KR20190017460A - semantic web potal system and search system using multi ontology. - Google Patents

Abstract

The present invention relates to a semantic search system using multi-ontology, capable of easily and correctly finding desired information of a user through an intelligent search service. According to the present invention, the semantic search system using multi-ontology comprises an ontology framework unit, a component architecture unit, a required item data collection means, a portal search service providing means, and a semantic portal search means.

Description

Semantic web search system using multi-ontology.

The present invention relates to a semantic web portal using a multiple ontology and a semantic search system using the semantic web portal. More particularly, the present invention relates to a semantic web portal system, The present invention relates to a semantic web portal using a multi-ontology and a semantic search system using the same, which provides a semantic search method that can easily and accurately search through an intelligent search service such as semantic reasoning.

The next generation Web (or Semantic Web) refers to the next generation of webs proposed by Tim Berners-Lee, the creator of the Web. The Semantic Web is now a well-defined semantic extension of the Web, It can be defined as a web that can be done.

In other words, it means semantic web environment in which information acquisition and integration, which was done through one-sided interaction between the existing person and the web, can be semantically realized between a computer and a computer between a person and a computer.

The detailed technical composition for the realization of the Semantic Web is defined by the W3C led by Tim Berners-Lee, the proponent of the Semantic Web, and presented as a Recommendation, which is used as a practical international standard.

As shown in FIG. 1, the layered Cake is a layered Cake based on a layer-by-layer basis in consideration of the inter-technology relationships.

Layer Cake is composed of technologies for integrating various information on the web under the goal of "Data Integration at Web Scale" as a method for realizing Semantic Web.

The structure of such a technology not only provides appropriate access to the context in which the information is used, but also provides a standard data expression framework in which a user or a computer can access information semantically by integrating various information.

In recent years, the source of competitive advantage has been changing from a traditional industry, which was tangible and production-oriented, to a 'knowledge-based economy,' where knowledge and other intangible assets are transformed into industry.

At the core of the process of evolving into a knowledge-based economy is a new awareness of the value of knowledge and technological innovation through it. In particular, knowledge exchange activities within and between industries are recognized as an important element of technological innovation process.

This activity of intra-industry or inter-industry exchange of knowledge is seen as a paradigm of convergence and is recognized as a key concept to create new markets globally or to maximize the competitiveness of existing industries. In a situation where users' requirements are changing rapidly, the type of convergence between industries is expected to become more diverse in future.

Semantic Web technology is the core technology of the technology infrastructure for inter-industry knowledge fusion in that it provides international standards for information sharing and integration. The semantic and integrated approach of knowledge provided by semantic web technology can integrate knowledge that is scattered and unusable in the industry into a standardized knowledge convergence system to facilitate the generation of new knowledge, And can be used as a core to provide users with higher value added services to users including enterprises.

The Semantic Web, which is attracting attention as a next generation web, is currently being applied in various fields using the Internet, providing new opportunities in the field, and leading various related researches.

Among them, semantic web portal is attracting attention as an application field integrating semantic web technologies.

In general, web portals are the starting point for web users and provide hyperlinks to useful information and services.

In addition, it has recently been used as a means of communication or information exchange.

However, these current web portals have increased inaccurate and heterogeneous information due to the explosion of information due to the development of the web.

Also, due to the limitation of current web technology, communication and information exchange problems arise.

The Semantic Web portal is a more advanced form of web portal built using the Semantic Web technology.

A Web application that provides information and services for a specific domain is based on a knowledge system that can be understood and processed by the machine by giving meaning to data and information resources.

In other words, because the semantic web portal is based on an ontology that is basically a knowledge system that enables semantic information processing, a different information search method is required because of the characteristics of this semantic web portal, which is different from the existing portal .

The present invention has been made to solve the above problems, and an object of the present invention is to reconstruct conventional information using a standard ontology to easily search hidden information desired by a user to provide a service.

Another object of the present invention is to grasp the correlation between the viewpoints or the viewpoints of each person, company, product, etc., and to extract the associativity through intelligent search service such as semantic reasoning.

