CN103699542A - Construction Method of Standard Ontology Library of Natural Gas and Pipeline Technology - Google Patents

Abstract

The invention discloses a natural gas and pipeline technical standard ontology library construction method, and relates to the technical field of digital data processing devices and pipeline systems. The method comprises (1) determining the domain and range of the body; (2) collecting and analyzing domain information; (3) determining a concept; in the determination of concepts, synonyms are supplemented; (4) establishing a body frame; (5) ontology custom integration, including the reference of existing ontologies, and the integration of new ontologies; (6) determining a concept logic relationship; when the concept logic relationship is determined, the existing ontology is combined; (7) establishing a complete body surface; (8) confirming and evaluating; (9) evolution; participating in (3) determination of concepts and (6) determination of concept logical relations after evolution; (10) and finishing the establishment of the ontology. The ontology base established by the invention can realize efficient standard information retrieval from 'basic field information' to 'important technical indexes'.

Description

Construction Method of Standard Ontology Library of Natural Gas and Pipeline Technology

technical field

The invention relates to a method for constructing a natural gas and pipeline technology standard ontology library, and relates to the technical fields of digital data processing devices and pipeline systems.

Background technique

With the development of information technology and network technology, information sharing systems have been widely used in various industries, greatly improving the labor efficiency and convenience of various industries. The natural gas and pipeline industries have been at the forefront of the industry in the application of information technology. With the With the further development of business and technology, it is no longer enough to provide functions such as document-level retrieval and browsing to meet the needs of users, and further excavation and services must be carried out to meet user needs. At present, the commonly used standard retrieval method is "basic field information" retrieval, which generally can only realize full-text retrieval by searching standard names and subject terms, and cannot realize precise positioning and retrieval of technical indicators, and comparison of the same technical indicators in different standards . To sum up, traditional retrieval methods have limitations in the following aspects in terms of the effect of using technical standards.

(1) It is impossible to accurately retrieve the content of technical standards

The traditional database retrieval method is to process the bibliographic data of standard documents by means of classification, title, abstract and descriptors to realize the retrieval of technical standards and technical regulations. However, technical indicators are generally scattered in different technical standards and technical regulations. The traditional search method can only retrieve relevant standards through bibliographic databases, and read the content of the original technical indicators one by one. But such a method is time-consuming, and it is difficult to guarantee the recall rate.

(2) The technical indicators of different standards cannot be retrieved at the same time, and the comparison of the same technical indicators of different standards can be realized.

During the retrieval process, it often happens that the technical indicators of the same product exist in different standards such as international standards, national standards, industry standards, local standards and enterprise standards, and users often need to compare the technical indicators of the same product in different standards research, which cannot be satisfied by traditional retrieval methods.

Standard content extraction and display system is the latest development direction of standard information retrieval. This new type of retrieval method can realize efficient standard information retrieval from "basic field information" to "important technical indicators" through the systematic extraction and effective organization of standard technical indicators. For project managers and implementers in charge of oil and gas pipeline construction, key indicators can be used to control pipeline design and construction; for oil and gas pipeline operators and managers, query and comparison of operating parameters and methods can be realized; for scientific research personnel , which can realize the difference analysis of key indicators of domestic and foreign standards, judge the technical differences, and analyze the coordination among the standards in the system.

CN102591878A discloses a method for establishing a technical standard content extraction and display system, and "Petroleum Planning and Design", Volume 22, No. 6, 2011, "Research on the Development Scheme of Natural Gas and Pipeline Standard Information Management System" discloses a natural gas and pipeline standard information The development plan of the management system, but the system can only search the full text of the standard, and cannot extract and display the content of the standard. It has not disclosed the construction method of the natural gas and pipeline standard ontology library, and its technology is not perfect.

Standard content extraction and display technology is a new standard retrieval technology, and there is no commercial database developed by this technology in foreign countries. In China, only the China National Institute of Standardization has initially applied standard content extraction and display technology to national standards and industry standards for food and agricultural products, and built a corresponding extraction and display system platform to realize the extraction and display of standard content indicators. At present, there is no method of establishing a content extraction and display system for natural gas and pipeline technical standards at home and abroad.

The oil and gas pipeline standard ontology library is a standardized object, which can cover all effective retrieval objects in natural gas and pipeline standards, and can define the positions of different ontology objects and their affiliation relationships through the ontology library, which can play an important role in the retrieval of standard content indicators. to support.

