JP5992497B2 - Realization method of situation awareness ontology for providing user interest information service based on context awareness - Google Patents

Description

  The present invention relates to a method for predicting a user's interest based on context awareness technology, and more particularly, to accurately determine what information is required in the user's current situation based on context awareness technology. The present invention relates to a method for realizing a context awareness ontology based on context awareness in order to provide a user interest information service based on context awareness configured to be able to predict and provide relevant information to a user.

  In addition, the present invention constructs context ontology and tour ontology based on context awareness and defines relationships between core classes in order to accurately predict and provide necessary information to users. In addition, the present invention relates to a user interest information service providing method based on context awareness, configured to be able to accurately predict and provide information desired by a user through a semantic data mining algorithm (Semantic Data Mining Algorithm).

  Recently, various information exists on networks such as the Internet due to the arrival of the information age. For this reason, in order to accurately determine what kind of information is useful from among such a large amount of information and select and provide only the necessary information, there is an increasing demand for information filtering technology. It's getting on.

  Furthermore, recently, in addition to existing desktop personal computers, with the development of technologies related to WiFi, 3G, 4G, GPS, sensors, etc., for example, notebook personal computers (Notebooks), tablet PCs (Tablet PCs), PDAs (Personal Digital Assistants). ), PCS (Personal Communications Service), Smartphone (Smart Phone), etc., ubiquitous computing technology and mobile computing technology considering mobility are actively developed.

  In addition, with such a trend, many changes have been caused in the information search method in such a mobile environment. In other words, the existing information retrieval and information transmission method using a desktop personal computer has been mainly information retrieval methods in a fixed space such as an office or a house. With the development of technology for making information, the information transmission method is changing to a method capable of information retrieval or information transmission considering mobility and portability.

  More specifically, for example, according to [Kim 2008] (see Non-Patent Document 1), a ubiquitous web application environment is that many users and various devices such as desktop personal computers, office devices, and home appliances have web applications. Used to define an interacting environment. In such an environment, the majority of users can be provided with a desired service using a portable device such as a mobile phone.

  Further, according to [Michalis 2012] (see Non-Patent Document 2), an information search method using wired and an information search method using wireless are compared. Although it shows a general trend of decline except for the time in the room and after leaving work, the information retrieval method using wireless shows that it continues to increase except for the time when users sleep. Disclosure.

  As described above, the development of the wireless mobile device and the information search method using the same enables the user to search for desired information anytime and anywhere, but when searching for information desired by the user, There is a problem that it is the same as the method using the existing desktop that the information that is not desired is more often provided than the accurate information desired by the user.

  In addition, the conventional search method does not present only the desired interest information as described above, but in addition to the annoyance that the user has to filter the search results one by one, In order to know the position, there is another inconvenience that the search must be repeated in different categories for the same contents until the desired information is obtained, such as further searching the map.

  Therefore, in order to solve the problems of the search method in the conventional mobile environment as described above, it is possible to provide information search results optimized and personalized for user search requests anytime and anywhere. In addition, based on context awareness technology that recognizes the user's current situation using the current user's position and surrounding situation, and provides appropriate information, the user's interest level is predicted. There is a need to develop a context ontology model that allows more accurate prediction of what search results are desired.

  In addition, in order to provide search results that are more accurately summarized to suit the limited screens of mobile devices, considering the mobility, portability and display limitations of users using mobile devices such as smartphones When a search request desired by a user is input, a query pattern that is similar to or matches the user's query pattern is searched based on a previously stored context ontology database, and the query requested by the user is searched. Leverage contextual data, configured to provide users with more accurate and quick search results using a user query pattern search system that provides matching or similar results after comparison with patterns Semantic Day The development of the mining algorithm (Semantic Data Mining Algorithm) is required.

  Moreover, in order to overcome the shortcomings of the existing search method that shows an excessively large number of search results in response to a user search request and to provide a more accurate and concise result to the user search request Recognize the situation based on user search results and provide customized information, and at the same time using the GPS and map functions built into mobile devices such as smartphones, the current location of the user and the location of the information desired by the user It is preferable to provide a new information retrieval method configured so that information desired by a user can be quickly and accurately searched by providing information on a search result together with a search result. No device or method to satisfy has been presented yet.

Korean Published Patent Publication No. 10-2013-0097476 Korean Registered Patent Publication No. 10-1000096 Korea Registered Patent Publication No. 10-0868331 Korean Registered Patent Publication No. 10-0688087

