CN112785347A - Intelligent customer service question and answer recommendation method and system based on knowledge graph - Google Patents
Intelligent customer service question and answer recommendation method and system based on knowledge graph Download PDFInfo
- Publication number
- CN112785347A CN112785347A CN202110170658.7A CN202110170658A CN112785347A CN 112785347 A CN112785347 A CN 112785347A CN 202110170658 A CN202110170658 A CN 202110170658A CN 112785347 A CN112785347 A CN 112785347A
- Authority
- CN
- China
- Prior art keywords
- entity
- words
- question
- list
- relation
- 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
- 238000010276 construction Methods 0.000 claims description 8
- 238000001228 spectrum Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 description 9
- 230000004044 response Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000012545 processing Methods 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
- G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
-
- 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
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- 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
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- 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
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
- G06Q30/016—After-sales
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses an intelligent customer service question and answer recommendation method and system based on a knowledge graph, relates to the technical field of artificial intelligence, and can be used for pushing out the most relevant questions or answers based on the knowledge graph when an intelligent customer service system cannot identify the intention of a user to ask questions. The method comprises the following steps: extracting common entity words, relation words, question entity sentences and answer entity sentences from sample data, and constructing a common entity list, a relation list, a common entity-question entity relation list and a common entity-answer entity relation list; when a question of a user is identified as missing by the intelligent customer service system, extracting common entity words and relation words in the question according to the common entity list and the relation list; traversing the common entity list based on the extracted common entity words, and searching corresponding question entity sentences through the common entity and question entity relation list and feeding back the corresponding question entity sentences to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding back the corresponding answer entity sentences to the user when the common entity list is traversed to the corresponding entities.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to an intelligent customer service question and answer recommendation method and system based on a knowledge graph.
Background
The human customer service department in the online platform, as a labor-intensive post, faces a number of problems. On one hand, the operation and maintenance costs such as the training cost, the communication cost, the collective office field cost and the like required by customer service greatly increase the enterprise operation burden; on the other hand, due to the difference of personnel levels and subjective individuality of manual operation, the customer service level is unreliable and customers feel unstable; in addition, with the increasing number of clients of modern enterprises, the workload of manual customer service is increasing due to the huge business consultation requirements. There is an increasingly prominent conflict between the limited efficiency of manual processing and the persistence of business growth. The intelligent customer service system is just produced in the contradiction between supply and demand.
The knowledge map technology is applied to an intelligent customer service system, an automatic question-answering system assisting artificial intelligence can be provided for the customer service industry, the working pressure of artificial customer service is relieved, the labor cost of enterprises is reduced, meanwhile, the user experience is improved, and the accuracy, the normalization and the stability of the enterprise service are improved.
Disclosure of Invention
The invention aims to provide an intelligent customer service question and answer recommendation method and system based on a knowledge graph, which can be used for deducing the most relevant questions or answers based on the knowledge graph when an intelligent customer service system cannot identify the intention of a user to ask questions.
In order to achieve the above object, a first aspect of the present invention provides an intelligent customer service question-answer recommendation method based on a knowledge graph, including:
extracting common entity words, relation words, question entity sentences and answer entity sentences from sample data, constructing a common entity list, a relation list, a common entity-question entity relation list and a common entity-answer entity relation list, and storing the common entity-question entity relation list and the answer entity relation list in a graph database;
when a question of a user is identified as missing by an intelligent customer service system, extracting common entity words and relation words in the question according to the common entity list and the relation list;
traversing the common entity list based on the extracted common entity words, and searching corresponding question entity sentences through the common entity and question entity relation list and feeding back the corresponding question entity sentences to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding back the corresponding answer entity sentences to the user when the common entity list is traversed to the corresponding entities.
Preferably, the method for extracting the general entity words, the relation words, the question entity sentences and the answer entity sentences from the sample data comprises the following steps:
after segmenting words according to each sample data, identifying and extracting common entity words and relation words in the sample data through parts of speech, or identifying and extracting common entity words and relation words in the sample data through a preset regular expression;
and identifying the sample data as question entity sentences or answer entity sentences.
