CN114841347A - Self-service fault diagnosis method and device based on knowledge graph - Google Patents
Self-service fault diagnosis method and device based on knowledge graph Download PDFInfo
- Publication number
- CN114841347A CN114841347A CN202210342224.5A CN202210342224A CN114841347A CN 114841347 A CN114841347 A CN 114841347A CN 202210342224 A CN202210342224 A CN 202210342224A CN 114841347 A CN114841347 A CN 114841347A
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- description
- query statement
- picture
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a self-service fault diagnosis method and device based on a knowledge graph, and relates to the technical field of knowledge graphs, the technical field of image-text recognition and the financial field. The method comprises the following steps: acquiring fault description data input by a user, wherein the fault description data comprises at least one of text data, voice data and picture data; analyzing fault description data input by a user by using a preset description statement generation rule to obtain a fault query statement; and obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge map, wherein the fault diagnosis result comprises a fault reason or a fault solution. The method and the device can help the staff of the enterprise administration unit to quickly locate the fault reason and the fault solution, improve the fault repair speed, improve the technical operation and maintenance levels of different regions administered by the enterprise, and optimize the fault processing capacity of the enterprise.
Description
Technical Field
The application relates to the technical field of knowledge maps, the technical field of image-text recognition and the financial field, in particular to a self-service fault diagnosis method and device based on knowledge maps.
Background
With the vigorous development of economy, large, medium and small enterprises have more and more facilities, and technical operation and maintenance personnel need to master more and more abundant operation and maintenance knowledge. Various faults are often encountered during the operation of various devices, systems and software. Once a fault occurs, an enterprise administration unit usually determines the fault reason through the experience of technical operation and maintenance personnel and then solves the fault. If the fault is only known, and the fault reason and the solution are not clear, the department of jurisdiction will report the fault and give the fault to a special person in the superior department or department for processing. This process is referred to herein as a "manual fault reporting and handling procedure".
There are three ways of reporting fault by manual work, including making a call, sending a mail to a public mailbox, and reporting fault by chat group feedback. The defect of low processing efficiency exists in manual fault reporting, and the specific expression is as follows:
1) the response time is relatively long. Enterprises generally set special service desks to receive fault reporting information such as telephone calls, mails and the like, and transfer the task of solving the fault to corresponding technical operation and maintenance personnel. During the period, a transfer process is carried out, and the response speed is influenced by the regulation and control capability of the service desk. Once a plurality of jurisdictions simultaneously carry out a plurality of fault reporting processes, the processing response speed of the plurality of jurisdictions is tested.
2) It is difficult to reflect the specific situation. Some faults are difficult to describe by using oral languages, when fault reporting personnel call for fault reporting, technicians are difficult to restore fault scenes and make corresponding solution strategies, and meanwhile, solutions are difficult to describe in a telephone.
3) Messages are easily ignored. The failure reporting personnel of the jurisdictions report failures in the chat group, when the group personnel can not pay attention to the messages in time during working, the problems are more difficult to screen out from a plurality of messages, and the problem is considered in charge of the system. Moreover, different members of the group have knowledge and professional levels in different fields, and if a problem is simultaneously replied, the fault-reporting personnel are easily misled.
Disclosure of Invention
In order to improve the efficiency of fault diagnosis and solution, the application provides a self-service fault diagnosis method and device based on a knowledge graph, and relates to the technical field of knowledge graphs, the technical field of image-text recognition and the financial field.
In a first aspect, the present application provides a self-service fault diagnosis method based on a knowledge graph, the method comprising:
acquiring fault description data input by a user, wherein the fault description data comprises at least one of text data, voice data and picture data;
Analyzing fault description data input by a user by using a preset description statement generation rule to obtain a fault query statement;
and obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge map, wherein the fault diagnosis result comprises a fault reason or a fault solution.
In an embodiment, when the fault description data is text data, analyzing the fault description data input by the user by using a preset description statement generation rule to obtain a fault query statement, where the fault query statement includes:
performing semantic analysis on the text data to obtain fault description keywords;
determining a query statement structure corresponding to the fault description keyword according to the mapping relation between the fault description keyword and the query statement structure;
and obtaining a fault query statement according to the fault description keyword and the query statement structure.
In an embodiment, when the fault description data is voice data, analyzing the fault description data input by the user by using a preset description statement generation rule to obtain a fault query statement, including:
carrying out voice recognition on the voice data to obtain corresponding voice text data;
Performing semantic analysis on the voice text data to obtain fault description keywords;
matching a corresponding query statement structure according to the fault description keyword;
and obtaining a fault query statement according to the fault description keyword and the query statement structure.
In an embodiment, when the fault description data is picture data, analyzing the fault description data input by the user by using a preset description statement generation rule to obtain a fault query statement, where the fault query statement includes:
converting the picture data into picture text data by using a pre-established picture recognition text model;
semantic analysis is carried out on the picture text data to obtain fault description keywords;
matching a corresponding query statement structure according to the fault description keyword;
and obtaining a fault query statement according to the fault description keyword and the query statement structure.
In one embodiment, the step of creating the picture recognition text model comprises:
acquiring a plurality of fault equipment pictures and corresponding description sentences to obtain a training data set;
training a preset encoder-decoder model by using the training data set to obtain the picture recognition text model; the picture identification text model can generate corresponding picture text data according to the input fault equipment picture.