Kim H. W. , Nam J .; Y. Lee H .; Shin D., et al. M.M. "A Device Coordination Mechanism with Context Ontology for Ubiquitous Web Application Engineering", Journal of Engineering, Technology. 17, 2008 Michelis K. , “An Introduction to Android”, Tutorial: hy439 & hy539, 2012 Ushold M.M. F. Jasper R .; J. et al. , “A Framework for Understanding and Classifying Ontology Applications”, Proceedings of the IJCAI-99, Workshop on Ontologies and Probable SolP. 1-12, 1999 Fischer G. , Ostwald J .; L. "Knowledge Management: Problems, Promises, Realities, and Challenges", IEEE Intelligent Systems, Vol. 16, no. 1, pp. 60-72, 2001 Hwang H.H. S. Lee J .; Y. , “A Study of a Knowledge Inference Algorithm using an Association Mining Method based on Ontologies”, Journal of Korea Society Multimedia. 11, no. 11 pp. 1566-1574, 2008 Mostefoui S. K. , Hirsbrunner B., et al. "Towards a Context-Based Service Composition Framework", Proceedings of the International Conferencing on Web Services, pp. 42-45, ICWS'03, Las Vegas, Nevada, 23-26 June 2003 Yilmaz N. Alptekin G .; , “An Ontology-Based Data Mining Approach for Strategic Marketing Decisions”, The 10th International Conference on Cybertechnics and Information Technology, A13. Kim J. et al. H. Kim H., et al. C. "Information Retrieval System for Mobile Devices", Journal of the Korean Society of Marine Engineering, Vol. 33, no. 4, pp. 569-577, 2009 Park H. S. Park M .; H. , Cho S. B. “Information Recommendation in Mobile Environment using a Multi-Criteria Decision Making”, Journal of KIISE. Computing Practices and Letters, Vol. 14, no. 3, pp. 306-310, 2008 Seo H. S. Lee S. Y. , “User Behavior Pattern Inference Model for Context-Awareness-based Personalized Recommended Services”, Thesis (Master's degree), Kongju National. Graduate School: Computer Science Department, 2012 Jeong J. E. , “Ontology based Preprocessing Scheme for Mining Data Streams from Sensor Networks”, Intelligence Information Research, Vol. 15, no. 3, pp. 67-80, 2009 Ko S. K. , Choi S. M.M. Eom H .; S. Cha J. et al. W. Cho H .; C. Kim L., et al. H, Han Y. S. "A Smart Movie Recommendation System", Human Interface and the Management of Information. Interacting with Information Lecture Notes in Computer Science, Vol. 6771, pp. 558-566, 2011 Sarkaleh M. K. , Mahdavi M .; , Biandalan M .; , “Designing a Tourist Recommendation System Based on Location, Mobile Device and User Features in Museuum”, International Journal of Maningol. 4, no. 2, May 2012 Abraham T.W. "Knowledge Discovery in Spatial-Temporal Databases", School of Computer and Information Science, University of South of Australia, Ph.D. D dissertation, 1999 Gu M. S. , Hwang J. et al. H. Ryu K .; H. “Weighted-FP-Tree Based XML Query Pattern Mining”, 6th International Conference, ADMA 2010, Vol. 6440, pp. 417-428, 2011.11 Chen, L.C. Bhowmick, S .; S. Chia, L.A. T.A. , “Minning Maximum Frequent Changing Substructure from XML Documents”, In Proceedings of the 6th International Conference on Data Warehouse. Chen, L.C. Bhowmick, S .; S. Chia, L.A. T.A. , “FRACTURE-Minning: Mining Frequently and Concurrently Mutual Structures from Historical XML Documents”, Elsevier Science Journal in D & E. 320-347, 2006 Han, J .; Pei, J .; Yin, Y. , “Ming Frequent Patterns Without Candade Generation”, Proceedings of the 2000 ACM SIGMOD international conference on Management of data, 2000 Hunyadi D. , “Performance comparison of apriori and FP-growth algorithms in generating association rules.” Proceedings of the encep ense on europe. 376-381, 2011 http: // www. android. com / http: // www. jetbrains. com / webstorm /

  Therefore, the present invention is for solving the problems of the conventional techniques as described above, and its purpose is to provide the user with unnecessary search results in response to a search request from the user. In order to solve the problem of the prior art search method, which had the disadvantage that only the information had to be further filtered, the user's interest level was predicted based on the context awareness technique, and what kind of search the user was at what location. By constructing a context ontology model for more accurately predicting what is desired and predicting user interest information based on such context ontology model, only information related to the user's interest in response to a user's search request Context that is configured to be able to present and provide more accurate and concise search results To provide situational awareness ontology implemented method for providing a user interest information service based on awareness.

  Another object of the present invention is to use a semantic data mining algorithm using a context ontology in order to overcome the shortcomings of a limited screen and a small memory of a mobile terminal such as a smartphone. It is an object of the present invention to provide a context-aware ontology implementation method for providing a user interest information service based on context awareness, which is configured to be able to provide a search result summarized more accurately.

  Another object of the present invention is to provide a search result for the search request information of the user so that not only the content of the desired information but also the position can be easily searched by only one search. User interest information based on context awareness configured to be able to provide the current location of the user and the location of the search result requested by the user together with the search result using GPS and map functions built into the mobile device It is to provide a situation recognition ontology realization method for providing a service.

  In order to achieve the above object, according to the present invention, in response to a user's search request, information unnecessary for the user is provided as a search result, whereby only the necessary information is filtered and selected by the user. In order to solve the problems of the prior art search system that had the disadvantage of having to do so, by presenting only the interest information of the user as a search result based on the context awareness technology, A situation recognition ontology implementation method for realizing a search service configured to provide a more accurate and concise search result than a search system, collecting information on the region of interest of the user and a database Before the user interest information database construction stage to build Based on the ontology inference stage for setting the relationship and attributes between each data stored in the user interest information database constructed in the user interest information database construction stage, and the relationship and attributes of each data set in the ontology inference stage And a ontology database construction stage for constructing an ontology database. A method for realizing a situation recognition ontology is provided.