Another object of the present invention is to fuse knowledge by applying various kinds of knowledge scattered on the web by applying semantic web technology in order to collect knowledge for improving competitiveness of companies.

In order to achieve the above object, there is provided a semantic web portal construction method using a multi-ontology of the present invention,

A system for searching an ontology repository defining a semantic system of knowledge fusion,

An ontology framework part for defining an ontology to be constructed for data integration, and a type and scope of an ontology to be used in a seamonic portal;

A component architecture part for deriving from each layer using a model layer, a control layer, and a view layer model, and configuring each component of the system as a component and designing the component;

A requirement data collection means for collecting analyzed data by analyzing necessary target data according to requirements in a semantic portal to derive requirements and analyzing the collected data;

Deriving the concepts based on the meta information extracted from the data collected through the requirement data collection means and defining a relationship between the derived concepts to provide a search service capable of tracking information to users by building knowledge A portal search service providing means for searching the portal search service;

Searching for and searching for seed candidates containing the keyword selected by the user by selecting a service provided by the portal search service providing means, searching for concepts related to the searched seeds, selecting only the connections to be extended, And a semantic portal search unit for outputting a search result.

As described above, the construction of the semantic web portal using the multi-ontology according to the present invention and the semantic-type search system using the same provide various knowledge scattered on the web in order to collect knowledge for improving the competitiveness of the enterprise, It is possible to create more meaningful knowledge on the company's activities through the fusion of knowledge.

In addition, when a user searches on the Semantic Web, it is not searched for a search term when a search term is entered for more accurate search, but a search result is searched for a technology, a patent, a corporation, a thesis linked to a search term through an ontology It is possible to provide more accurate and meaningful knowledge on the activities of the company, which can be usefully used for future strategies and technology development utilizing them.

In addition, the semantic portal search service provides a variety of service methods to the users, thereby enabling accurate and efficient tracking so that the user can easily find the desired information.

In addition, it not only enables users to search for desired information accurately, but also provides the effect of quickly finding necessary information through ranking of search results.

In addition, it is possible not to rely on the query language simply, but to suggest an algorithm for the user to track the search.

In addition, it provides the effect of extracting associations through intelligent search services such as semantic reasoning by grasping the interrelation among the viewpoints or viewpoints of each person, company, and product.

Also, in order to collect knowledge for enhancing competitiveness of companies, various knowledge which is scattered on the web is applied to apply the semantic web technology to provide a fusion effect of knowledge.

1 is a block diagram of a semantic web layer Cake.
FIG. 2 is a block diagram of a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and a semantic search system using the same.
3 is a block diagram of a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and an ontology framework of a semantic search system using the same.
FIG. 4 is a diagram illustrating an operation flow of a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and an ontology framework section of a semantic search system using the same.
5 is a block diagram illustrating a component architecture of a semantic search system using a multi-ontology and a semantic web portal according to an exemplary embodiment of the present invention.
FIG. 6 is a diagram illustrating an exemplary construction of a semantic web portal using a multi-ontology according to an exemplary embodiment of the present invention and components defined in a control layer of the semantic search system using the same.
FIG. 7 is a diagram illustrating an exemplary construction of a semantic web portal using a multi-ontology according to an exemplary embodiment of the present invention and components defined in a view layer of a semantic search system using the same.
FIG. 8 is a diagram illustrating an example of a conventional search method of Naver and Google.
9 is a diagram illustrating a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and a semantic search method of a semantic search system using the same.
FIG. 10 is a diagram illustrating an example in which a semantic web portal constructed using a multiple ontology according to an embodiment of the present invention and a meta information knowledge expression unit of a semantic search system using the same are represented.
FIG. 11 is a diagram illustrating a result of searching through a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and a semantic search system using the semantic web portal.
FIG. 12 is a diagram illustrating an example of building a semantic web portal using a multi-ontology according to an embodiment of the present invention and extracting results of an operation processing unit of the semantic search system using the same.
FIG. 13 is a diagram illustrating a result of processing by a central subject processing unit of a semantic type search system using the construction of a semantic web portal using a multi-ontology according to an embodiment of the present invention.
FIG. 14 is a diagram illustrating a timeline function of a semantic type search system using the construction of a semantic web portal using a multi-ontology according to an embodiment of the present invention.