In one embodiment, the step of establishing the failure knowledge-graph comprises:
acquiring common fault types of equipment and fault reasons and fault solving methods corresponding to the common fault types; the fault type, the fault reason and the fault solving method are all data in text format;
mining the fault type, the fault reason, the fault solution and the corresponding relation thereof by a data mining method to obtain a plurality of entities and the relation among the entities;
carrying out entity disambiguation and parallel similarity calculation on the entities to obtain parallel similarity of the entities;
and establishing a fault knowledge graph according to the plurality of entities, the relationship among the entities and the similarity of the entities.
In an embodiment, obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge graph includes:
inputting the fault query statement into the fault knowledge graph, and determining a target fault entity corresponding to the fault query statement and a plurality of result fault entities corresponding to the target fault entity;
and outputting the plurality of result fault entities according to the sequence from high similarity to low similarity of the target fault entity and the result fault entity to obtain the fault diagnosis result.
In a second aspect, the present application provides a self-service fault diagnosis device based on a knowledge-graph, the device comprising:
the data acquisition module is used for acquiring fault description data input by a user, wherein the fault description data comprises at least one of text data, voice data and picture data;
the fault query statement generation module is used for analyzing fault description data input by a user by using a preset description statement generation rule to obtain a fault query statement;
and the fault diagnosis module is used for obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge map, wherein the fault diagnosis result comprises a fault reason or a fault solution method.
In one embodiment, the fault query statement generation module includes:
the semantic analysis unit is used for carrying out semantic analysis on the text data to obtain fault description keywords when the fault description data are the text data;
the query statement structure matching unit is used for determining a query statement structure corresponding to the fault description keyword according to the mapping relation between the fault description keyword and the query statement structure;
And the fault query statement generating unit is used for obtaining a fault query statement according to the fault description keyword and the query statement structure.
In an embodiment, the fault query statement generating module further includes a voice conversion unit, configured to perform voice recognition on the voice data to obtain corresponding voice text data when the fault description data is the voice data; and then the semantic analysis unit, the query sentence structure matching unit and the fault query sentence generating unit respectively process the voice text data to obtain a fault query sentence corresponding to the voice data.
In an embodiment, the fault query statement generating module further includes an image converting unit, configured to convert the picture data into picture text data by using a pre-created picture recognition text model when the fault description data is the picture data; and then the semantic analysis unit, the query sentence structure matching unit and the fault query sentence generating unit are used for respectively processing the picture text data to obtain a fault query sentence corresponding to the picture data.
In an embodiment, the self-service fault diagnosis apparatus based on a knowledge graph further includes a model building module for creating the image recognition text model, where the model building module is specifically configured to:
Acquiring a plurality of fault equipment pictures and corresponding description sentences to obtain a training data set;
training a preset encoder-decoder model by using the training data set to obtain the picture recognition text model; the picture identification text model can generate corresponding picture text data according to the input fault equipment picture.
In an embodiment, the self-service fault diagnosis apparatus based on a knowledge graph further includes a knowledge graph establishing module for creating the fault knowledge graph, and the knowledge graph establishing module is specifically configured to:
acquiring common fault types of equipment and fault reasons and fault solving methods corresponding to the common fault types; the fault type, the fault reason and the fault solving method are all data in text format;
mining the fault type, the fault reason, the fault solution and the corresponding relation thereof by a data mining method to obtain a plurality of entities and the relation among the entities;
carrying out entity disambiguation and parallel similarity calculation on the entities to obtain parallel similarity of the entities;
and establishing a fault knowledge graph according to the plurality of entities, the relationship among the entities and the similarity of the entities.
In an embodiment, the fault diagnosis module is specifically configured to:
inputting the fault query statement into the fault knowledge graph, and determining a target fault entity corresponding to the fault query statement and a plurality of result fault entities corresponding to the target fault entity;
and outputting the plurality of result fault entities according to the sequence from high similarity to low similarity of the target fault entity and the result fault entity to obtain the fault diagnosis result.
In a third aspect, the present application provides an electronic device, comprising:
the self-service fault diagnosis method based on the knowledge graph comprises a central processing unit, a storage and a communication module, wherein a computer program is stored in the storage, the central processing unit can call the computer program, and the central processing unit realizes any self-service fault diagnosis method based on the knowledge graph when executing the computer program.
In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program, which when executed by a processor implements any of the methods for knowledgegraph-based self-service fault diagnosis provided herein.
The self-service fault diagnosis method and device based on the knowledge map can help workers of the enterprise administration units to quickly locate fault reasons and a fault solution, and technical operation and maintenance personnel can take the initiative in processing faults according to recommended optimal solution steps. The method and the device can greatly improve the fault repairing speed, improve the technical operation and maintenance levels of different regions governed by the enterprise, and optimize the fault processing capacity of the enterprise.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic diagram of a self-service fault diagnosis method based on a knowledge graph provided in the present application.
Fig. 2 is another schematic diagram of the self-service fault diagnosis method based on the knowledge graph provided in the present application.
Fig. 3 is another schematic diagram of the self-service fault diagnosis method based on the knowledge graph provided in the present application.