  Here, the ontology includes user status information including a user identifier U (User identification), spatial information L (Location), time information T (Time), and service information S (Service) related to the user. It is configured to be expressed by spatio-temporal situation information O = {U, L, T, S} having four dimensions.

  The time information is information divided at a predetermined time interval, the spatial information is generalized position information corresponding to coordinates (xi, yi) on the Euclidean space, and the coordinates are the user. Means the longitude value and the latitude value of the current location.

  In addition, the database construction stage collects information on sightseeing spots as interest information of the user for each region, designates a root class as a top class, and designates each region as a subclass. , Specify the name of the tourist destination for each of the regions as a subclass for each of the regions, and specify the detailed description for the facility and the relevant tourist destination in each tourist region as a subclass for the name of the tourist destination, A process for constructing a tourism database having a hierarchical structure by setting a synonym or equivalence class relationship for initial voice search to the area or the tourist spot is performed. It is characterized by.

  Here, the tourism database includes a plurality of fields including id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, long, and the add_1 field, the add_2 field, the add_3 field, and the add_3 field, The add_4 field and the add_5 field indicate the address of each sightseeing spot, the tour_site field stores the name of the sightseeing spot, the syn field is a field for synonym or initial word processing, and the lat field The long field is configured to mean latitude and longitude necessary for setting a map.

  In addition, the ontology inference stage includes “Place” composed of the name of the area to be visited; “Attraction” composed of the name of the tourist attraction in each area; Buildings, relics, ruins in each tourist area "Resource" composed of seaside and dining room; and "Cultural activity" meaning cultural action, "Scenery view" meaning act of enjoying the scenery, and "Dining out" meaning act of eating out ”, And an attribute of each data is determined on the basis of“ Activity ”which means an action actually performed by the user in each sightseeing spot.

  Here, in the ontology inference stage, the attribute relationship between the “Place” and the “Attraction” is designated as “hasAttraction” and “IsAttractionOf”, and the attribute relationship between the “Attraction” and the “Resource” is “hasResource” and “IsResourceOf” is designated, and the attribute relationship between “Activity” and “Resource” is designated as “hasActivity” and “IsActivityOf”.

  In addition, the ontology inference step is configured to consider a multi-attribute relationship when a plurality of attribute relationships exist in the data constituting each class.

  In addition, the ontology inference step is performed using Protage_4.0.2 version.

  In addition, the ontology database construction stage is configured to use the database construction scheme (OWL) to construct the database schema construction (OWL), which is configured to be configured as a database description scheme (OWD) that is configured to be a resource description framework (RDF) (Resource Description Framework) and an RDF (Resource Description Framework Framework). It is characterized by.

  In addition, according to the present invention, using the situation recognition ontology constructed by the situation recognition ontology implementation method described above, only the user's interest information is presented as a search result, compared with the existing search methods. A method for providing a user interest information service based on context awareness, characterized in that it is configured to provide a more accurate and concise search result to be suitable for a mobile environment including a smartphone and a tablet PC. Provided.

  In addition, according to the present invention, by using the situation recognition ontology constructed by the situation recognition ontology implementation method described above, only the interest information of the user is presented as a search result, so that compared to the existing search system. Provided a user interest information service system based on context awareness, characterized by being configured to provide more accurate and concise search results, and suitable for mobile environments including smartphones and tablet PCs Is done.

  As described above, according to the present invention, a context ontology model for predicting a user's interest level based on context awareness technology and more accurately predicting at what position a user desires is constructed. The user interest information can be predicted based on the context ontology model, and only the information related to the user's interest can be presented in response to the user's search request, thereby providing a more accurate and concise search result. By providing a situation awareness ontology implementation method for providing user interest information services based on context awareness, it is necessary for users to provide a huge amount of search results in response to user search requests. Conventional technology that had the disadvantage of having to further filter only sensitive information It is possible to solve the problems of the search method.

  In addition, according to the present invention, it is possible to provide context-awareness that is configured to be able to provide a search result summarized more accurately using a semantic data mining algorithm using a context ontology. By providing a method for realizing a situation recognition ontology for providing a user interest information service based on the above, it is possible to overcome the shortcomings of a limited screen and a small memory of a mobile terminal such as a smartphone.

  In addition, according to the present invention, the search result for the user search request information is provided, and at the same time, the current location of the user and the location of the search result requested by the user are searched using the GPS and the map function built in the mobile device. By providing a context awareness ontology implementation method for providing a user-aware information service based on context awareness, which is configured to be able to be provided along with the results, the user can obtain the content of the desired information with only one search. You can easily find not only the position but also the position.