According to an aspect of the present invention, there is provided a semantic web portal using a multi-ontology,

A system for searching an ontology repository defining a semantic system of knowledge fusion,

An ontology framework part for defining an ontology to be constructed for data integration, and a type and scope of an ontology to be used in a seamonic portal;

A component architecture part for deriving from each layer using a model layer, a control layer, and a view layer model, and configuring each component of the system as a component and designing the component;

A requirement data collection means for collecting analyzed data by analyzing necessary target data according to requirements in a semantic portal to derive requirements and analyzing the collected data;

Deriving the concepts based on the meta information extracted from the data collected through the requirement data collection means and defining a relationship between the derived concepts to provide a search service capable of tracking information to users by building knowledge A portal search service providing means for searching the portal search service;

Searching for and searching for seed candidates containing the keyword selected by the user by selecting a service provided by the portal search service providing means, searching for concepts related to the searched seeds, selecting only the connections to be extended, And a semantic portal search unit for outputting a search result.

Here, the requirement data collection means may include:

A requirement analysis unit for analyzing necessary target data according to requirements in the semantic portal to derive a requirement,

A data collecting unit for collecting analyzed data through the requirement analyzing unit;

And a requirement information database for storing and managing the data collected by the data collection unit.

Here, the portal search service providing means may include:

A meta information knowledge expression unit for deriving concepts based on the meta information extracted from the collected data through the requirement data collection unit and defining a relationship between the derived concepts and building knowledge about the corresponding domain,

And a portal search service providing unit for providing a search service for tracking accurate information to a user.

In this case, the semantic portal searching unit

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

And a ranking unit for setting a priority according to the number of connections between the seed and the final node of the result path.

In this case, according to an additional aspect,

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

A ranking unit for setting a priority according to the number of connections between the seed and the final node of the result path;

And an arithmetic processing unit for executing a search using an AND operator between two keywords when the user wants to search concepts related to the two words.

According to another aspect of the present invention,

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

A ranking unit for setting a priority according to the number of connections between the seed and the final node of the result path;

An operation processor for performing a search using an AND operator between two keywords when the user wants to find concepts related to two words,

And a central subject processing unit for performing a search by setting the subject as a central keyword when the user desires to obtain additional results related to one topic.

At this time, the portal search service providing unit,

A navigation service unit for searching for scattered knowledge by providing a function for navigating among information constructed in the semantic portal according to a result searched by the semantic portal searching unit,

A trend analysis service unit for reconstructing a search result based on a time line so that trends can be easily recognized according to the results searched by the semantic portal search means,

A personalization service unit for changing a search result of the semantic portal search unit according to a central keyword when a central keyword is set according to a tendency of an individual in a navigation search by the navigation service unit;

And a keyword search service unit for providing a search result by the semantic portal search unit according to a keyword when the keyword search is set in the navigation search by the navigation service unit, To the user.

At this time, the ontology framework unit,

A service analyzing unit for analyzing necessary data according to requirements in a semantic portal collection unit in the requirement data collection unit and analyzing and verifying requirements when deriving a requirement;

A domain analysis unit for analyzing a current system configuration when collecting and analyzing necessary target data according to a requirement in a semantic portal in a requirement data collection means,

And a data modeling unit for defining the class name and the structure of the data to define the class relation and then defining an attribute constraint when defining the attribute list to create an instance.

At this time,

A control layer that consists of components defined as contents of services and functions that are commonly used by each service, components for implementing the contents of services provided by the semantic portal,

A model layer for providing an associated API for use in the control layer,

And a view layer made up of components necessary for constructing a user interface.

Hereinafter, a semantic web portal using a multi-ontology according to the present invention and a semantic search system using the same will be described in detail with reference to the accompanying drawings.

FIG. 2 is a block diagram of a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and a semantic search system using the same.

