopted in the electric power entry library to mine relations among entries.
(2) The electric power document material library and the electric power question and answer experience library have an association relation, professional words appearing in the files can be linked to corresponding entries of the electric power entry library, and meanwhile, the relation between documents is mined through manual association and intelligence in the electric power document material library.
(3) The electric power question-answering experience library refers to files of an electric power document material library, professional words appearing in the electric power question-answering experience library can be linked to corresponding entries of the electric power entry library, and meanwhile, similar question-answering experiences are mined in the electric power question-answering experience library through manual association and intelligence.
The storage layer mainly adopts a database main-standby mode and comprises a basic database and a backup database. The scheme does not need related development and modification, various databases support the mode, the deployment and maintenance cost is low, and additional system complexity and bottleneck are not introduced. The scheme meets the operation requirement of the system at the present stage.
The consumption layer is mainly used for providing applications such as power document retrieval, power document keyword query and the like aiming at the power document material library according to the acquired related power knowledge; the method comprises the following steps of providing applications such as power entry retrieval and hot entry query for a power entry library; the method comprises the following steps of providing applications such as power question and answer experience retrieval and hot question and answer experience query aiming at a power question and answer experience library; and establishing relevant application scenes such as knowledge retrieval, knowledge recommendation, intelligent question answering, training examinations and the like. The consumer layer functions as follows:
(1) the power marketing knowledge base is used for scientifically classifying, storing and expressing power marketing knowledge and supporting a user to quickly search required knowledge through different dimensions;
(2) the electric power marketing file library is used for structuring electric power marketing files of all the years, so that retrieval and consumption are facilitated;
(3) the electric power marketing help-seeking question-answer library integrates the intelligent problem solving, experience sharing and topic discussion of people, and stimulates knowledge innovation enthusiasm and sediment knowledge;
(4) the electric marketing training examination module is used for realizing individual learning examination by innovating in artificial intelligence technology by using the past years of gamification, fragmentation and the like for reference;
(5) the electric power marketing expert question-answer library enables business experts to participate in the creation of knowledge;
(6) the power marketing operation guidance library is used for providing multimedia power marketing typical service scene operation guidance;
(7) the personal intelligent knowledge assistant can enable employees to quickly enter an assistant page through an APP, a common computer webpage and a WeChat, and provides multimedia interactive online intelligent knowledge assistance and data support services.
The electric power marketing knowledge system meets the capability requirements of a knowledge base such as ordered knowledge classification, atomized knowledge analysis and application, clear label management, autonomous knowledge acquisition and compilation and the like, optimizes the customer experience and reduces the cost of human resources. The embodiment of the invention also provides an application method of the power marketing knowledge, which comprises the following steps:
acquiring original power knowledge data;
processing the acquired original power knowledge data to generate knowledge points;
auditing the knowledge points to generate a knowledge subject and publishing the knowledge subject;
applying the generated knowledge topic.
Further, the processing the acquired raw power knowledge data to generate knowledge points includes:
adopting a predefined processing mode, and cleaning data by using different preprocessing scripts aiming at different texts;
performing semantic analysis on the cleaned data, including Chinese word segmentation, Chinese pinyin fault tolerance, Chinese part-of-speech tagging and Chinese syntax dependence analysis;
semantic elements are extracted from the content after semantic analysis, and the extracted semantic elements comprise names of people, names of places, names of organizations, time, prices and numerical words;
and automatically generating potential knowledge points according to the extracted semantic elements and a predefined template.
Further, auditing the knowledge points to generate a knowledge topic and publishing the knowledge topic, including:
selecting a matched knowledge template according to the knowledge content of the knowledge points, and checking and modifying the knowledge content according to the requirement of knowledge template specification; the knowledge template is a document template, a lexical template and a question and answer template;
adding knowledge labels and attributes to the checked knowledge points, filling in the association relation between the knowledge points and the professional words, and then storing the knowledge points in the form of the atomized knowledge;
filling corresponding attributes into the same classified atomized knowledge point content according to a preset entry knowledge format to generate a knowledge theme.