It is a figure which compares and shows the characteristic, advantage, and fault of an existing data mining method as a table | surface. It is a flowchart which shows the process which designates a specific sightseeing spot as the range of an area, and builds a context ontology. It is a figure for demonstrating the process inferred for construction of a sightseeing ontology using the situation recognition data of the sightseeing spot in each area in order to provide a user with the service with respect to a sightseeing spot. FIG. 4 is a diagram schematically showing a part of the hierarchical structure of the tourism ontology shown in FIG. 3. It is a figure for demonstrating the process of setting the attribute value and relationship of the data value which comprises each class for reasoning of ontology. It is a figure which shows the multiple attribute relationship between the data which comprise each class. It is a figure which shows roughly the process of inferring ontology using a protage. It is a figure which shows a part of sightseeing database constructed | assembled by the Example of this invention. It is a figure which shows a part of sightseeing database table constructed | assembled by the Example of this invention. It is a figure which shows the structure of the tourism ontology based on the Example of this invention. It is a figure which shows the ontology structure which extended add_2 class of the sightseeing ontology shown in FIG. It is a figure which shows the ontology structure which extended add_3 class of the sightseeing ontology shown in FIG. It is a figure which shows the ontology structure which expanded add_4 class of the sightseeing ontology shown in FIG. In accordance with an embodiment of the present invention, tour_4.0 constructed using OWL (Web Ontology Language), RDF and RDFS using protage_4.0.2. It is a figure which shows a part of ow ontology source. 1 is a diagram schematically illustrating an overall configuration of a method for realizing a situation recognition ontology for providing a user interest information service based on context awareness according to an embodiment of the present invention; FIG. It is a figure which shows the example of the XML query pattern tree (Query pattern tree) and the XML query subtree (Query Rooted subtree) which has a route, respectively. It is a figure which shows the structure of the frequent subtree which has a query pattern tree and a root, respectively. It is a figure which shows roughly the process of sharing the prefix (prefix) of an FP-Tree construction process, and the process of showing each transaction. It is a figure which shows the process of constructing conditional FP-Tree of m which is arbitrary items in the process of performing data mining by performing FP-growth algorithm. It is a figure which shows the conditional pattern obtained as a result of repeating the process of searching for a conditional pattern as a table | surface collectively. It is a figure which shows the example which produced the query made in the Korean language which the user performed in the form of XML. It is a figure which shows Book DTD Tree which is an XML document created on the basis of XQuery. It is a figure which shows the frequent item set (transaction frequency itemset) of each transaction in a sightseeing ontology database. It is a figure which shows roughly a part of FP-tree structure using the sightseeing ontology database which has the frequent appearance structure which generate | occur | produces for each area. It is a figure which respectively shows the case where a conditional FP-tree is constructed | assembled on the basis of global (Global) FP-tree and the "Mala road" in a sightseeing area. It is a figure which shows a part of FP-growth algorithm source. It is a figure which shows the result of having performed the data mining by applying minimum support degree to 0.4 to FP-growth algorithm. 1 is a diagram schematically illustrating an overall configuration of a mobile-based information search system that utilizes a pattern matching technique using a semantic data mining algorithm. FIG. It is a figure which shows roughly a series of processing steps performed when the search request with respect to a sightseeing spot enters from a user. When a search request is received from a user, the user's requirements are analyzed, the user's query pattern is analyzed, and the pattern that matches the analyzed query pattern is compared with the pattern structure stored in the ontology database of the search system. It is a figure explaining the process of searching for the similar or the same pattern structure. It is a figure which shows schematically the whole structure of the user interest information search method using the semantic data mining which concerns on the Example of this invention. It is a figure which shows a part of source | sauce for implement | achieving the initial screen of a search window. It is a figure which shows the source which implement | achieved the event action with respect to a search button with a script function. It is a figure which shows a part of source which performs the map button action for providing a map service. It is a figure which shows the structural example of the main screen of the mobile information search system which concerns on the Example of this invention. When a regional value of “Taejeon Metropolitan City” is entered in the search window and the map button is clicked, the map of the relevant region is displayed, and the cafeteria and sightseeing in the relevant region according to the input value entered in the search window It is a figure which shows each the structure by which the screen which introduces an area is displayed. It is a figure which shows the map implementation | achievement source which implement | achieves the map which recognizes the user's present position value automatically as a latitude value and a longitude value, and shows the map of the area where the user is currently located with a latitude value and a longitude value. It is a figure which shows the result screen implement | achieved via the source as shown in FIG. It is a figure which collectively shows the experimental result which applied the FP-growth algorithm to the ontology database which concerns on the Example of this invention, and performed the data mining process as a table | surface. It is a figure which shows collectively the experimental result which applied the FP-growth algorithm to the existing relational database, and performed the data mining process as a table | surface. 6 is a graph showing a process of changing memory usage according to a change in minimum support degree critical value of ontology data and an existing relational database according to an embodiment of the present invention. It is a graph which shows the change of the number of frequent patterns by the change of the minimum support level value of ontology data and the existing relational database which concern on the Example of this invention. It is a graph which shows the change of the execution time by the change of the minimum support degree of ontology data and the existing relational database which concern on the Example of this invention. It is a graph which shows the change of execution time by the change of the transaction of ontology data and the existing relational database which concerns on the Example of this invention. It is a graph which compares and shows the change of the memory usage by the change of the transaction of ontology data based on the Example of this invention, and the existing relational database. It is a graph which shows the change of the precision which shows search efficiency with the change of the transaction which is a user query pattern. It is a figure which shows the screen which displays a search result when "Keron mountain" is input into the search window in the existing portal site. It is a figure which shows the screen which displays a search result when "Mount Kelon" is input into the search window with the mobile information search system which concerns on the Example of this invention. It is a figure which respectively shows the screen which displays the detailed description relevant to a search result, and the map of a surrounding area, when a search word is input with the mobile information search system which concerns on the Example of this invention.