Fig. 4 is another schematic diagram of the self-service fault diagnosis method based on the knowledge graph provided in the present application.
Fig. 5 is a schematic diagram of steps for creating a picture recognition text model according to the present application.
FIG. 6 is a schematic diagram of the steps provided herein for establishing a failure knowledge map.
Fig. 7 is another schematic diagram of the self-service fault diagnosis method based on the knowledge graph provided in the present application.
Fig. 8 is a schematic diagram of a self-service fault diagnosis device based on knowledge graph provided in the present application.
Fig. 9 is a schematic diagram of a self-service fault diagnosis device based on knowledge graph provided in the present application.
Fig. 10 is a schematic diagram of a self-service fault diagnosis device based on knowledge graph provided in the present application.
Fig. 11 is a schematic diagram of a self-service fault diagnosis device based on knowledge graph provided in the present application.
Fig. 12 is a schematic diagram of an electronic device provided in the present application.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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.
As shown in fig. 1, the present application provides a self-service fault diagnosis method based on a knowledge graph, which includes the following steps S101 to S103:
step S101, obtaining fault description data input by a user, wherein the fault description data comprises at least one of text data, voice data and picture data.
Specifically, the user can be understood as an error reporting person (i.e., a person who directly or indirectly reports a fault), fault description data is reported by the error reporting person, and the fault description data in three formats input by the error reporting person can be identified by the application, namely, the three formats are a text format, an audio format and a picture format. The data in the formats can be input by using the prior art, in practical application, the fault reporting method can be realized by depending on a computer system, and at the moment, fault description data in a text format is input by fault reporting personnel through input and output equipment such as a keyboard, a touch display screen and the like of the computer system, fault description data in a voice format is input through an audio acquisition module of the computer system, or fault transmission data in a picture format is uploaded through a camera or other data transmission interfaces. Since several data uploading methods in the examples have been developed, the present application does not make much description.
Step S102, analyzing fault description data input by a user by using a preset description statement generation rule to obtain a fault query statement.
Specifically, the description statement generation rule includes, but is not limited to, the steps of performing keyword parsing on the fault description data, performing query statement structure matching, and performing keyword filling. In fact, the step is to recombine the fault description sentences input by the fault reporting personnel and convert the fault description sentences into fault query sentences which have a specific sentence structure and can be identified by the fault knowledge map.
Step S103, obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge map, wherein the fault diagnosis result comprises a fault reason or a fault solving method.
Specifically, inputting a fault query statement into a pre-established knowledge graph to obtain a fault reason and a fault solution corresponding to a fault described in the fault query statement; the fault diagnosis result of the step not only outputs a group of fault reasons and a fault solution, but also outputs a plurality of fault reasons which may cause the fault described in the fault query statement in the order of high relevance to low relevance, and the fault solution is also output in the same way. The user can determine the most possible reason for the fault according to the correlation degree, and proceed to process the fault according to the optimal solution method of the fault, thereby improving the fault solution efficiency.
The method and the system can help fault reporting personnel to quickly locate the fault reason and obtain the fault solution, do not need to wait for the response and the troubleshooting of related technical personnel, greatly improve the fault repairing speed, improve the technical operation and maintenance levels of different regions governed by the enterprise, and optimize the fault processing capacity of the enterprise.
In an embodiment, as shown in fig. 2, when the fault description data is text data, step S102, analyzing the fault description data input by the user by using a preset description statement generation rule to obtain a fault query statement, where the method includes:
and step S1021, performing semantic analysis on the text data to obtain fault description keywords.
Specifically, the text data may be in various forms such as phrases, sentences, paragraphs, and the like, and the fault description keywords obtained by parsing the text data are words, or phrases representing core contents of the text data.
When semantic analysis is carried out, firstly, word segmentation operation is carried out on text data input by fault reporting personnel to obtain a plurality of single words, words or phrases; and then, carrying out stop words, and filtering stop words in the single words, the words or the phrases obtained by word segmentation, wherein for example, the words such as "in" and "in" are stop words, and the words such as "the" in "and" at "in" are stop words. And obtaining keywords after filtering, namely the fault description keywords corresponding to the text data.
For example, suppose that the text data entered by the error reporting person is "how does the apparatus 1 has the failure a? "first, the text data is cut into words to obtain" device 1 "," occurrence "," failure a "," how to do "; and then, removing the stop word from the result of the word segmentation, and removing the word to obtain the fault description keywords 'equipment 1', 'occurrence', 'fault A' and 'how to do' corresponding to the text data.
Step S1022, determining the query sentence structure corresponding to the fault description keyword according to the mapping relationship between the fault description keyword and the query sentence structure.