Preferably, the method for constructing a common entity list, a relationship list between common entities and question entities, and a relationship list between common entities and answer entities stored in a graph database further comprises:
acquiring a plurality of common entity words of the sample data to construct a common entity list, acquiring a plurality of relation words of the sample data to construct a relation list, acquiring a plurality of common entity words, relation words and corresponding question entity sentences of the sample data to construct a common entity and question entity relation list, and acquiring a plurality of common entity words, relation words and corresponding answer entity sentences of the sample data to construct a common entity and answer entity relation list.
Preferably, the method for identifying the common entity words and the relation words in the words by parts of speech comprises the following steps:
and identifying the nouns in the sample data as common entity words, and identifying prepositions and/or verbs behind the nouns as relation words.
Preferably, the relation words in the sample data are expanded by near-synonym to enrich the relation words in the relation list, the relation list of the common entity and the question entity and the relation list of the common entity and the answer entity.
Preferably, the method for identifying the user's question and question intent according to the common entity list and the relationship list includes:
and identifying the intention of the user to ask a question by using a textcnn algorithm model, automatically replying the user to ask a question by the intelligent customer service system if the identification is successful, extracting common entity words and relation words in the question according to the common entity list and the relation list if the identification is failed, and replying the user to ask a question after inquiring the knowledge graph database.
Compared with the prior art, the intelligent customer service question and answer recommendation method based on the knowledge graph has the following beneficial effects:
the invention provides an intelligent customer service question and answer recommendation method based on a knowledge graph, which comprises the steps of firstly extracting common entity words, relationship words, question entity sentences and answer entity sentences from a plurality of sample data, constructing a common entity list, a relationship list of the common entities and the question entities and a relationship list of the common entities and the answer entities, correspondingly storing the common entity lists, the relationship list of the common entities and the question entities and the relationship list of the common entities and the answer entities in a graph database, judging whether the users intend to lose the intentions when the users send question sentences in an intelligent customer service system, namely whether the intentions of the user question sentences can be identified by the intelligent customer service system, automatically replying the user questions by the intelligent customer service system if the identifications succeed, extracting the common entity words and the relationship words in the question sentences according to the common entity list and the relationship list if the identifications fail, inquiring the user question by the graph database through the knowledge graph, particularly extracting the common entity words and the relationship words in the question sentences, and traversing the common entity list based on the extracted common entity words, and searching corresponding question entity sentences through the common entity and question entity relation list and feeding back the corresponding question entity sentences to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding back the corresponding answer entity sentences to the user when the common entity list is traversed to the corresponding entities.
Therefore, when the intelligent customer service system identifies that the question is absent, the most relevant questions or answers can be pushed out through the assistance of the knowledge graph, and the customer service question-answering experience of the user is improved.
A second aspect of the present invention provides an intelligent customer service question and answer recommendation system based on a knowledge graph, which is applied to the intelligent customer service question and answer recommendation method based on a knowledge graph in the above technical solution, and the system includes:
the system comprises a construction unit, a graph spectrum database and a database, wherein the construction unit is used for extracting common entity words, relation words, question entity sentences and answer entity sentences from sample data, constructing a common entity list, a relation list of common entities and question entities and a relation list of common entities and answer entities and storing the common entity list, the relation list and the answer entity list in the graph spectrum database;
the extraction unit is used for extracting common entity words and relation words in the question sentences according to the common entity list and the relation list when the question sentences of the users are identified as missing intentions by the intelligent customer service system;
and the searching unit is used for traversing the common entity list based on the extracted common entity words, searching corresponding question entity sentences through the common entity and question entity relation list and feeding the question entity sentences back to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding the answer entity sentences back to the user when the common entity list is traversed to the corresponding entities.
Preferably, the construction unit comprises:
the word segmentation module is used for identifying and extracting common entity words and relation words in the sample data by parts of speech after segmenting words of the sample data, or identifying and extracting the common entity words and the relation words in the sample data by adopting a preset regular expression;
and the identification module is used for identifying and extracting common entity words and relation words in the sample data by parts of speech after segmenting the sample data into words, or identifying and extracting the common entity words and the relation words in the sample data by adopting a preset regular expression.