As shown in Fig. 2, the system of the present invention includes:

A system for searching an ontology repository defining a semantic system of knowledge fusion,

An ontology framework 110 for defining an ontology to be constructed for data integration, and determining a type and a scope of the ontology to be used in the web portal;

A component architecture unit 120 for deriving and constructing each component of the system as a component using a model layer, a control layer, and a view layer model;

A requirement data collection means 130 for analyzing necessary target data according to requirements in the semantic portal to derive the requirements and collecting the analyzed data;

Deriving the concepts based on the meta information extracted from the data collected through the requirement data collection means and defining a relationship between the derived concepts to provide a search service capable of tracking information to users by building knowledge A portal search service providing means (140)

Searching for and searching for seed candidates containing the keyword selected by the user by selecting a service provided by the portal search service providing means, searching for concepts related to the searched seeds, selecting only the connections to be extended, And a semantic portal search unit 150 for outputting a search result.

In order to construct the system of the present invention, an ontology to be constructed for data integration is defined, an ontology framework section 110 for determining the type and scope of the ontology to be used in the portal site, a model layer, And a component architecture unit 120 for designing each component of the system as a component by constructing the component architecture unit 120 to provide a service provided by the portal search service providing unit 140 The semantic portal search unit 150 searches for seed candidates including the keyword selected by the user and searches for concepts related to the searched seeds to select and process only the connections to be extended to set priorities The search result is output.

Hereinafter, each of the above means and portions will be described in detail.

3 is a block diagram of a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and an ontology framework of a semantic search system using the same.

FIG. 4 is a diagram illustrating an operation flow of a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and an ontology framework section of a semantic search system using the same.

As shown in FIG. 3, the ontology framework section 110 includes:

A service analysis unit 110a for analyzing necessary data according to the requirements in the semantic portal in the requirement data collection unit and analyzing and verifying the requirements when deriving the requirements,

A domain analysis unit 110b for analyzing the current system configuration when collecting and analyzing necessary target data according to the requirements in the semantic portal in the requirement data collection means,

And a data modeling unit 110c for defining the class name and the structure of the data to define the class relation and then defining an attribute constraint when defining the attribute list to create an instance.

That is, the ontology framework unit 110 defines a methodology for constructing the ontology, and determines the type and scope of the ontology to be used in the semantic portal.

For example, when an electronic component company uses a service, an ontology used in a semantic portal includes a service ontology for linking data to be used in a service, a service ontology for collecting terms of the electronic component field (BLU) Domain ontology.

As the methodology, EOE (Evolving Ontology Engineering) methodology, which is proposed by Korea Computer Science Agency, is used. The EOE methodology takes into consideration the domestic development environment as an ontology construction methodology that gradually builds the ontology as the requirements become clear over time , Because the performing agency has used it several times, it is used to secure the reliability.

As shown in FIG. 4, the service analyzing unit 110a analyzes required data in the semantic portal according to the requirements in the requirement data collecting unit, analyzes and analyzes the requirements when deriving the requirements, The modeling unit 110c defines a class name and a structure of data to define a class relationship, defines an attribute list, and defines an attribute constraint to create an instance.

In addition, the domain analysis unit 110b is responsible for analyzing the current system configuration when collecting and analyzing necessary data according to requirements in the semantic portal in the requirement data collection means.

5 is a block diagram illustrating a component architecture of a semantic search system using a multi-ontology and a semantic web portal according to an exemplary embodiment of the present invention.

FIG. 6 is a diagram illustrating an exemplary construction of a semantic web portal using a multi-ontology according to an exemplary embodiment of the present invention and components defined in a control layer of the semantic search system using the same.

FIG. 7 is a diagram illustrating an exemplary construction of a semantic web portal using a multi-ontology according to an exemplary embodiment of the present invention and components defined in a view layer of a semantic search system using the same.

As shown in FIG. 5, the component architecture section 120 includes:

A control layer 120b composed of components defined as functions of services and functions commonly used by the respective services,

A model layer 120c for providing an associated API for use in the control layer,

And a view layer 120a made up of components necessary for constructing a user interface.

In other words, the component architecture part is a component configuration of the system configuration. Components are derived from each layer using MVC (Model, View, Control) model, and included in the model level for ontology. In particular, the component architecture part has the advantage that it is more flexible than other design architectures because each component of the entire system can be viewed as a component and independently designed.