Specifically, the query statement structure may be understood as a query statement template, and the query statement structure is a statement structure recognizable by the failure knowledge graph. For example, the present application provides two examples of query statement structures herein: a cause of < failed entity > in < equipment entity >, a solution for < failed entity > in < equipment entity >. The mapping relationship between the fault description keyword and the query statement structure may be stored in the memory in the form of a mapping relationship table, which may be referred to as table 1:
table 1: mapping relation table of fault description key words and query statement structures
Here, the process of determining the structure of the query statement is illustrated by way of example, and is also illustrated in step S1021: suppose that the text data entered by the error reporting person is "how do the device 1 has a fault a? "and the fault description keywords" device 1 "," occurrence "," fault a ", and" how to "are obtained, and whether the above fault description keywords appear in the fault description keywords listed in table 1 is respectively queried; in this example, the fault description keyword "how to" is found in table 1, and the query sentence structure having the mapping relation with "how to" can be determined as "< solution of < fault entity > in device entity >. From this it is possible to determine the text data "how do the device 1 has a fault a? "the corresponding query statement structure is" < solution of < failed entity > in device entity >.
In fact, since all the expression manners cannot be included in table 1, when none of the fault description keywords corresponding to the text data exists in table 1, the similarity between each fault description keyword corresponding to the text data and the fault description keyword defined in table 1 may be calculated, and the query sentence structure corresponding to the fault description keyword having the highest similarity in table 1 and corresponding to the fault description keyword corresponding to the text data is taken as the query sentence structure corresponding to the text data.
The above examples are merely illustrative and are not intended to limit the present application. In practical applications, the fault reporting personnel can input more detailed text data than the above example, and the query sentence structure and the mapping relation in table 1 can be further refined or added, which is not limited in the present application.
And step S1023, obtaining a fault query statement according to the fault description keyword and the query statement structure.
Specifically, the fault query statement can be obtained by filling part or all of the fault description keywords to corresponding positions in the corresponding query statement structure, and the filled keywords and the filled positions thereof depend on the structure of the fault query statement.
In the previous example, the fault description keywords are "device 1", "occurrence", "fault a" and "what, and the corresponding query statement structure is" solution of < fault entity > in < device entity >, so that the "device 1" is filled to the position of the < device entity ", and the" fault a "is filled to the position of the < fault entity >, so as to obtain the" solution of fault a in device 1 "of the fault query statement.
Through the steps S1021 to S1023, the conversion between the text data of the fault description input by the fault reporting personnel and the fault query sentence recognizable by the knowledge map can be realized. In consideration of different use habits of fault reporting personnel, the application also provides a method for converting audio data and picture data such as voice data into fault query sentences, namely, the voice data and the picture data are converted into text data, and then the text data is processed according to the steps S1021 to S1023, so that the corresponding fault query sentences can be obtained. See the following two examples for details.
In an embodiment, as shown in fig. 3, when the fault description data is voice data, step S102, analyzing the fault description data input by the user by using a preset description statement generation rule to obtain a fault query statement, including:
and step S1024, performing voice recognition on the voice data to obtain corresponding voice text data.
Specifically, the speech recognition method used in this embodiment may be any method that can realize conversion between speech data and text data in the prior art, and the speech recognition method is not improved in this application, so that it is not described herein too much, and a person skilled in the art may implement step S1024 according to any speech recognition method in the prior art to convert audio data such as speech data input by an obstacle reporting person into corresponding speech text data.
And step S1021, performing semantic analysis on the voice text data to obtain fault description keywords.
Step S1022, matching the corresponding query statement structure according to the fault description keyword.
And step S1023, obtaining a fault query statement according to the fault description keyword and the query statement structure.
The specific implementation of steps S1021 to S1023 can refer to the description of the foregoing embodiments, and is not repeated here.
In an embodiment, as shown in fig. 4, when the fault description data is picture data, step S102, analyzing the fault description data input by the user by using a preset description statement generation rule to obtain a fault query statement, where the fault query statement includes:
step S1025, using a pre-created picture recognition text model to convert the picture data into picture text data.
Specifically, the picture data in this step mainly refers to a real shot picture of the faulty device. The picture recognition text model is obtained by training a training data set consisting of a large number of equipment pictures with text labels. The input data of the picture identification text model is a picture of equipment with faults input by a fault reporter, the output data is fault description text data corresponding to the content shown by the picture of the equipment, and similarly, the fault description text data can be in various forms such as phrases, sentences and paragraphs.
And step S1021, performing semantic analysis on the picture text data to obtain fault description keywords.
Step S1022, matching the corresponding query statement structure according to the fault description keyword.
And step S1023, obtaining a fault query statement according to the fault description keyword and the query statement structure.
The specific implementation of steps S1021 to S1023 can refer to the description of the foregoing embodiments, and is not repeated here.
The three embodiments respectively analyze and process the fault description data in the text format, the audio format and the picture format to obtain corresponding fault query statements. According to the fault diagnosis method and device, through the provision of multiple fault data input formats, fault reporting personnel can conveniently select different modes to realize accurate description of the fault, and the accuracy of the fault diagnosis result is improved.
In one embodiment, as shown in fig. 5, the step of creating the picture recognition text model comprises:
step S201, acquiring a plurality of failure device pictures and corresponding description sentences to obtain a training data set.
Specifically, the fault equipment pictures are a large number of fault equipment pictures collected in historical working time, and description sentences corresponding to each fault equipment picture are given and calibrated by multiple experts together. The expert determines the description sentence corresponding to the picture of the fault equipment according to a uniform description rule, wherein the description rule includes but is not limited to the use of professional words, the structure and the detail degree of the description sentence, and the like, and the description rule can be determined according to the difference of faults and the personal habits of the expert.