Preferably, the method for identifying the common entity words and the relation words in the words by parts of speech comprises the following steps:
and identifying the nouns in the sample data as common entity words, and identifying prepositions and/or verbs behind the nouns as relation words.
Compared with the prior art, the beneficial effects of the intelligent customer service question and answer recommendation system based on the knowledge graph provided by the invention are the same as the beneficial effects of the intelligent customer service question and answer recommendation method based on the knowledge graph provided by the technical scheme, and the detailed description is omitted here.
A third aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, executes the steps of the above-mentioned intelligent customer service question-and-answer recommendation method based on a knowledge graph.
Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the invention are the same as the beneficial effects of the intelligent customer service question and answer recommendation method based on the knowledge graph provided by the technical scheme, and the detailed description is omitted here.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flow chart of an intelligent customer service question and answer recommendation method based on a knowledge graph in the embodiment of the invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
Referring to fig. 1, the present embodiment provides an intelligent customer service question and answer recommendation method based on a knowledge graph, including:
extracting common entity words, relation words, question entity sentences and answer entity sentences from sample data, constructing a common entity list, a relation list, a common entity-question entity relation list and a common entity-answer entity relation list, and storing the common entity-question entity relation list and the answer entity relation list in a graph database; when a question of a user is identified as missing by the intelligent customer service system, extracting common entity words and relation words in the question according to the common entity list and the relation list; traversing the common entity list based on the extracted common entity words, and searching corresponding question entity sentences through the common entity and question entity relation list and feeding back the corresponding question entity sentences to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding back the corresponding answer entity sentences to the user when the common entity list is traversed to the corresponding entities.
In the method for recommending intelligent customer service question-answering based on the knowledge graph provided by this embodiment, first, common entity words, relationship words, question entity sentences and answer entity sentences are extracted from a plurality of sample data, a common entity list, a relationship list of the common entities and the question entities, and a relationship list of the common entities and the answer entities are constructed and stored in a graph database, when a user sends a question sentence in an intelligent customer service system, the intelligent customer service system first determines whether the user intends to be absent, that is, whether the intention of the user to ask the question sentence can be identified, if the identification is successful, the intelligent customer service system automatically replies the user question, if the identification is failed, the common entity words and the relationship words in the question sentence are extracted according to the common entity list and the relationship list, the graph database is called to reply the user question after being queried through the knowledge graph, specifically, the common entity words and the relationship words in the question sentence are extracted, and traversing the common entity list based on the extracted common entity words, and searching corresponding question entity sentences through the common entity and question entity relation list and feeding back the corresponding question entity sentences to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding back the corresponding answer entity sentences to the user when the common entity list is traversed to the corresponding entities.
Therefore, according to the embodiment, when the intelligent customer service system identifies that the question is absent, the most relevant questions or answers can be pushed out in an auxiliary mode through the knowledge graph, and the customer service question and answer experience of the user is improved.
In the above embodiment, the method for extracting the general entity words, the relation words, the question entity sentences and the answer entity sentences from the sample data includes:
after segmenting words for each sample data, identifying and extracting common entity words and relation words in the sample data through parts of speech, or identifying and extracting common entity words and relation words in the sample data through a preset regular expression; and identifying the sample data as question entity sentences or answer entity sentences.
In specific implementation, a plurality of sample data needs to be collected in advance, each sample data is segmented by adopting a jieba, and each segmented word is labeled in part of speech, a jieba part of speech labeling function based on hmm is used here, the part of speech after the sample data segmentation generally comprises a noun, a preposition and a verb, the noun is used as a common entity word of the sample data, the preposition and the verb are used as related words, for example, the sample data is "pay how to pay for easy payment", wherein "pay as a noun corresponds to the common entity word," pay "as a preposition, and" pay "as a verb, that is," pay "corresponds to the related word. In addition, the regular expression can automatically identify the sentence structure of the sample data and extract common entity words and relation words. The recognition of the question entity sentence or the answer entity sentence can be recognized by whether the question entity sentence or the answer entity sentence includes the query word or not, and can also be recognized based on model training, which is not limited in this embodiment.