It is also widely used because it reduces the complexity of document creation and program implementation in design and implementation.

The model layer 120c is composed of a service ontology and a domain ontology. The model layer 120c is designed on the basis of the service ontology and the domain ontology. The model layer 120c provides related APIs for use in the control layer 120b.

The data used in the semantic portal includes electronic newspaper articles, papers in the field of electronics, patents, and terms and company information of the electronic field (BLU) provided by the company.

As shown in FIG. 6, the control layer 120b is a component for implementing the contents of the service provided by the semantic portal, and is composed of components defined by the contents of the service and the functions commonly used in the respective services.

As shown in FIG. 7, the view layer 120a is a part that is visible to a user and is composed of components necessary for implementing a user interface.

The requirement data collection means of the present invention comprises:

A requirement analysis unit for analyzing necessary target data according to requirements in the semantic portal to derive a requirement,

A data collecting unit for collecting analyzed data through the requirement analyzing unit;

And a requirement information database for storing and managing the data collected by the data collection unit.

The requirements for building ontologies are gathered through interviews with business stakeholders, analysis of target data, and analysis of the companies that will use the semantic portal.

The data collection unit derives key concepts and relationships for constructing an ontology for each based on a number of electronic newspapers, articles, patents, and company information.

For example, we extract the major entities such as newspapers, authors, and publishers with the metadata of the electronic newspaper and define the relationship between them. In addition, since there are several classification methods such as section classification, article classification, and article classification, the article performs functions to accommodate such a classification system. After extracting title, author, publisher, classification code of an academic article And the relationship between them is extracted and defined.

In particular, domestic papers have a classification system according to the KDC code, and the title is defined as a class having both Korean and English, and the metadata of the thesis is maintained, and the original text is configured to access the Internet directly by storing the URL.

In addition, there are many information related to patents such as the filing date, application number, registration date, and registration number in addition to the information of title, author, and the like. In particular, authors have several subclasses such as inventors, applicants, agents (patent attorneys), and testers (patent examiners), as well as maintaining multiple metadata that can contain patent data.

The applied company information analyzes the relation between the part information produced by the company and the information related to the company.

The data modeling unit 110c defines the structure of the class based on the concept and basic relations derived from the data analysis process, and models the top-bottom relationship as well as the relationship between the classes.

It also defines attribute lists and attribute constraints.

In accordance with the above-described EOE methodology, it is necessary to repeatedly adjust the process of [class name / structure definition] - [class relation definition] - [property list definition] - [property constraint definition] - [instance creation] .

In order to retrieve data corresponding to a keyword desired by the user in the semantic portal configured as described above,

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

And a ranking unit for setting a priority according to the number of connections between the seed and the final node of the result path.

FIG. 8 is a diagram illustrating an example of a conventional search method of Naver and Google.

9 is a diagram illustrating a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and a semantic search method of a semantic search system using the same.

The ontology-based semantic search described in the present invention refers to a search service that displays more intelligent results using an ontology containing concepts and relationship information for information accumulated in a distributed information system.

Semantic search provides users with results of semantic search method rather than existing keyword matching method. When searching for a specific company, it starts with Seed and tracks related information related to company, thesis and newspaper related to it.

As shown in FIG. 8, in a general search engine, information is gathered and data is constructed as an inverted file in a tree form through a text-based index, and information matching with a keyword is searched for in a search request, And then return the result. So, to be a good search engine, it's up to you to index as much data as possible and to be able to show search results faster and more accurately.

Especially in case of Google, we are improving the accuracy of the search results by developing the PageRank algorithm.

However, these search engines only handle data that is publicly available on the Web, making them unsuitable for relatively closed business environments. In addition, since search engines perform searches only in their own repositories with independent repositories, they are less flexible in terms of linking and convergence of information.

On the other hand, as shown in FIG. 9, in the semantic search, an ontology is used for more accurate retrieval. In the ontology, an information entity in each information resource is specified and a relation between the information entities is described.

Therefore, when a user inputs a search term 'BLU', it is possible to search for and search for technologies, patents, companies, and articles linked to the product through an ontology instead of searching only the LED BLU and light guide plate from the product DB of the company or other companies .