In fact, besides the faulty device picture and the description sentence corresponding to the faulty device picture, the normal device picture and the description sentence corresponding to the faulty device picture can also be used as a part of the training data set to participate in the subsequent model training, and the recognition result of the model can be more accurate.
Step S202, training a preset encoder-decoder model by using the training data set to obtain the picture recognition text model; the picture identification text model can generate corresponding picture text data according to the input fault equipment picture.
Specifically, the encoder-decoder model of the present application is an attention-based encoder-decoder model, wherein the encoder is a deep convolutional neural network, the decoder network is a stacked cyclic neural network based on a long-short term memory network structure, and the performance of the finally obtained picture recognition text model is improved by allowing the decoder to focus on the corresponding position in the image when generating the words described by the picture. In practical application, the training of the image recognition text model can be accelerated by adopting a pre-training model mode. The encoder-decoder model is trained by using a training data set consisting of the fault equipment picture and the corresponding description sentence thereof as well as the normal equipment picture and the corresponding description sentence thereof, so as to obtain the picture recognition text model based on the encoder-decoder, and the possibility of obtaining the corresponding picture text data according to the input fault equipment picture can be maximized.
In one embodiment, as shown in fig. 6, the step of establishing the failure knowledge-graph comprises:
step S301, acquiring common fault types of equipment, fault reasons corresponding to the common fault types and fault solution methods; the fault type, the fault reason and the fault solving method are all data in text format.
The fault type, the corresponding fault reason, the fault solution method and other fault data comprise actually generated historical fault data and supplementary fault data provided by experts and workers. The fault types in the fault data need to contain as many fault types as possible to ensure the comprehensiveness and availability of the finally obtained knowledge graph.
In the above fault data, for the same fault type, there may be more than one corresponding fault cause, and similarly, there may be more than one corresponding fault solution.
The essence of the step is to acquire a data set for creating the knowledge graph, and after the data set is obtained, the steps of knowledge extraction, knowledge representation, knowledge fusion and the like can be performed on the data in the data set to obtain the knowledge graph.
Step S302, the fault type, the fault reason, the fault solution and the corresponding relation are mined through a data mining method, and a plurality of entities and the relation among the entities are obtained.
Specifically, the plurality of entities include a fault entity established according to a fault type, a generation cause entity established according to a fault cause, and a solution entity established according to a fault solution, wherein a relationship between the fault entity and the generation cause entity is "generation, cause", and a relationship between the fault entity and the solution entity is "solution, remove". One of the attributes of the fault entity is a name, which can be a fault name; the attribute of the generation reason entity comprises a name, content and the like, wherein the content is the specific reason for generating the fault; the attributes of the solution entity include name, content, etc., and the content is a specific solution for solving the fault.
In this step, knowledge extraction is actually performed on the data set in step S301, and mainly includes entity extraction, attribute and attribute value extraction, and relationship extraction. Since the entity is the most basic element in the knowledge graph, the completeness, accuracy, recall rate and the like of the extraction of the entity directly influence the quality of the construction of the knowledge graph. Therefore, entity extraction is the most basic and critical step in knowledge extraction. For example, the entity extraction method adopted by the present application includes but is not limited to a rule and dictionary based extraction method, a statistical machine learning based extraction method, and the like, and the rule based method generally needs to write a template for a target entity and then perform matching in an original corpus; the method based on statistical machine learning is mainly to train the original corpus by the machine learning method and then to identify the entity by the trained model. The attributes and attribute value extraction of the present application can be extracted from sentences using manually defined or automatically generated patterns. The specific method of relationship extraction adopted in the present application includes, but is not limited to, a deep hidden relationship extraction method based on Markov logic network and ontology reasoning.
Step S303, carrying out entity disambiguation and parallel similarity calculation on the plurality of entities to obtain the parallel similarity of each entity.
The parallel similarity calculation method used in the present application may be, for example, a distributed similarity method (distributed similarity), a pattern Matching method (pattern Matching), or other parallel similarity calculation methods in the related art.
And step S304, establishing a fault knowledge graph according to the plurality of entities, the relationship among the entities and the similarity of the entities.
Based on the fault entities, the generation cause entities, the solution method entities and the relations between the entities obtained in step S302 and the parallel similarities of the entities obtained in step S303, a complete knowledge graph is constructed. The knowledge map can quickly locate the fault type according to a fault query statement converted from fault description data input by fault reporting personnel, query the generation reason or solution corresponding to the fault type and feed back the result to the fault reporting personnel.
In an embodiment, as shown in fig. 7, in step S103, obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge map, where the fault diagnosis result includes:
Step S1031, inputting the fault query statement into the fault knowledge map, and determining a target fault entity corresponding to the fault query statement and a plurality of result fault entities corresponding to the target fault entity.
Specifically, the target fault entity is determined according to the content corresponding to the < device entity > and the < fault entity > in the input fault query statement. The result fault entity comprises two types, one is a generation reason of the target fault entity, the other is a solution of the target fault entity, when the query statement structure of the fault query statement is "< the solution of the fault entity in the equipment entity >, the relation between the target fault entity and the result fault entity is determined to be 'solution' from the fault query statement, therefore, the knowledge graph queries the solution capable of solving the corresponding fault of the target fault entity based on the relation; when the query statement structure of the fault query statement is "< the generation cause of the fault entity in the device entity >, it is determined from the fault query statement that the relationship between the target fault entity and the result fault entity is" generated ", and therefore, the knowledge graph can solve the generation cause of the fault corresponding to the target fault entity based on the query of the relationship.