To realize the extraction and display of the technical content of natural gas and pipeline standards, it is necessary to summarize the standard objects in the technical standards and clarify the relationship between each object, and then form a unified retrieval rule and an effective collection of retrieval points that can achieve precise positioning. In addition, as natural gas and pipeline technical standards become more and more powerful in supporting business and the scope of application of standards becomes wider and wider, it is an urgent need for users to establish a unified, standardized, and complete standardized object, that is, an ontology library, and will be used in natural gas It plays an important role in the process of information sharing and integration with the pipeline field. However, there is no ready-made ontology database available at present, and there is no effective method to extract standard objects to build ontology database after retrieval.

Therefore, decomposing the natural gas and pipeline standards and extracting the core standard objects to build ontology database is the basis for standard content extraction and display.

Contents of the invention

The purpose of the present invention is to invent an efficient standard information retrieval from "basic field information" to "important technical indicators", which is easy to understand and feasible, can effectively decompose and extract natural gas and pipeline standard objects, and build a unified and complete Construction method of natural gas and pipeline technology standard ontology library.

The present invention is a construction method of an ontology library in a natural gas and pipeline technical standard content extraction and display system. The establishment method of the natural gas and pipeline technical standard content extraction and display system is shown in Figure 1, and mainly includes the following steps:

(1) Use optical character recognition (OCR) tools to digitize the full text of natural gas and pipeline standard documents to digitize the standard documents;

(2) Establish natural gas and pipeline standard ontology database, style database, bibliographic database;

(3) Establish a content database of natural gas and pipeline standards;

(4) Develop a natural gas and pipeline standard content extraction and display system platform for standard content extraction, display and comparison.

The specific establishment steps are:

(1) According to the digitization specification of standard documents, after digitizing the full text of the determined natural gas and pipeline standards, it also includes identification of document content and quality review, so that the standards can be edited to meet the needs of extraction and display;

(2) Establish the natural gas and pipeline standard ontology database, the natural gas and pipeline standard bibliographic database, and the natural gas and pipeline standard style database. The establishment of these three databases is a parallel process without interfering with each other;

①Establish the natural gas and pipeline standard ontology library: analyze and summarize the subject concepts of natural gas and pipeline standards, aiming at the concept connotation of the main category, according to the query results of professional knowledge and the affiliation relationship between the standardized objects enumerated in the standard documents, the ontology is carried out Conceptual analysis, clarifying the concept system and its hierarchical relationship in the standard, and establishing the natural gas and pipeline standard ontology library;

②Establishment of natural gas and pipeline style library: classify natural gas and pipeline standard documents, summarize the structured information of standards; carry out standard document style analysis according to standard documents with the same structure, abstract the core concepts and their characteristic description terms, and establish style library;

③Natural gas and pipeline standard bibliography database: analyze and summarize standard technical indicators for natural gas and pipeline technical standards, conduct terminology and conceptual relationship research on technical indicator systems, standardize technical indicator concepts, system construction and indexing, and standard literature Indexing of technical indicators, classification and labeling of style elements, establishment of a database of standard technical indicators, establishment of auxiliary databases such as quantities and units; use the Chinese standard name, English standard name, standard number, standard type, and technical direction of the technical field in the standard text , Adoption relationship, substitution relationship, superseded relationship, cited documents, standard status, project approval date, release date, implementation date, confirmation date, importance level, focal unit, drafting unit, abstract, Chinese subject terms, English subject terms, translation , Remarks, text and other information as the basis to establish a standard bibliographic database;

(3) Establish a natural gas and pipeline standard content database

The original standard system table is the same as the three databases of ontology database, bibliography database and style database established in step (2), forming a natural gas and pipeline standard content database;

(4) Develop a natural gas and pipeline standard content extraction and display system platform. In addition to general retrieval system functions, the system should also have standard content index retrieval functions such as basic retrieval functions, management functions, online feedback functions, and help functions. , Standard index processing function.

The construction process of the natural gas and pipeline standard ontology library is shown in Figure 2, which is:

(1) Determine the domain and scope of the ontology;

(2) Collection and analysis of field information;

(3) Determination of concepts; in the determination of concepts, synonyms should be added;

(4) Establish an ontology framework;

(5) Ontology custom integration, including references to existing ontologies, and integration of new ontologies;

(6) Determine the conceptual logical relationship; when determining the conceptual logical relationship, it is necessary to combine the existing ontology;

(7) Establish a complete ontology table;

(8) Confirmation and evaluation;

(9) Evolution; Participate in the determination of (3) concepts and (6) determine the logical relationship of concepts after evolution;

(10) Complete ontology establishment.