Further, the method also comprises the following steps:
storing the power document knowledge and the related labels generated according to the document template verification in a power document material library;
storing the electric power entry knowledge and the related labels generated according to the examination of the entry template in an electric power entry library;
and storing the electric power question and answer generated according to the question and answer template examination and the related tags in an electric power question and answer experience library.
Further, the generated knowledge topic is applied, and the method comprises the following steps:
carrying out power document retrieval and power document keyword query application on a power document material library;
carrying out electric entry retrieval and hot entry query application on the electric entry library;
and carrying out electric power question-answering experience retrieval and hot question-answering experience query application on the electric power question-answering experience base.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (15)
1. An electricity marketing knowledge system platform, comprising:
the data layer is used for acquiring original power knowledge data;
the processing layer is used for processing the original power knowledge data of the data layer to generate knowledge points;
the examination and check layer is used for examining and checking the knowledge points to generate a knowledge subject and issuing the knowledge subject;
and a consumption layer for applying the generated knowledge topic.
2. The power marketing knowledge system platform of claim 1, wherein the data layer is specifically configured to obtain:
legal and legal policy documents refer to documents which are issued at the national level, the national network company level and the province company level and have guiding function on work;
95598 knowledge base, including power knowledge points of various service types of electricity charge and business service;
95598 work order data refers to 95598 work orders of business consultation and fault repair accepted by the national network customer service center through multiple channels;
an intelligent question and answer system (FAQ) library refers to a customer frequently-asked question and answer library in an intelligent customer service robot;
and the client label library is generated from the user dimension, the knowledge dimension and the answer dimension and comprises the service type, the time, the professional, the applicable unit and whether the label content is opened for the client.
3. The electricity marketing knowledge system platform of claim 1, wherein the processing layer is specifically configured to,
adopting a predefined processing mode, and cleaning data by using different preprocessing scripts aiming at different texts;
performing semantic analysis on the cleaned data, including Chinese word segmentation, Chinese pinyin fault tolerance, Chinese part-of-speech tagging and Chinese syntax dependence analysis;
semantic elements are extracted from the content after semantic analysis, and the extracted semantic elements comprise names of people, names of places, names of organizations, time, prices and numerical words;
and automatically generating potential knowledge points according to the extracted semantic elements and a predefined template.
4. The power marketing knowledge system platform of claim 1, wherein the processing layer is further configured to autonomously create knowledge points according to a knowledge point modification requirement or a new knowledge point aggregation requirement provided by a user.
5. The power marketing knowledge system platform of claim 1, wherein the auditing layer is specifically configured to,
selecting a matched knowledge template according to the knowledge content of the knowledge points, and checking and modifying the knowledge content according to the requirement of knowledge template specification; the knowledge template is a document template, a lexical template and a question and answer template;
adding knowledge labels and attributes to the checked knowledge points, filling in the association relation between the knowledge points and the professional words, and then storing the knowledge points in the form of the atomized knowledge;
filling corresponding attributes into the same classified atomized knowledge point content according to a preset entry knowledge format to generate a knowledge theme.
6. The electric marketing knowledge system platform of claim 5, wherein the processing layer returns if the knowledge points submitted by the processing layer do not meet the specification requirements of the knowledge template.
7. The power marketing knowledge system platform of claim 5, wherein the storing of the knowledge points in the form of the atomized knowledge means that the knowledge contents of the knowledge points are divided according to their essential attributes into the first class and the second class, and the region and the profession to which the knowledge contents belong.
8. The power marketing knowledge system platform of claim 1, further comprising a storage layer, wherein the storage layer is specifically configured to store a power document material library, a power vocabulary entry library and a power question and answer experience library;
the power document material library is used for storing power document knowledge and related labels generated by the auditing layer according to the document template;
the electric power entry library is used for storing electric power entry knowledge and related labels generated by the examination layer according to the entry template examination;
the electric power question-answering experience base is used for storing electric power question-answering and related labels generated by the examination and verification layer according to the question-answering template examination.