Step S1032, output the multiple result fault entities according to the sequence from high similarity to low similarity between the target fault entity and the result fault entity, and obtain the fault diagnosis result.
In view of various reasons causing the same fault on the same equipment, the method for solving the same fault on the same equipment is more than one, therefore, the method outputs a plurality of result fault entities for a user to refer according to the sequence of the similarity between a target fault entity and a result fault entity from high to low, wherein the content of the result fault entity with the highest similarity is the optimal result. The user can jointly judge the generation reason and the solution of the equipment fault by combining the reference result output by the knowledge graph and the working experience of the user.
The self-service fault diagnosis method based on the knowledge graph can help workers of enterprise administration units to quickly locate fault reasons and a fault solution method, and technical operation and maintenance personnel can take the initiative to process faults according to recommended optimal solution steps. The method and the device can greatly improve the fault repairing speed, improve the technical operation and maintenance levels of different regions governed by the enterprise, and optimize the fault processing capacity of the enterprise.
Based on the same inventive concept, the embodiment of the present application further provides a self-service fault diagnosis device based on a knowledge graph, which can be used to implement the methods described in the above embodiments, as described in the following embodiments. The self-service fault diagnosis device based on the knowledge graph has the advantages that the problem solving principle is similar to that of the self-service fault diagnosis method based on the knowledge graph, so the implementation of the self-service fault diagnosis device based on the knowledge graph can refer to the implementation of the self-service fault diagnosis method based on the knowledge graph, and repeated parts are not repeated. As used hereinafter, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. While the system described in the embodiments below is preferably implemented in software, implementations in hardware, or a combination of software and hardware are also possible and contemplated.
In a second aspect, the present application provides a self-service fault diagnosis apparatus based on a knowledge graph, as shown in fig. 8, the apparatus comprising:
a data obtaining module 801, configured to obtain fault description data input by a user, where the fault description data includes at least one of text data, voice data, and picture data;
A fault query statement generating module 802, configured to analyze fault description data input by a user using a preset description statement generating rule to obtain a fault query statement;
and the fault diagnosis module 803 is configured to obtain a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge graph, where the fault diagnosis result includes a fault cause or a fault solution.
In one embodiment, as shown in fig. 9, the fault query statement generation module 802 includes:
the semantic parsing unit 8021 is configured to, when the fault description data is text data, perform semantic parsing on the text data to obtain a fault description keyword;
a query statement structure matching unit 8022, configured to determine, according to the mapping relationship between the fault description keyword and the query statement structure, a query statement structure corresponding to the fault description keyword;
a fault query statement generating unit 8023, configured to obtain a fault query statement according to the fault description keyword and the query statement structure.
In an embodiment, as shown in fig. 10, the fault query statement generating module 802 further includes a speech converting unit 8024, configured to perform speech recognition on the speech data to obtain corresponding speech text data when the fault description data is speech data; and then the semantic analysis unit, the query sentence structure matching unit and the fault query sentence generating unit are used for respectively processing the voice text data to obtain a fault query sentence corresponding to the voice data.
In an embodiment, please continue to refer to fig. 10, the fault query statement generating module further includes an image converting unit 8025, configured to, when the fault description data is picture data, convert the picture data into picture text data by using a pre-created picture recognition text model; and then the semantic analysis unit, the query sentence structure matching unit and the fault query sentence generating unit are used for respectively processing the picture text data to obtain a fault query sentence corresponding to the picture data.
In an embodiment, as shown in fig. 11, the self-service fault diagnosis apparatus based on a knowledge graph further includes a model building module 804 for creating the picture recognition text model, where the model building module 804 is specifically configured to:
acquiring a plurality of fault equipment pictures and corresponding description sentences to obtain a training data set;
training a preset encoder-decoder model by using the training data set to obtain the picture recognition text model; the picture identification text model can generate corresponding picture text data according to the input fault equipment picture.
In an embodiment, with continued reference to fig. 11, the self-service troubleshooting apparatus based on a knowledge graph further includes a knowledge graph establishing module 805 configured to create the failure knowledge graph, where the knowledge graph establishing module 805 is specifically configured to:
Acquiring common fault types of equipment and fault reasons and fault solving methods corresponding to the common fault types; the fault type, the fault reason and the fault solving method are all data in text format;
mining the fault type, the fault reason, the fault solution and the corresponding relation thereof by a data mining method to obtain a plurality of entities and the relation among the entities;
carrying out entity disambiguation and parallel similarity calculation on the entities to obtain parallel similarity of the entities;
and establishing a fault knowledge graph according to the plurality of entities, the relationship among the entities and the similarity of the entities.
In an embodiment, the fault diagnosis module 803 is specifically configured to:
inputting the fault query statement into the fault knowledge graph, and determining a target fault entity corresponding to the fault query statement and a plurality of result fault entities corresponding to the target fault entity;
and outputting the plurality of result fault entities according to the sequence from high similarity to low similarity of the target fault entity and the result fault entity to obtain the fault diagnosis result.