The construction process of the natural gas and pipeline standard ontology library is as follows:

(1) Determining the field and scope of the ontology library: it is necessary to clarify the professional fields to be covered by the ontology library to be built, the purpose, function and application objects of the ontology;

(2) Collection and analysis of field information: fully understand the field knowledge by collecting information in the field of oil and gas pipelines; information sources include experts, books, standards, networks and other ontologies;

(3) Determination of concepts: After fully understanding the domain knowledge of natural gas and pipelines, determine the concepts and the relationship between concepts in this domain, express them in precise terms, and confirm them with domain experts, and use them as the core concept set of domain ontology. The basic requirements that should be met are:

① The defined concepts and relationships must be domain-related; the boundaries of domains are often fuzzy, and the concepts included in the boundaries need to be determined according to actual needs;

② The terms used should be precise and have unique meanings;

③ There are corresponding natural language descriptions and synonyms for each term;

(4) Establish an ontology library framework; for a large number of concepts in the field organized in step (3), they should be grouped according to certain logical rules to form different small professional fields. In addition, the importance of each concept should be evaluated, key terms should be selected, unnecessary or beyond the scope of the field should be discarded, and the knowledge of the field should be expressed as accurately and concisely as possible;

(5) Custom integration of ontology library; when creating an ontology library, it can be customized or reuse an existing ontology library in the field; when reusing an ontology library, you need to pay attention to viewing the meta-ontology library, selecting and semantics in your own conceptual model To achieve a consistent definition of terms; the key technology involved is ontology mapping; for each integrated ontology library, its meta-ontology library, term set, formal ontology library description, and the location integrated in its own ontology library should be determined and other attributes;

(6) Determine the logical relationship of concepts; mainly based on professional knowledge and scientific classification, according to the subject method and taxonomy in taxonomy, determine the logical relationship of concepts;

(7) Establish a complete ontology library; combine the natural gas and pipeline standard ontology library with the effective retrieval points of standard documents to form a framework system of domain knowledge and obtain the framework structure of the domain ontology library;

When establishing the natural gas and pipeline ontology database, the ontology division should follow the following basic rules: a) The sum of the extensions of each subitem should be equal to the extension of the parent item; b) The extensions of the divided subitems should be mutually exclusive; c) Each time The division should be carried out according to the same principle; d) The division should be carried out step by step, from high to low, from simple to complex, and should be combined with the main business of natural gas and pipeline; e) It should be continuously updated and supplemented;

The division and setting of categories should highlight the main business, and the categories with relatively high content should be set as close as possible; for some non-exclusive categories, comprehensive basic standards with general guiding significance can be set separately according to the content category; the technical requirements of the category at the upper level must be met at the category at the lower level;

(8) Confirmation and evaluation: Ontology database should have correctness, consistency, scalability and validity;

(9) Evolution: The ontology library needs to be continuously updated during use, and the evolution of the ontology library can be by integrating new ontology libraries or defining new concepts and relationships;

(10) Complete the establishment of ontology database.

Beneficial effects of the present invention:

The invention achieves the following beneficial effects for the construction method of the natural gas and pipeline standard ontology library:

(1) The invention is simple, easy to understand and feasible, and can effectively decompose and extract natural gas and pipeline standard objects, and build a unified and complete ontology library;

(2) The ontology library constructed by the present invention is applied to the natural gas and pipeline standard content extraction and display system, which can be used as a collection of effective retrieval points to realize accurate positioning and retrieval of technical standard content, and realize the transformation from "basic field information" to "important technical indicators" efficient standard information retrieval;

(3) The ontology extracted by the present invention is accurate, unique and scientific, and can be used as a part of the terminology database, which plays an important role in information sharing and communication in the fields of natural gas and pipelines.

Description of drawings

Figure 1 Flowchart of establishing the content extraction and display system for natural gas and pipeline standards

Figure 2 Ontology library construction flow chart

Detailed ways

Embodiment. This example is an experimental method, and its flow chart is shown in Figure 2.

This example mainly includes the following steps:

(1) Determine the domain and scope of the ontology;

(2) Collection and analysis of domain information;

(3) Determination of concepts; in the determination of concepts, synonyms should be added;

(4) Establish an ontology framework;

(5) Ontology custom integration, including references to existing ontologies, and integration of new ontologies;

(6) Determine the conceptual logical relationship; when determining the conceptual logical relationship, it is necessary to combine the existing ontology;

(7) Establish a complete ontology table;

(8) Confirmation and evaluation;

(9) Evolution; Participate in the determination of (3) concepts and (6) determine the logical relationship of concepts after evolution;

(10) Complete ontology establishment.