9. The power marketing knowledge system platform of claim 7, wherein the power document material library, the power entry library and the power question-answering experience library have an association relationship of:
the electric power entry library refers to files in the electric power document material library, hot experience of the electric power question and answer experience library is correlated, and meanwhile, manual correlation and intelligent mining of relations among entries are adopted inside the electric power entry library;
professional words appearing in the power document material library are linked to corresponding entries in the power entry library, hot experience of the power question-answering experience library is correlated, and meanwhile, manual correlation and intelligent mining of relations among documents are adopted inside the power document material library;
the electric power question-answering experience library refers to files in the electric power document material library, professional words appearing in the electric power question-answering experience library can be linked to corresponding entries in the electric power entry library, and meanwhile, similar question-answering experiences are manually related and intelligently mined in the electric power question-answering experience library.
10. The power marketing knowledge system platform of claim 1, wherein the consumption layer is specifically configured to,
providing power document retrieval and power document keyword query applications aiming at a power document material library;
providing electric entry retrieval and hot entry query application aiming at the electric entry library;
and providing power question-answering experience retrieval and hot question-answering experience query application aiming at the power question-answering experience library.
11. An application method of power marketing knowledge, comprising:
acquiring original power knowledge data;
processing the acquired original power knowledge data to generate knowledge points;
auditing the knowledge points to generate a knowledge subject and publishing the knowledge subject;
applying the generated knowledge topic.
12. The method for applying electric power marketing knowledge according to claim 11, wherein the processing the acquired raw electric power knowledge data to generate knowledge points comprises:
adopting a predefined processing mode, and cleaning data by using different preprocessing scripts aiming at different texts;
performing semantic analysis on the cleaned data, including Chinese word segmentation, Chinese pinyin fault tolerance, Chinese part-of-speech tagging and Chinese syntax dependence analysis;
semantic elements are extracted from the content after semantic analysis, and the extracted semantic elements comprise names of people, names of places, names of organizations, time, prices and numerical words;
and automatically generating potential knowledge points according to the extracted semantic elements and a predefined template.
13. The method for applying electric power marketing knowledge according to claim 11, wherein auditing the knowledge points to generate a knowledge topic and publishing the knowledge topic comprises:
selecting a matched knowledge template according to the knowledge content of the knowledge points, and checking and modifying the knowledge content according to the requirement of knowledge template specification; the knowledge template is a document template, a lexical template and a question and answer template;
adding knowledge labels and attributes to the checked knowledge points, filling in the association relation between the knowledge points and the professional words, and then storing the knowledge points in the form of the atomized knowledge;
filling corresponding attributes into the same classified atomized knowledge point content according to a preset entry knowledge format to generate a knowledge theme.
14. The method for applying electric power marketing knowledge according to claim 13, further comprising:
storing the power document knowledge and the related labels generated according to the document template verification in a power document material library;
storing the electric power entry knowledge and the related labels generated according to the examination of the entry template in an electric power entry library;
and storing the electric power question and answer generated according to the question and answer template examination and the related tags in an electric power question and answer experience library.
15. The method for applying electric power marketing knowledge according to claim 14, wherein the applying the generated knowledge topic comprises:
carrying out power document retrieval and power document keyword query application on a power document material library;
carrying out electric entry retrieval and hot entry query application on the electric entry library;
and carrying out electric power question-answering experience retrieval and hot question-answering experience query application on the electric power question-answering experience base.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911085091.2A CN110929007A (en) | 2019-11-08 | 2019-11-08 | Electric power marketing knowledge system platform and application method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911085091.2A CN110929007A (en) | 2019-11-08 | 2019-11-08 | Electric power marketing knowledge system platform and application method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110929007A true CN110929007A (en) | 2020-03-27 |