The self-service fault diagnosis device based on the knowledge graph can help workers of the enterprise administration units to quickly locate fault reasons and a fault solving method, and technical operation and maintenance personnel can take the initiative in fault treatment according to recommended optimal solving steps. The method and the device can greatly improve the fault repairing speed, improve the technical operation and maintenance levels of different regions governed by the enterprise, and optimize the fault processing capacity of the enterprise.
In a third aspect, the present invention further provides an electronic device, and referring to fig. 12, the electronic device 100 specifically includes:
a central processing unit (processor)110, a memory (memory)120, a communication module (Communications)130, an input unit 140, an output unit 150, and a power supply 160.
The memory (memory)120, the communication module (Communications)130, the input unit 140, the output unit 150 and the power supply 160 are respectively connected to the central processing unit (processor) 110. The memory 120 stores a computer program, the central processing unit 110 can call the computer program, and the central processing unit 110 implements all steps of the script-based API dynamic programming method in the above embodiments when executing the computer program.
In a fourth aspect, embodiments of the present application further provide a computer-readable storage medium for storing a computer program, the computer program being executable by a processor. The computer program, when executed by a processor, implements any of the script-based API dynamic orchestration methods provided by the present invention.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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.
As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description 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 so forth) having computer-usable program code embodied therein. The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment. In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of an embodiment of the specification.
In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction. The above description is only an example of the embodiments of the present disclosure, and is not intended to limit the embodiments of the present disclosure. Various modifications and variations to the embodiments described herein will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present specification should be included in the scope of the claims of the embodiments of the present specification.
Claims (10)
1. A self-service fault diagnosis method based on knowledge graph is characterized by comprising the following steps:
acquiring fault description data input by a user, wherein the fault description data comprises at least one of text data, voice data and picture data;
analyzing fault description data input by a user by using a preset description statement generation rule to obtain a fault query statement;
and obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge map, wherein the fault diagnosis result comprises a fault reason or a fault solution.
2. The self-service fault diagnosis method based on the knowledge graph of claim 1, wherein when the fault description data is text data, the fault description data input by a user is analyzed by using a preset description statement generation rule to obtain a fault query statement, and the method comprises the following steps:
performing semantic analysis on the text data to obtain fault description keywords;
determining a query statement structure corresponding to the fault description keyword according to the mapping relation between the fault description keyword and the query statement structure;
and obtaining a fault query statement according to the fault description keyword and the query statement structure.
3. The self-service fault diagnosis method based on the knowledge graph of claim 1, wherein when the fault description data is voice data, the analyzing the fault description data input by the user by using a preset description statement generation rule to obtain a fault query statement comprises:
carrying out voice recognition on the voice data to obtain corresponding voice text data;
performing semantic analysis on the voice text data to obtain fault description keywords;
matching a corresponding query statement structure according to the fault description keyword;
And obtaining a fault query statement according to the fault description keyword and the query statement structure.
4. The self-service fault diagnosis method based on the knowledge graph of claim 1, wherein when the fault description data is picture data, the fault description data input by a user is analyzed by using a preset description statement generation rule to obtain a fault query statement, and the method comprises the following steps:
converting the picture data into picture text data by using a pre-established picture recognition text model;
semantic analysis is carried out on the picture text data to obtain fault description keywords;
matching a corresponding query statement structure according to the fault description keyword;
and obtaining a fault query statement according to the fault description keyword and the query statement structure.
5. The self-service intellectual graph fault diagnosis method according to claim 4, wherein the step of creating the picture recognition text model comprises:
acquiring a plurality of fault equipment pictures and corresponding description sentences to obtain a training data set;
training a preset encoder-decoder model by using the training data set to obtain the picture recognition text model; the picture identification text model can generate corresponding picture text data according to the input fault equipment picture.
6. The self-service fault diagnosis method based on knowledge-graph according to any of claims 1 to 5, characterized in that the step of establishing the fault knowledge-graph comprises:
acquiring common fault types of equipment and fault reasons and fault solving methods corresponding to the common fault types; the fault type, the fault reason and the fault solution are all data in text format;
mining the fault type, the fault reason, the fault solution and the corresponding relation thereof by a data mining method to obtain a plurality of entities and the relation among the entities;
carrying out entity disambiguation and parallel similarity calculation on the entities to obtain parallel similarity of the entities;
and establishing a fault knowledge graph according to the plurality of entities, the relationship among the entities and the similarity of the entities.
7. The self-service fault diagnosis method based on the knowledge graph of claim 6, wherein obtaining the fault diagnosis result corresponding to the fault description data according to the fault query statement and the fault knowledge graph established in advance comprises:
inputting the fault query statement into the fault knowledge graph, and determining a target fault entity corresponding to the fault query statement and a plurality of result fault entities corresponding to the target fault entity;
And outputting the plurality of result fault entities according to the sequence from high similarity to low similarity of the target fault entity and the result fault entity to obtain the fault diagnosis result.