In the above embodiment, the method for constructing the general entity list, the relationship list between the general entity and the question entity, and the relationship list between the general entity and the answer entity, and storing them in the graph database further includes:
the method comprises the steps of collecting common entity words of a plurality of sample data to construct a common entity list, collecting relation words of the plurality of sample data to construct a relation list, collecting common entity words, relation words and corresponding question entity sentences of the plurality of sample data to construct a common entity and question entity relation list, and collecting common entity words, relation words and corresponding answer entity sentences of the plurality of sample data to construct a common entity and answer entity relation list.
In specific implementation, common entities in sample data of unstructured data are extracted, after the extraction, common entity words can be obtained through manual examination, the common entity list at least comprises three dimensional information which are respectively a common entity word ID, the common entity words and at least one similar meaning word corresponding to the common entity words, and each common entity word ID and each common entity word in the common entity list are unique. The common entity and problem entity relation list at least comprises three dimensional information which are common entity words, problem IDs and problem entity sentences corresponding to the common entity words and the relation words respectively; the common entity and answer entity relationship list at least comprises three dimensional information which are respectively a common entity word, an answer ID and an answer entity sentence corresponding to the common entity word and the relationship word; the list files are all saved in the map database through the CSV format.
In the above embodiment, the method for identifying the common entity words and the relation words by parts of speech includes:
a noun in the sample data is recognized as a normal entity word, and a preposition word and/or a verb behind the noun is recognized as a relation word. For another example, for the sample data of "how to log on to the easy pay bank account", it is obvious that "the easy pay bank account" is a noun, "how" is a preposition, "log on" is a verb, and based on the above extraction rule, the "easy pay bank account" can be extracted as a common entity word, and "how to log on" is an associated word.
In the above embodiment, the relation words in the sample data are expanded by the similar meaning words to enrich the relation words in the relation list, the relation series list of the common entity and the question entity, and the relation list of the common entity and the answer entity. For example, the synonym of "how" can be expanded to "why", "how", etc.
In the above embodiment, the method for identifying a question and question intention of a user according to the common entity list and the relationship list includes: and identifying the intention of the user to ask a question by using a textcnn algorithm model, automatically replying the user to ask a question by the intelligent customer service system if the identification is successful, extracting common entity words and relation words in the question according to a common entity list and a relation series list if the identification is failed, and replying the user to ask a question after inquiring the knowledge map by using a calling map database.
In specific implementation, the textcnn algorithm is adopted to identify the intention probability of a user question, when the intention probability is smaller than a threshold value, the intention identification is successful, when the intention probability is larger than the threshold value, the intention identification is failed, illustratively, the probabilities of the user question in 17 categories are identified through the textcnn algorithm model, then the variance of the probabilities of the 17 categories is calculated, and if the variance is larger than 0.2, the intention identification is failed. And when the intention identification fails, calling the spectrum database to inquire through the knowledge graph and then replying a user question.
In the above embodiment, the common entity list is traversed based on the extracted common entity words, and when the common entity list is traversed to the corresponding entity, the corresponding question entity statement is searched for through the common entity and question entity relationship list and fed back to the user, or the corresponding answer entity statement is searched for through the common entity and answer entity relationship list and fed back to the user. When the method is specifically implemented, the common entity words extracted from the question sentences are traversed in the common entity list, if the common entity list can traverse the common entity words, the question questions can be searched for relevant questions or answers through the knowledge graph, then the question entity sentences consistent with the question questions are searched for in a traversing mode through the common entity and question entity relation list and fed back to the user, or the corresponding answer entity sentences are searched for in a traversing mode through the common entity and answer entity relation list and fed back to the user.
In order to make the solution of this embodiment simultaneously applicable to the customer service systems with three terminals of APP, WAP and PC, common entity words, relation words, question entity statements and answer entity statements corresponding to the three terminals are deployed in the common entity list, the common entity-to-question entity relation list and the common entity-to-answer entity relation list, respectively.