FIG. 10 is a diagram illustrating an example in which a semantic web portal constructed using a multiple ontology according to an embodiment of the present invention and a meta information knowledge expression unit of a semantic search system using the same are represented.

The meta information knowledge expression unit for deriving concepts based on the meta information extracted from the data collected through the requirement data collection means and defining the relationship between the derived concepts and constructing the knowledge about the corresponding domain includes an ontology- In order to enable semantic search, it derives concepts based on meta information extracted from distributed information resources, and defines the relationship between these concepts to construct knowledge about the domain.

This knowledge is expressed through the ontology. In Fig. 10, the nodes represent the instances of the ontology, and the relationships between the instances are connected to the arc. This means that the instance is linked to other instances.

Meanwhile, the portal search service provider of the present invention,

A navigation service unit for searching for scattered knowledge by providing a function for navigating among information constructed in the semantic portal according to a result searched by the semantic portal searching unit,

A trend analysis service unit for reconstructing a search result based on a time line so that trends can be easily recognized according to the results searched by the semantic portal search means,

A personalization service unit for changing a search result of the semantic portal search unit according to a central keyword when a central keyword is set according to a tendency of an individual in a navigation search by the navigation service unit;

And a keyword search service unit for providing a search result by the semantic portal search unit according to a keyword when the keyword search is set in the navigation search by the navigation service unit, To the user.

The semantic portal of the present invention provides four types of search services, namely, a navigation service, a trend analysis service, a personalization service, and a keyword search service.

The core of the search for a semantic portal proposed in the present invention is to provide a service that enables a user to track accurate information based on an ontology. Therefore, among the above four services, a service focused on the semantic portal is a navigation service.

Therefore, in order to maximize the function of the navigation service, other services operate as a part of the navigation service.

The navigation service unit provides a function of navigating among the information constructed in the portal, thereby enabling efficient search of scattered knowledge.

The trend analysis service unit reconstructs search results based on a timeline so that trends can be easily grasped from the navigation search results of news, articles, and patents.

The personalization service unit provides a " central keyword " function so that search can be reconfigured according to the tendency of the individual at the time of navigation search.

That is, when the user sets a keyword to be searched by himself or herself, all the searches to be performed later are changed according to the central keyword.

The keyword search service unit is configured to accommodate a keyword search algorithm such as a general search engine in conjunction with a navigation search service.

Therefore, the navigation search service has not only a general keyword search service but also a semantic search function based on the ontology.

Wherein the semantic portal searching means comprises:

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

And a ranking unit for setting a priority according to the number of connections between the seed and the final node of the result path.

In other words, the essence of semantic search is to enable the user to easily find the information they want by allowing them to accurately and efficiently track related information based on the ontology when the user enters a search word. This is different from simple keyword-based mapping method in existing search engine and it can be a method to maximize linkage and convergence of information through search result tracking technique.

When the search keyword is inputted by the user, the semantic portal search means searches for the ontology with the search keyword. In the first step, a keyword-based matching method such as a general search engine is used as a basis. However, in addition to simple keyword matching, if a query is a class, it also finds the synonym of the class, and also gets information about the subclass of the class. In this way, the related ontology instance is retrieved using the class hierarchy of the ontology. The set of instances is called the Seed instance in the sense that it is the start (seed) of the search.

The seed search result from the ontology seed search unit searches the concepts related to the found seeds and defines a hierarchical structure between the seeds by a semantic search processing unit for selecting and processing only the links to be extended.

That is, the semantic search processing unit tracks related information for each Seed instance.

The ontology basically has a triple-based graph structure called SPO (Subject, Predicate, Object)

Tracking information follows the graph of this ontology and finds the desired information.

This cycle is performed for a specified depth, and when a cycle occurs on a specific path, the path stops traversing.

The information gathered through the ontology tracking through the ranking unit to set the priority according to the number of connections between the final nodes of the result path is first classified into seven representative groups defined in the semantic web portal (news, patent, papers, company, Product, technology).

If the user inputs two or more search terms and an AND operation is performed, the duplicate result of the search result is removed, and then ranking and sorting are performed for each group.