  Hereinafter, a specific embodiment of a situation recognition ontology implementation method for providing a user interest information service based on context awareness according to the present invention will be described with reference to the accompanying drawings.

  Here, it should be noted that the content described below is only one example for carrying out the present invention, and the present invention is not limited to the content of the example described below. .

  Further, in the following description of the embodiments of the present invention, the description of the portions that are the same as or similar to the contents of the prior art, or that can be easily understood and implemented by those skilled in the art will be briefly described. It should be noted that the detailed description thereof has been omitted in order to achieve the above.

  That is, as described later, the present invention has a drawback in that it has a disadvantage that it provides a huge amount of search results in response to a search request from the user and further filters only the information that the user needs. In order to solve the problem of the system, a context ontology model for predicting a user's interest level based on a context awareness technique and more accurately predicting what kind of search the user desires at which position By predicting user interest information based on such context ontology model, only information related to user interest is presented in response to user search requests, and more accurate and concise search results are provided. Based on context awareness, configured to be able to It relates to the situation recognition ontology realization method for providing over The interest information service.

  In addition, the present invention summarizes more accurately by using a semantic data mining algorithm (Context Data Ontology) to overcome the shortcomings of limited screens and low memory of mobile terminals such as smartphones. The present invention relates to a method for realizing a situation awareness ontology for providing a user interest information service based on context awareness, which is configured to be able to provide a searched search result.

  In addition, the present invention provides a search result for a user's search request information and at the same time for a mobile device so that not only the content of desired information but also a position can be easily searched by a single search. To provide a user interest information service based on context awareness configured to be able to provide a user's current position and a position of a search result requested by the user together with the search result using a built-in GPS and map function The present invention relates to a situation recognition ontology realization method.

  Here, before describing a specific embodiment of a method for realizing a situation awareness ontology for providing a user interest information service based on context awareness according to the present invention, the ontology and context awareness applied to the present invention are described. (Context awareness) and the basic concept of a data mining algorithm will be described.

  First, ontology means a method of conceptualizing information within a specific range and classifying or defining concepts with a logical structure and expressing them in a hierarchical structure format. Conventionally, [Ushold 1999]. According to [Fischer 2001] and [Hwang 2008], Ontology is a data representation method for analyzing and integrating data scattered by the web and providers, etc. RDFS (Resource Description Framework Schema Specification), RDF (Resource Description), which is an element for expressing an ontology in order to construct an ontology, generally defined as a stylized explicit proposition ramework), OWL (technical content for Web Ontology Language) has been proposed (see Non-Patent Documents 3 to 5).

  That is, RDFS is the most basic language for indicating an ontology, and defines a basic vocabulary for using an RDF application language. RDF is a data model for all types of information and metadata. It is a useful language that can display, and consists of source and sentences.

  More specifically, an RDF sentence is composed of a subject, a predicate (predictate (property)), and an object (object (value)). OWL (Web Ontology Language) is based on an initial ontology language such as RDFS. A standard ontology language proposed by the W3C to define an ontology in the semantic web to be created, which can represent object classes and upper and lower hierarchical relationships between classes for the purpose of representing the ontology's hierarchical structure. Since any attribute can be defined and the defined attribute can be assigned to a class, it is more standardized than RDFS, and can be logically inferred from the relationship between objects. Similarly, information is expressed in RDF text. .

  Here, also in the embodiment of the present invention described below, RDFS, RDF, OWL, etc. as described above were used to construct a tourism ontology database.

  Next, the concept of context awareness (context awareness) will be described. The context information may be personal, such as the user's current activity, and may be technical, such as the device currently in use. It may be environmental, such as temperature, location or time.

  In other words, personalization of context-aware service recognizes through the collection and exchange of situation information, and provides the user with a customized service suitable for the situation (context) through processing processes such as analysis and inference. For this, various surrounding situation information must be appropriately modeled, collected and distributed.

  Therefore, the situation information collected to provide context-aware services must be analyzed and inferred into high-dimensional information that is combined with user information and expressed in one unit of life. The method of storing and systematically and efficiently sharing is an important technique for determining the success or failure of the context awareness system.

  Moreover, recently, as the number of mobile users has increased, there has been an increasing demand for applications and services that can adapt to a rapidly changing environment. However, a context awareness system suitable for such demands is still being constructed. In order to overcome the problem that it was difficult to do, an ontology-based context model using OWL has been presented as a method for expressing semantic context, inferring situations, sharing knowledge, and classifying situations .

  More specifically, so-called SOCAM (Service-Oriented Context-Aware Middleware) has been presented for context-aware services that allow various situation inferences, that is, according to, for example, [Mostefoui 2003] A CB-Sec (Context-Based Service Composition) framework (framework) in which context-aware computing and agent technology are combined is presented (see Non-Patent Document 6).