In summary, the embodiment provides a set of solutions for assisting the intelligent customer service system in answering the user questions through the knowledge map, and when the intelligent customer service system cannot identify the intentions of the user questions asked, the relevant questions and/or answers can be recommended for the user to select and view, so that the accuracy and efficiency of the customer service questions and answers are improved. The innovation points of the embodiment are as follows:
1. based on the application of the scheme of the embodiment, the system can assist the manual customer service to undertake a part of customer consultation work, particularly high-efficiency response under a high-concurrency scene, relieves the working pressure of the customer service, and reduces the operation and maintenance cost of the customer service;
2. based on the application of the scheme of the embodiment, because the standardized business knowledge base is used, the response of the customer problem is given according to the established program and the mathematical model, the difference between the personnel service quality and the professional level of the manual customer service can be avoided, and the continuous, stable and standardized service output is ensured;
3. based on the application of the scheme of the embodiment, the clustering deployment can be realized, the fault tolerance of the response service of the intelligent customer service system is improved, and the 7 x 24 continuous and stable service output of the intelligent customer service system is ensured;
4. the network service framework adopted by the scheme ensures the independence of the intelligent customer service and the user front-end service (APP, WAP and PC), so that the deployment of the intelligent customer service does not depend on the original environment of an enterprise subscribing the service, and the intelligent customer service has high production compatibility.
Example two
The embodiment provides an intelligent customer service question-answer recommendation system based on a knowledge graph, which comprises:
the system comprises a construction unit, a graph spectrum database and a database, wherein the construction unit is used for extracting common entity words, relation words, question entity sentences and answer entity sentences from sample data, constructing a common entity list, a relation list of common entities and question entities and a relation list of common entities and answer entities and storing the common entity list, the relation list and the answer entity list in the graph spectrum database;
the system comprises an extraction unit, a question analysis unit and a question analysis unit, wherein the extraction unit is used for extracting common entity words and relation words in a question according to a common entity list and a relation series list when the question of a user is identified as missing intention by an intelligent customer service system;
and the searching unit is used for traversing the common entity list based on the extracted common entity words, searching corresponding question entity sentences through the common entity and question entity relation list and feeding the question entity sentences back to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding the answer entity sentences back to the user when the common entity list is traversed to the corresponding entities.
Preferably, the construction unit comprises:
the word segmentation module is used for identifying and extracting common entity words and relation words in the sample data by parts of speech after segmenting words of the sample data, or identifying and extracting the common entity words and the relation words in the sample data by adopting a preset regular expression;
and the identification module is used for identifying and extracting common entity words and relation words in the sample data by parts of speech after segmenting the sample data into words, or identifying and extracting the common entity words and the relation words in the sample data by adopting a preset regular expression.
Preferably, the method for identifying the common entity words and the relation words in the words by parts of speech comprises the following steps:
and identifying the nouns in the sample data as common entity words, and identifying prepositions and/or verbs behind the nouns as relation words.
Compared with the prior art, the beneficial effects of the intelligent customer service question and answer recommendation system based on the knowledge graph provided by the embodiment of the invention are the same as the beneficial effects of the intelligent customer service question and answer recommendation method based on the knowledge graph provided by the first embodiment, and the detailed description is omitted here.
EXAMPLE III
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the above-mentioned intelligent customer service question-and-answer recommendation method based on a knowledge graph.
Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the embodiment are the same as the beneficial effects of the intelligent customer service question and answer recommendation method based on the knowledge graph provided by the technical scheme, and are not repeated herein.
It will be understood by those skilled in the art that all or part of the steps in the method for implementing the invention may be implemented by hardware instructions related to a program, the program may be stored in a computer-readable storage medium, and when executed, the program includes the steps of the method of the embodiment, and the storage medium may be: ROM/RAM, magnetic disks, optical disks, memory cards, and the like.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (10)
1. An intelligent customer service question and answer recommendation method based on a knowledge graph is characterized by comprising the following steps:
extracting common entity words, relation words, question entity sentences and answer entity sentences from sample data, constructing a common entity list, a relation list, a common entity-question entity relation list and a common entity-answer entity relation list, and storing the common entity-question entity relation list and the answer entity relation list in a graph database;
when a question of a user is identified as missing by an intelligent customer service system, extracting common entity words and relation words in the question according to the common entity list and the relation list;
traversing the common entity list based on the extracted common entity words, and searching corresponding question entity sentences through the common entity and question entity relation list and feeding back the corresponding question entity sentences to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding back the corresponding answer entity sentences to the user when the common entity list is traversed to the corresponding entities.