After the ranking is performed, grouping of the same information among the applied results is performed to simplify the result displayed in the browser.

In addition, the ranking algorithm used in the semantic portal is Word Similarity, which multiplies Word Similarity, which calculates how similar the search result is to the Seed result, by Ranking Constant, and adding the Navigation Path Weight to give a higher ranking to be.

FIG. 11 is a diagram illustrating a result of searching through a semantic web portal construction using a multi-ontology according to an embodiment of the present invention and a semantic search system using the semantic web portal.

The search result is output by portal search service provider. Each search result shows the extracted result similar to the input information by pressing the "+" button.

Additional tracking results show that Ajax technology can help users reduce the amount of time they feel, and in addition to viewing additional information, paging can also improve user experience with Ajax technology.

Since the Ajax technique is already well known to those skilled in the art, a detailed description thereof will be omitted.

FIG. 12 is a diagram illustrating an example of building a semantic web portal using a multi-ontology according to an embodiment of the present invention and extracting results of an operation processing unit of the semantic search system using the same.

In the case where the user of the present invention desires to find concepts related to two words, the operation processing unit for performing a search using the AND operator between two keywords is different from the keyword search method which is an existing search method.

The method of keyword retrieval method returns all the results that exist for the inputted keywords. However, the And operation provided by the semantic portal is a method of searching the nodes that are commonly visited in the process of being traced through each seed, .

This allows you to discover new information that is related to both operands, rather than just the information that contains the two operands.

FIG. 12 is a diagram illustrating a method of performing an And operation and a method of extracting a result of an And operation.

FIG. 13 is a diagram illustrating a result of processing by a central subject processing unit of a semantic type search system using the construction of a semantic web portal using a multi-ontology according to an embodiment of the present invention.

The central subject processing unit is responsible for performing a search by setting the subject as a central keyword in order to obtain additional results related to one subject.

The central keyword search is a search method that provides information related to the Seed instance selected as the central keyword in the semantic search result value.

In this way, new relevance can be derived through the central keyword and the input keyword, and there is an advantage that new knowledge can be provided to the user. 13 shows a method of registering a central keyword and a search result screen in a state registered as a central keyword.

FIG. 14 is a diagram illustrating a timeline function of a semantic type search system using the construction of a semantic web portal using a multi-ontology according to an embodiment of the present invention.

The timeline function served by the trend analysis service unit is a function provided only for a thesis, a news, and a patent, and provides a timeline in which each document appears.

Preferably, the Timeline API is implemented using the Timeline API developed by the MIT SMILE lab.

At present, there is a lot of information accumulated in various industries in Korea. In the meantime, various studies and projects have been carried out to efficiently share such information, but the fusion of knowledge has not succeeded.

Accordingly, a new paradigm for knowledge sharing has been required to overcome this problem. Through the present invention, it is possible to provide a means for sharing and linking knowledge with the semantic web, which has been popular as a next generation web.

Knowledge fusion method in semantic web is using ontology.

Semantic Web ontology is described using Web Ontology Language (OWL), a web ontology language that is standardized by the W3C. The language follows the RDF (Resource Description Framework) format, which is a standard for describing Web resources. It is based on the Description Logics and Frame language. As described above, the ontology language based on the standard is proposed in the present invention, and the scalability and compatibility of knowledge to be constructed can be enhanced.

It will be appreciated by those skilled in the art that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is to be understood, therefore, that the embodiments described above are to be considered in all respects as illustrative and not restrictive.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

110: ontology framework part
120: Component architecture section
130: Requirements data collection means
140: portal search service providing means
150: Semantic portal search means

Claims (9)

A system for searching an ontology repository defining a semantic system of knowledge fusion,

An ontology framework part for defining an ontology to be constructed for data integration, and a type and scope of an ontology to be used in a seamonic portal;

A component architecture part for deriving from each layer using a model layer, a control layer, and a view layer model, and configuring each component of the system as a component and designing the component;

A requirement data collection means for collecting analyzed data by analyzing necessary target data according to requirements in a semantic portal to derive requirements and analyzing the collected data;