8. A self-service fault diagnosis device based on knowledge graph, characterized by comprising:
the data acquisition module is used for acquiring fault description data input by a user, wherein the fault description data comprises at least one of text data, voice data and picture data;
the fault query statement generation module is used for analyzing fault description data input by a user by using a preset description statement generation rule to obtain a fault query statement;
and the fault diagnosis module is used for obtaining a fault diagnosis result corresponding to the fault description data according to the fault query statement and a pre-established fault knowledge map, wherein the fault diagnosis result comprises a fault reason or a fault solution.
9. An electronic device, comprising:
the self-service fault diagnosis method based on the knowledge graph comprises a central processing unit, a storage and a communication module, wherein a computer program is stored in the storage, the central processing unit can call the computer program, and when the central processing unit executes the computer program, the self-service fault diagnosis method based on the knowledge graph in any one of claims 1 to 7 is realized.
10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method of self-service knowledge-graph-based fault diagnosis of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210342224.5A CN114841347A (en) | 2022-04-02 | 2022-04-02 | Self-service fault diagnosis method and device based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210342224.5A CN114841347A (en) | 2022-04-02 | 2022-04-02 | Self-service fault diagnosis method and device based on knowledge graph |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114841347A true CN114841347A (en) | 2022-08-02 |
Family
ID=82563353
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210342224.5A Pending CN114841347A (en) | 2022-04-02 | 2022-04-02 | Self-service fault diagnosis method and device based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114841347A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115033679A (en) * | 2022-08-10 | 2022-09-09 | 深圳联友科技有限公司 | Method for searching automobile maintenance data based on knowledge graph |
| CN116893924A (en) * | 2023-09-11 | 2023-10-17 | 江西南昌济生制药有限责任公司 | Equipment fault processing method, device, electronic equipment and storage medium |
| CN117171365A (en) * | 2023-11-02 | 2023-12-05 | 北京纷扬科技有限责任公司 | Intelligent fault problem positioning method and system based on knowledge graph |
-
2022
- 2022-04-02 CN CN202210342224.5A patent/CN114841347A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115033679A (en) * | 2022-08-10 | 2022-09-09 | 深圳联友科技有限公司 | Method for searching automobile maintenance data based on knowledge graph |
| CN115033679B (en) * | 2022-08-10 | 2023-01-13 | 深圳联友科技有限公司 | Method for searching automobile maintenance data based on knowledge graph |
| CN116893924A (en) * | 2023-09-11 | 2023-10-17 | 江西南昌济生制药有限责任公司 | Equipment fault processing method, device, electronic equipment and storage medium |
| CN116893924B (en) * | 2023-09-11 | 2023-12-01 | 江西南昌济生制药有限责任公司 | Equipment fault processing method, device, electronic equipment and storage medium |
| CN117171365A (en) * | 2023-11-02 | 2023-12-05 | 北京纷扬科技有限责任公司 | Intelligent fault problem positioning method and system based on knowledge graph |
| CN117171365B (en) * | 2023-11-02 | 2024-02-02 | 北京纷扬科技有限责任公司 | Intelligent fault problem positioning method and system based on knowledge graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114841347A (en) | Self-service fault diagnosis method and device based on knowledge graph | |
| US11847422B2 (en) | System and method for estimation of interlocutor intents and goals in turn-based electronic conversational flow | |
| US11270081B2 (en) | Artificial intelligence based virtual agent trainer | |
| US11418461B1 (en) | Architecture for dynamic management of dialog message templates | |
| JP2019020893A (en) | Non-factoid type question answering machine | |
| TW202009749A (en) | Human-machine dialog method, device, electronic apparatus and computer readable medium | |
| US8818795B1 (en) | Method and system for using natural language techniques to process inputs | |
| WO2021218029A1 (en) | Artificial intelligence-based interview method and apparatus, computer device, and storage medium | |
| WO2018171953A1 (en) | A method for retrieving a recommendation from a knowledge database of a ticketing system | |
| CN105706092B (en) | The method and system of four values simulation | |
| CN111930792A (en) | Data resource labeling method and device, storage medium and electronic equipment | |
| CN112270182A (en) | Analysis idea derivation method, system and storage medium based on multi-turn dialogue question-answer mode | |
| US20220350690A1 (en) | Training method and apparatus for fault recognition model, fault recognition method and apparatus, and electronic device | |
| CN117331730A (en) | Fault diagnosis method, device, electronic equipment and storage medium | |
| CN110738056A (en) | Method and apparatus for generating information | |
| CN113326367A (en) | Task type dialogue method and system based on end-to-end text generation | |
| CN117035099A (en) | Enterprise knowledge base question-answer pair generation method based on large model and knowledge graph | |
| CN115510213A (en) | Question answering method and system for working machine and working machine | |
| CN115269862A (en) | Electric power question-answering and visualization system based on knowledge graph | |
| Dikshit et al. | Automating Questions and Answers of Good and Services Tax system using clustering and embeddings of queries | |
| CN117453895B (en) | Intelligent customer service response method, device, equipment and readable storage medium | |
| US20240127514A1 (en) | Artificial Intelligence Model Generation of Meeting Tapestries | |
| Abela et al. | Development of an augmentative and alternative communication app for the Maltese language | |
| WO2023119520A1 (en) | Estimation device, estimation method, and program | |
| US20220101094A1 (en) | System and method of configuring a smart assistant for domain specific self-service smart faq agents |
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 |