2. The method of claim 1, wherein the method of extracting general entity words, relation words, question entity sentences and answer entity sentences from the sample data comprises:
after segmenting words according to each sample data, identifying and extracting common entity words and relation words in the sample data through parts of speech, or identifying and extracting common entity words and relation words in the sample data through a preset regular expression;
and identifying the sample data as question entity sentences or answer entity sentences.
3. The method of claim 1 or 2, wherein the method of constructing a list of common entities, a list of relationships between common entities and question entities, and a list of relationships between common entities and answer entities stored in a graph database further comprises:
acquiring a plurality of common entity words of the sample data to construct a common entity list, acquiring a plurality of relation words of the sample data to construct a relation list, acquiring a plurality of common entity words, relation words and corresponding question entity sentences of the sample data to construct a common entity and question entity relation list, and acquiring a plurality of common entity words, relation words and corresponding answer entity sentences of the sample data to construct a common entity and answer entity relation list.
4. The method of claim 2, wherein the method of identifying common entity words and related words by part of speech comprises:
and identifying the nouns in the sample data as common entity words, and identifying prepositions and/or verbs behind the nouns as relation words.
5. The method of claim 2, wherein near-synonym expansion is performed for the relation words in the sample data to enrich the relation words in the relation list, the common entity and question entity relation list, and the common entity and answer entity relation list.
6. The method of claim 1, wherein the method of identifying the user's question and question intent from the common entity list and the relationship list comprises:
and identifying the intention of the user to ask a question by using a textcnn algorithm model, automatically replying the user to ask a question by the intelligent customer service system if the identification is successful, extracting common entity words and relation words in the question according to the common entity list and the relation list if the identification is failed, and replying the user to ask a question after inquiring the knowledge graph database.
7. An intelligent customer service question-answer recommendation system based on a knowledge graph is characterized by comprising:
the system comprises a construction unit, a graph spectrum database and a database, wherein the construction unit is used for extracting common entity words, relation words, question entity sentences and answer entity sentences from sample data, constructing a common entity list, a relation list of common entities and question entities and a relation list of common entities and answer entities and storing the common entity list, the relation list and the answer entity list in the graph spectrum database;
the extraction unit is used for extracting common entity words and relation words in the question sentences according to the common entity list and the relation list when the question sentences of the users are identified as missing intentions by the intelligent customer service system;
and the searching unit is used for traversing the common entity list based on the extracted common entity words, searching corresponding question entity sentences through the common entity and question entity relation list and feeding the question entity sentences back to the user or searching corresponding answer entity sentences through the common entity and answer entity relation list and feeding the answer entity sentences back to the user when the common entity list is traversed to the corresponding entities.
8. The system of claim 7, wherein the building unit comprises:
the word segmentation module is used for identifying and extracting common entity words and relation words in the sample data by parts of speech after segmenting words of the sample data, or identifying and extracting the common entity words and the relation words in the sample data by adopting a preset regular expression;
and the identification module is used for identifying and extracting common entity words and relation words in the sample data by parts of speech after segmenting the sample data into words, or identifying and extracting the common entity words and the relation words in the sample data by adopting a preset regular expression.