  This included the history and overcame the drawback of certifying only the location as the context parameter in most context-aware applications and ignoring the temporal aspect. . In addition, existing context recognition includes user context (user context) including the user's role, identity, age, location, and preferences, and computing context (Computing context) including printers, displays, workstations, etc. in the network connection and surroundings. ), Time status including time such as day, week, month, season, etc. (Time context), physical status including weather, temperature, etc. (Physical context), and user status, computing status, and physical status for a certain period of time Five elements of context history were included that can be recorded and usefully applied to service configuration.

  Subsequently, the data mining technique will be described. Although the existing text mining technique can be applied to apply the semantic data mining technique, the tourism ontology database applied to the embodiment of the present invention described below is applied. In order to perform data mining using the ontology, a data mining technique for an XML document having the basic structure of the ontology will be described here.

  That is, referring to FIG. 1, FIG. 1 is a table comparing the features, advantages and disadvantages of existing data mining methods.

  As shown in FIG. 1, various data mining techniques for XML data and various techniques for XML query pattern mining have been studied. Each method has advantages and disadvantages.

  Also, according to [Hwang 2008], the current search system may provide too many search results due to a large amount of data, or may not be able to provide search results. In order to compensate for the shortcomings, we propose an algorithm that supports efficient search by expressing related knowledge inferred by OWL ontology constraints and association mining technique in SWRL inference language, and discovering new knowledge via Jess engine. did. That is, when a personal preference ontology by domain is constructed and information is retrieved on the basis of association and ontology using personal preference data as an example, a process of providing association and inference information has been shown (see Non-Patent Document 5).

  Moreover, according to [Yilmaz 2013], an ontology-based data mining technique using the Priori algorithm is presented, that is, a framework based on the data mining technique is used to determine the operating system preferred by consumers. Proposing, constructing a domain ontology, investigating questionnaires to users and measuring their preferences, then transmitting data, analyzing clustering, applying association rule algorithms and analyzing the results, At this time, a protage was used for ontology construction. The Apriori algorithm is performed in two stages in which a frequent item set is searched, the item set is classified into two types, an precedent and a result, and a rule satisfying the confidence level is searched (non-patented). Reference 7).

  Next, the information retrieval system in the mobile environment will be described. Recently, with the development of ubiquitous technology, as a study on context awareness technology that provides useful services to users, it recognizes the user situation and provides useful information. Context-aware recommendation services have been studied, but the existing context-aware recommendation service recommends only consistent conceptual level information as recommended information based on the situation, so that the recommended information matches the user's current situation. If not, it takes considerable effort and time because the user has to search for the desired information directly from the recommended information.

  Moreover, recently, with the development of the wireless Internet environment, Internet search using terminals has become popular, and in order to efficiently input such search terms, functional aspects of hardware such as a keyboard are considered. And methods that take into account the functional aspects of software using user favorites and saved search terms were presented.

  Here, the ontology model can be used for storing and managing various status information of the mobile device such as GPS, illuminance sensor, user information, etc. The mobile device is a hardware sensor included in the device. Depending on the type, it is possible to acquire various types of situation information for the user's position, environment, etc., assign this to the ontology model, grasp the meaning between the situation information and process it into information of the desired form or situation Can be used for inference.

  In addition, the mobile environment has different characteristics from the general wired environment. That is, it has the following characteristics. First, since the mobile environment has no space-time restrictions, there is a characteristic that the user tries to find desired information anytime and anywhere while carrying the mobile device. Second, the mobile environment is very personal, most mobile devices are owned by individuals, and the desired information differs from one another depending on their personal interests and areas of interest, so depending on the personal characteristics Although it looks the same in appearance, it has a characteristic that the query content may be substantially different. Third, mobile devices are small and have many restrictions on input and output, so that only the contents desired by the user are extracted rather than showing many pages through a small screen in a mobile environment. It is even more effective to show it. Fourth, in general, the mobile environment has a much lower data transmission rate and higher price than the wired environment. Fifth, in general, mobile devices have memory and processing capacity limitations, and the mobile device itself cannot process a large amount of information.

  Therefore, it is necessary to provide an information search system suitable for the mobile environment in consideration of such characteristics of the mobile environment.

  That is, according to [Kim 2009], a mobile information search model in which a document summary and a user profile are applied to a meta search model in consideration of such characteristics of the mobile environment is presented. The problem of small screen and low transmission rate can be alleviated, and the user profile can use an information retrieval model personalized in a mobile environment (see Non-Patent Document 8).

  In addition, since the degree of preference for the information recommendation service can vary depending on the situation, in order to provide the information recommendation service, first, the user's context information must be known.

  That is, according to [Park 2008], in order to model a preference system of individual users in a mobile environment after proposing a recommendation system that takes into account the preference of a large number of users in a mobile environment and applying it to restaurant recommendations. A Bayesian network was used. In many cases, the recommendation of restaurants uses AHP, which is a multi-criteria decision making method, in order to acquire a large number of preference levels based on the preference levels of a large number of users rather than individual users (see Non-Patent Document 9).