Family
ID=69852481
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911085091.2A Pending CN110929007A (en) | 2019-11-08 | 2019-11-08 | Electric power marketing knowledge system platform and application method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110929007A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111737412A (en) * | 2020-05-21 | 2020-10-02 | 电子科技大学 | Citizen visiting guiding method based on natural language processing and knowledge graph |
CN111858965A (en) * | 2020-08-04 | 2020-10-30 | 中国银行股份有限公司 | Knowledge base generation method and device |
CN112465143A (en) * | 2020-12-07 | 2021-03-09 | 广东电网有限责任公司佛山供电局 | Knowledge storage system |
CN112612841A (en) * | 2020-12-29 | 2021-04-06 | 重庆农村商业银行股份有限公司 | Knowledge extraction construction method, device, equipment and storage medium |
CN113672742A (en) * | 2021-08-19 | 2021-11-19 | 鲨鱼快游网络技术(北京)有限公司 | Knowledge co-construction method, electronic device and medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109189752A (en) * | 2018-10-12 | 2019-01-11 | 国网山东省电力公司电力科学研究院 | Power marketing knowledge base system based on intelligent Search Technique |
CN109710772A (en) * | 2018-11-13 | 2019-05-03 | 国云科技股份有限公司 | A kind of question and answer library Knowledge Management System and its implementation based on deep learning |
-
2019
- 2019-11-08 CN CN201911085091.2A patent/CN110929007A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109189752A (en) * | 2018-10-12 | 2019-01-11 | 国网山东省电力公司电力科学研究院 | Power marketing knowledge base system based on intelligent Search Technique |
CN109710772A (en) * | 2018-11-13 | 2019-05-03 | 国云科技股份有限公司 | A kind of question and answer library Knowledge Management System and its implementation based on deep learning |
Non-Patent Citations (1)
Title |
---|
俞阳: "基于知识图谱的电力知识平台关键技术研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111737412A (en) * | 2020-05-21 | 2020-10-02 | 电子科技大学 | Citizen visiting guiding method based on natural language processing and knowledge graph |
CN111858965A (en) * | 2020-08-04 | 2020-10-30 | 中国银行股份有限公司 | Knowledge base generation method and device |
CN112465143A (en) * | 2020-12-07 | 2021-03-09 | 广东电网有限责任公司佛山供电局 | Knowledge storage system |
CN112612841A (en) * | 2020-12-29 | 2021-04-06 | 重庆农村商业银行股份有限公司 | Knowledge extraction construction method, device, equipment and storage medium |
CN113672742A (en) * | 2021-08-19 | 2021-11-19 | 鲨鱼快游网络技术(北京)有限公司 | Knowledge co-construction method, electronic device and medium |
CN113672742B (en) * | 2021-08-19 | 2024-04-26 | 鲨鱼快游网络技术(北京)有限公司 | Knowledge co-building method, electronic equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110825882B (en) | Knowledge graph-based information system management method | |
US10896212B2 (en) | System and methods for automating trademark and service mark searches | |
CN111177569B (en) | Recommendation processing method, device and equipment based on artificial intelligence | |
CN110929007A (en) | Electric power marketing knowledge system platform and application method | |
US9116985B2 (en) | Computer-implemented systems and methods for taxonomy development | |
CN110888990B (en) | Text recommendation method, device, equipment and medium | |
KR20190113965A (en) | Systems and methods for using machine learning and rule-based algorithms to create patent specifications based on human-provided patent claims such that patent specifications are created without human intervention | |
US10366117B2 (en) | Computer-implemented systems and methods for taxonomy development | |
CA2807494C (en) | Method and system for integrating web-based systems with local document processing applications | |
CN109960756A (en) | Media event information inductive method | |
CN110321549B (en) | New concept mining method based on sequential learning, relation mining and time sequence analysis | |
CN112445894A (en) | Business intelligent system based on artificial intelligence and analysis method thereof | |
CN113919336A (en) | Article generation method and device based on deep learning and related equipment | |
Patil et al. | Novel technique for script translation using NLP: performance evaluation | |
CN111736804B (en) | Method and device for identifying App key function based on user comment | |
US11361032B2 (en) | Computer driven question identification and understanding within a commercial tender document for automated bid processing for rapid bid submission and win rate enhancement | |
TWI734085B (en) | Dialogue system using intention detection ensemble learning and method thereof | |
CN115017271B (en) | Method and system for intelligently generating RPA flow component block | |
CN115269771A (en) | Big data analysis system based on semantics | |
CN111949781B (en) | Intelligent interaction method and device based on natural sentence syntactic analysis | |
CN112559753A (en) | Management framework of natural language text processing and analyzing task based on business process management technology | |
CN113761081A (en) | Method and system for carrying out multi-dimensional combined retrieval on enterprise information | |
CN112579666A (en) | Intelligent question-answering system and method and related equipment | |
Kasmuri et al. | Building a Malay-English code-switching subjectivity corpus for sentiment analysis | |
CN110930189A (en) | Personalized marketing method based on user behaviors |
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 |