The system table of this example is as follows:

Table 1 Sample Ontology Table

In this example, the natural gas and pipeline ontology extraction technology established can effectively extract the ontology of natural gas and pipeline field after trial. Through the systematic extraction work, a knowledge ontology system in the field of natural gas and pipeline technology is formed, which is applied to the content extraction of natural gas and pipeline standards The system and display system can realize precise positioning and retrieval of technical indicators in standard content; directly display the required standard retrieval content or technical indicators in the search results, without requiring users to read the entire document to find the required information, thereby improving It improves the retrieval efficiency; it can realize the retrieval of standards related to technical indicators; when searching for standards, you can log in through the host, and when you retrieve specific standard technical indicators, you can also retrieve other related standards.

Claims (2)

1. rock gas and a pipe technology standard body base construction method, is characterized in that flow process is:

(1) determine field and the scope of body;

(2) Collection and analysis of realm information;

(3) concept determines; In the determining of concept, supplement synonym;

(4) set up body frame;

(5) body is self-defined integrated, comprises quoting of existing body, and new body is integrated;

(6) determine concept logic relation; Determine when concept logic is related to, be in conjunction with existing body;

(7) set up this complete body surface;

(8) confirm and evaluate;

(9) evolve; After evolving, participate in determining with (6) of (3) concept and determine concept logic relation;

(10) completing body sets up.

2. rock gas according to claim 1 and pipe technology standard body base construction method, is characterized in that building flow process and be specially:

(1) determine field and the scope of ontology library: the ontology library that clearly build is by the professional domain covering, object, effect and the application of body;

(2) Collection and analysis of realm information: fully understand this domain knowledge by collecting oil and gas pipeline realm information; Information source comprises expert, books, standard, network and other body;

(3) determining of concept: after fully understanding rock gas and pipeline domain knowledge, determine the relation between concept and concept in this field, express with accurate term, through domain expert's confirmation, as the Core Set of Concepts of domain body.The requirement that substantially should meet has:

1. definite concept and relation must be domain-specifics; The border in field is fuzzy often, need determine according to the actual requirements the concept that border comprises;

2. the term adopting is wanted accurately, and implication should have uniqueness;

3. to each term, there are corresponding natural language description and synonym to supplement;

(4) set up ontology library framework; For a large amount of concept in the field arranging in step (3), they are divided into groups according to certain logic rules, form different particular specialty fields, in the concept of same little career field, its correlativity should be more intense; In addition, to the importance of each concept wherein, to assess, select key term, abandon that those are unnecessary or exceed the concept of territory, the knowledge in the field that gives expression to of as far as possible accurately simplifying;

(5) ontology library is self-defined integrated; Create ontology library can be self-defined, can be also reusing of ontology library existing in field; While reusing ontology library, should be noted that and check meta-ontology storehouse, select with semanteme in own conceptual model with realize consistent term definition; The gordian technique wherein relating to is the mapping of body; For each integrated ontology library, should determine its meta-ontology storehouse, terminology, the description of formal ontology library and be integrated in the attributes such as position in own ontology library;

(6) determine concept logic relation; Main take professional knowledge with scientific classification as basis, according to the subject indexing method in taxonomy and classification, determine the logical relation of concept;

(7) set up complete ontology library; Rock gas and piping standards ontology library and normative document are effectively retrieved to a combination, thereby form the frame system of a domain knowledge, obtain the framed structure of field ontology library;

While setting up rock gas and pipeline body storehouse, body is divided should follow following primitive rule: a) the extension sum of each subitem should equal the extension of female; B) each subitem of dividing, its extension should be repelled mutually; C) each division should be undertaken by identity principle; D) divide should by level step by step, from high to low, go from the simple to the complex and carry out, should be combined with pipeline main business thickness in conjunction with rock gas; E) answer continuous updating to supplement;

The division of classification should be given prominence to main business with setting, by the larger classification of content relevance, should close on setting as far as possible; For some, without exclusive class, and the comprehensive basic standard with general directive significance can arrange separately respectively class according to content; The technical requirement lower floor classification of last layer time classification all will meet;

(8) confirm and evaluate: ontology library should have correctness, consistance, extensibility and validity;

(9) evolve: in use need ontology library to constantly update, the mode that ontology library is evolved can be integrated new ontology library or define new concept and relation;

(10) completing ontology library sets up.

Images