  This is because as most users have mobile devices recently, the mobile devices act as a passage through which personal information can be contacted, and the user always possesses it, so it collects user context information It can also play a role as a possible tool. Since information collected in this way can be usefully used for information recommendation for users, inference of mobile context is also an important research field related to information recommendation.

  Moreover, in order to provide a personalized recommendation service in a context-aware environment, it is necessary to be able to analyze the collected context information quickly and infer the user's purpose effectively. Since information differs depending on the context environment, it is not suitable to apply the existing inference algorithm as it is, and an efficient algorithm suitable for the mobile environment is required.

  That is, according to [Seo 2012], behavior patterns are classified using a naive Bayes classifier in order to minimize loss due to omission or error of information, and the user's tendency is effectively learned to infer behavioral purposes. In order to achieve this, a pattern matching technique was used (see Non-Patent Document 10). Also, according to [Jeong 2009], in order to collect environmental data in a specific space and discover more useful information and knowledge by developing various sensors and building a sensor network, data mining (Data mining) Research that uses the technique was introduced (see Non-Patent Document 11).

  Moreover, according to [Ko 2011], genre correlation is used so that preference information can be collected from many users using a collaborative filtering method and automatic prediction of user's interests can be performed. Has proposed a recommendation system based on (see Non-Patent Document 12). [Sarkaleh 2012] has characteristics such as mobility, independence, personality, and accessibility of mobile devices such as laptop computers, mobile phones, personal communication systems (PCS), and personal digital assistants (PDAs). Using these devices, you can personalize learning at the exact time and place, enrich and diversify learning content, and use GPS (Geographic Position System) function to provide information on various fields desired by the user Has been proposed (see Non-Patent Document 13).

  As described above, various methods related to an interest information recommendation system for providing only necessary and interested information to a user have been presented in the past. Has the following problems.

  More specifically, it can be said that mobility is the biggest difference when comparing an existing data search method with a data search method that is currently being actively applied in the mobile field. In other words, most of the existing data search methods use a desktop computer, so all searches and data applications have been performed in a single place, but considering the mobility and activity of the current society. At times, it is quite inefficient, and it can be said that it does not catch up with current state-of-the-art technology.

  Therefore, in order to overcome the disadvantages of the existing information retrieval methods as described above, there is an increasing demand for an information retrieval method suitable for the mobile environment, and the existing information retrieval method does not consider mobility. It was impossible to build a system that takes into account the current position of the user. However, recently, with the development of mobile technology, it is possible to apply a context awareness technique for finding a user's current location using a GPS function to an information search system. Therefore, the present invention uses this to provide a service for transmitting accurate information using the current position where the user makes a query.

  That is, for example, in order to be able to provide users with more accurate information transmission services through context awareness, a tourism ontology is constructed by collecting tourism data information in each region using a situation awareness ontology. It is possible to provide information suitable for the current situation of the user.

  Also, when a search result is obtained by inputting a search word in an existing information search system, more items that are not relevant than the accurate information desired by the user are searched, and a huge amount of search results are provided. In order to solve such a problem, in the present invention, a situation awareness data ontology is constructed, and a semantic data mining technique is provided to provide a more accurate and quick service to a user query. Adopted.

  That is, for example, by building a tourism ontology and storing a query pattern that a user often queries (questions), when a user query is performed, using a matching technique between the stored query pattern and the user query, By providing the user with a query pattern that matches or resembles the user's query (question), it is possible to provide only the information that the user actually desires.

  In addition, since the existing information retrieval system can only provide text information, the present invention provides a map service such as a Google map supported by mobile devices such as smartphones. It is intended to provide a map service that can more easily and quickly find the specific geographical location of the information requested by the user with a single search.

  Next, in order to solve the problems of the prior art as described above, a specific process for constructing an ontology using situation recognition information will be described in detail.

  First, the basic concept of context awareness ontology will be described. A spatio-temporal movement pattern is a position pattern of a moving object, which is personalized according to the position characteristics of a customer, Spatio-temporal rules that enable the provision of services, and such spatio-temporal relationships are very important for exploring causal relationships between events related to spatio-temporal objects, It is possible to provide appropriate services for user behavior through an ontology that considers both time and space.

  Therefore, in the present invention, an ontology that is useful for structuring situation information and capable of expressing information on interrelationship and partial situation is designed. The ontology according to the embodiment of the present invention is information related to user status information such as time information, space information, and service, and is expressed by O = {U, L, T, S}.

  Here, U (User identification) represents a user identifier, L (Location) represents spatial information, T (Time) represents time information, and S (Service) represents a service rule associated with the user.

  That is, the user U's spatio-temporal situation information includes user identification information, time information, spatial information, and service information, and has four dimensions.

  In more detail, the spatio-temporal information that identifies the user's space and time indicates the user's situation information, and serves as a reference that enables the service history according to the situation (context) to be classified. Consists of generalized values for coordinates in space for the valid time that the event occurred.

  At this time, the time information is information divided at regular time intervals, and the spatial information is generalized position information corresponding to coordinates (xi, yi) in the Euclidean space.