Deriving the concepts based on the meta information extracted from the data collected through the requirement data collection means and defining a relationship between the derived concepts to provide a search service capable of tracking information to users by building knowledge A portal search service providing means for searching the portal search service;

Searching for and searching for seed candidates containing the keyword selected by the user by selecting a service provided by the portal search service providing means, searching for concepts related to the searched seeds, selecting only the connections to be extended, And a semantic portal search means for outputting a search result. The semantic web search system using the semantic web portal is constructed by using the multiple ontology.
The method according to claim 1,

Wherein the requirement data collection means comprises:

A requirement analysis unit for analyzing necessary target data according to requirements in the semantic portal to derive a requirement,

A data collecting unit for collecting analyzed data through the requirement analyzing unit;

And a requirement information database for storing and managing the data collected by the data collecting unit, and constructing a semantic web portal using the multiple ontology and a semantic retrieval system using the same.
The method according to claim 1,

Wherein the portal search service providing means comprises:

A meta information knowledge expression unit for deriving concepts based on the meta information extracted from the collected data through the requirement data collection unit and defining a relationship between the derived concepts and building knowledge about the corresponding domain,

And a portal search service providing unit for providing a search service for tracking accurate information to a user. The semantic web portal using the multiple ontology and the semantic search system using the same.
The method according to claim 1,

Wherein the semantic portal searching means comprises:

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

And a ranking unit for setting priorities according to the number of connections between the seed and the final node of the result path. The semantic web portal system using the multiple ontology and the semantic search system using the same.
The method according to claim 1,

Wherein the semantic portal searching means comprises:

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

A ranking unit for setting a priority according to the number of connections between the seed and the final node of the result path;

And a calculation processing unit for performing a search using an AND operator between two keywords when the user wants to find concepts related to the two words. The construction of the semantic web portal using the multiple ontology and the use thereof brother
Search system.
The method according to claim 1,

Wherein the semantic portal searching means comprises:

An ontology seed search unit for searching, in the requirement information database, seed candidates including a keyword input by a user according to a service selected by the portal search service providing unit;

A semantic search processor for searching for concepts related to the seeds retrieved through the ontology seed searcher to select and process only the links to be extended,

A ranking unit for setting a priority according to the number of connections between the seed and the final node of the result path;

An operation processor for performing a search using an AND operator between two keywords when the user wants to find concepts related to two words,

And a central subject processing unit for performing a search by setting the subject as a central keyword when the user desires to obtain additional results related to a single topic, and constructing a semantic web portal using the multi-ontology Semantic search system.
The method of claim 3,

The portal search service providing unit,

A navigation service unit for searching for scattered knowledge by providing a function for navigating among information constructed in the semantic portal according to a result searched by the semantic portal searching unit,

A trend analysis service unit for reconstructing a search result based on a time line so that trends can be easily recognized according to the results searched by the semantic portal search means,

A personalization service unit for changing a search result of the semantic portal search unit according to a central keyword when a central keyword is set according to a tendency of an individual in a navigation search by the navigation service unit;

And a keyword search service unit for providing a search result by the semantic portal search unit according to a keyword when the keyword search is set in the navigation search by the navigation service unit, And a semantic search system using the constructed semantic web portal.
The method according to claim 1,

The ontology framework unit includes:

A service analyzing unit for analyzing necessary data according to requirements in a semantic portal collection unit in the requirement data collection unit and analyzing and verifying requirements when deriving a requirement;

A domain analysis unit for analyzing a current system configuration when collecting and analyzing necessary target data according to a requirement in a semantic portal in a requirement data collection means,

And a data modeling unit for defining an attribute constraint and creating an instance when defining a class relation by defining a class name and structure of data, and defining an attribute list, and constructing a semantic web portal using the multi-ontology And a semantic search system using the same.
The method according to claim 1,

The component architecture unit,

A control layer that consists of components defined as contents of services and functions that are commonly used by each service, components for implementing the contents of services provided by the semantic portal,

A model layer for providing an associated API for use in the control layer,

And a view layer composed of components necessary for constituting a user interface. The construction of a semantic web portal using a multiple ontology and a semantic search system using the same.

Images