9. The system of claim 8, wherein the method for identifying common entity words and related words by parts of speech comprises:
and identifying the nouns in the sample data as common entity words, and identifying prepositions and/or verbs behind the nouns as relation words.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of the claims 1 to 6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110170658.7A CN112785347A (en) | 2021-02-08 | 2021-02-08 | Intelligent customer service question and answer recommendation method and system based on knowledge graph |
| CA3148085A CA3148085A1 (en) | 2021-02-08 | 2022-02-08 | Knowledge-graph-based smart customer-service q&a recommendation method and system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110170658.7A CN112785347A (en) | 2021-02-08 | 2021-02-08 | Intelligent customer service question and answer recommendation method and system based on knowledge graph |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112785347A true CN112785347A (en) | 2021-05-11 |
Family
ID=75761230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110170658.7A Pending CN112785347A (en) | 2021-02-08 | 2021-02-08 | Intelligent customer service question and answer recommendation method and system based on knowledge graph |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112785347A (en) |
| CA (1) | CA3148085A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113392197A (en) * | 2021-06-15 | 2021-09-14 | 吉林大学 | Question-answer reasoning method and device, storage medium and electronic equipment |
| CN113792153A (en) * | 2021-08-25 | 2021-12-14 | 北京百度网讯科技有限公司 | Question and answer recommendation method and device |
| CN116955560A (en) * | 2023-07-21 | 2023-10-27 | 广州拓尔思大数据有限公司 | Data processing method and system based on thinking chain and knowledge graph |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109492077A (en) * | 2018-09-29 | 2019-03-19 | 北明智通（北京）科技有限公司 | The petrochemical field answering method and system of knowledge based map |
-
2021
- 2021-02-08 CN CN202110170658.7A patent/CN112785347A/en active Pending
-
2022
- 2022-02-08 CA CA3148085A patent/CA3148085A1/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109492077A (en) * | 2018-09-29 | 2019-03-19 | 北明智通（北京）科技有限公司 | The petrochemical field answering method and system of knowledge based map |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113392197A (en) * | 2021-06-15 | 2021-09-14 | 吉林大学 | Question-answer reasoning method and device, storage medium and electronic equipment |
| CN113392197B (en) * | 2021-06-15 | 2023-08-04 | 吉林大学 | Question-answering reasoning method and device, storage medium and electronic equipment |
| CN113792153A (en) * | 2021-08-25 | 2021-12-14 | 北京百度网讯科技有限公司 | Question and answer recommendation method and device |
| CN113792153B (en) * | 2021-08-25 | 2023-12-12 | 北京度商软件技术有限公司 | Question and answer recommendation method and device |
| CN116955560A (en) * | 2023-07-21 | 2023-10-27 | 广州拓尔思大数据有限公司 | Data processing method and system based on thinking chain and knowledge graph |
| CN116955560B (en) * | 2023-07-21 | 2024-01-05 | 广州拓尔思大数据有限公司 | Data processing method and system based on thinking chain and knowledge graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CA3148085A1 (en) | 2022-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109213910B (en) | A kind of government intelligence customer service method, system, electronic equipment and storage medium | |
| CN112785347A (en) | Intelligent customer service question and answer recommendation method and system based on knowledge graph | |
| CN108764480B (en) | Information processing system | |
| US20200257860A1 (en) | Semantic recognition method, electronic device, and computer-readable storage medium | |
| CN104679910A (en) | Intelligent answering method and system | |
| CN110175229B (en) | Method and system for on-line training based on natural language | |
| CN107807960B (en) | Intelligent customer service method, electronic device and computer readable storage medium | |
| CN110781280A (en) | Knowledge graph-based voice assisting method and device | |
| CN113297365B (en) | User intention judging method, device, equipment and storage medium | |
| CN113626573B (en) | Sales session objection and response extraction method and system | |
| CN112287082A (en) | Data processing method, device, equipment and storage medium combining RPA and AI | |
| CN109508367A (en) | Automatically extract the method, on-line intelligence customer service system and electronic equipment of question and answer corpus | |
| CN117493529B (en) | Anthropomorphic dialogue method and device based on natural language model and electronic equipment | |
| CN113672698A (en) | Intelligent interviewing method, system, equipment and storage medium based on expression analysis | |
| CN114969297A (en) | Method for analyzing power customer appeal relevancy | |
| US20210319481A1 (en) | System and method for summerization of customer interaction | |
| CN112988704A (en) | AI consultation database cluster building method and system | |
| CN117033560A (en) | Configuration item attribute alignment method based on large language model | |
| CN113111157B (en) | Question-answer processing method, device, computer equipment and storage medium | |
| US8051026B2 (en) | Rules collector system and method with user interaction | |
| CN113901839A (en) | User video information auditing method, device, equipment and storage medium | |
| CN110600016B (en) | File pushing method and device | |
| KR20220099690A (en) | Apparatus, method and computer program for summarizing document | |
| CN113239164A (en) | Multi-round conversation process construction method and device, computer equipment and storage medium | |
| CN112417113A (en) | Intelligent question-answering method and system based on voice recognition technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210511 |
|
| RJ01 | Rejection of invention patent application after publication |