  In addition, the position generalization is generalized to a certain zone with respect to the coordinate value of the space to identify the positions of the user and the object. Here, the coordinates of the space mean the longitude value and the latitude value of the area where the current position is located.

  For example, if the user is currently located in the town of “Jung-gu, Jung-gu, South Korea,” the spatial coordinates of the user are generalized to the area position value corresponding to the position of Jung-gu, Jung-gu, The latitude value and longitude value of the current position of the user can be used.

  Next, the ontology inference process and the tourism ontology construction process necessary for constructing the ontology database will be described.

  First, referring to FIG. 2, FIG. 2 is a flowchart showing a process of constructing a context ontology by designating a specific tourist spot as a region range.

  As shown in FIG. 2, in the process of constructing the context ontology, first, the ontology domain to be constructed is determined in order to construct the ontology (step S21).

  That is, for example, when providing information on a sightseeing spot, the user searches for sightseeing spot information desired by the user and collects information on the relevant sightseeing spot for each region.

  Thereafter, a class for each main concept is formed (step S22). At this time, each class is configured in a hierarchical structure including a superclass and a subclass.

  That is, for example, a route class (root class) called tour_sites is designated as the highest class, each region is designated as a subclass (subclass) below it, and a lower class is generated. Establish a tourism ontology that has a hierarchical structure by setting relationships.

  Subsequently, by using the attributes of the data values constituting each class, it is defined what relationship and what characteristics each data has (step S23), that is, for example, the characteristics of the regions constituting each class, Specify the attribute relationship of each class using the contents explaining the famous cafeteria or sightseeing spot included in the area.

  Next, with respect to the name of the tourist spot corresponding to each class, an attribute value is specified using the characteristics of the tourist spot to generate an instance of the tourist spot (step S24).

  Next, a consistency check is performed to check whether the newly constructed tourism ontology matches the axioms and attributes of the existing ontology (step S25), and the thus constructed ontology searches the user's information. Used for.

  Therefore, when the user tries to search for a desired tourist spot at the current position of the user by performing the above-described process, a tourist spot is selected for each area in order to recommend an appropriate tourist spot at the position requested by the user. You can classify and build a tourism ontology.

  Next, the process of inferring the attributes and relationships of each class of the tourism database in order to construct the tourism ontology as described above will be described.

  In general, a user desires to receive an optimal service at any position and any environment regardless of his / her own position and surrounding environment. In this case, a service is a unit that abstractly describes information about functions such as devices that can be provided, locations, and execution conditions, and is structured hierarchically.

  Further, if there is no service that meets the user's request in the existing service rule set, it is necessary to create a new service or provide a combination of existing services. In order to provide such services to users, it is possible to provide services related to each place in the ontology, sightseeing spots in each place, and restaurants, shopping malls, beaches, and walking paths in each sightseeing spot. For each content, the attributes of each class in the ontology must be specified through inference.

  The reasoning process of such a tourism ontology will be explained. First, referring to FIG. 3, FIG. 3 illustrates a process for inferring for the construction of a tourism ontology using situation recognition data of a tourist area in each region in order to provide a user with a service for the tourist area. FIG.

  In FIG. 3, “Place” means the name of the area where sightseeing is to be performed, for example, Jeju Special Self-Governing Province (Seju), Seoul Special City (Seoul), Daejeon Metropolitan City (Daejon), Busan Metropolitan City. (Busan), Chungcheongnam-do (Chungnam), Chungcheongbuk-do, etc.

  “Attraction” means a name of a sightseeing spot in each area, for example, National Museum, Kyeongbokgung, Haeundae, Mount Kelong, Shopping mall, etc. Consisting of the names of tourist attractions.

  “Resource” means a building, a relic, a ruin, a seaside, etc. in each sightseeing spot, that is, for example, an old building such as an Ancient Remains or a folk village in Kyeongbokgung, It consists of Haeundae or Jeju Ushijima seaside and a restaurant in a department store.

  In addition, “Activity” means an action that the user actually performs at each sightseeing spot, for example, “Cultural activity” means a cultural act of visiting a court in Gyeongbokgung, and “Scenery view”. Means an act of enjoying beautiful scenery at tourist sites such as Haeundae, Jeju Island, and Mount Hanra, and “Dining out” means an act of eating out at a restaurant in a department store.

  In the embodiment of the present invention, the protage_4.0.2 is used for ontology construction and ontology inference, and tour ontology, tour.ontology using OWL (Web Ontology Language), RDF, RDFS, or the like. owl was constructed.

  Next, referring to FIG. 4, FIG. 4 is a diagram schematically showing a part of the hierarchical structure of the tourism ontology shown in FIG.

  As shown in FIG. 4, the ontology according to an embodiment of the present invention specifies a subclass for each region for the construction of a tourism ontology, specifies a class of a smaller group belonging to the region, Below that, the name of the sightseeing spot was designated as a subclass, and another subclass was designated for explanation of the cafeteria in each sightseeing spot or the sightseeing spot.

  Moreover, in order to perform a synonym or initial voice search for each region or sightseeing spot, a synonym class relationship that means synonym processing was set.

  In other words, when a user inputs a search term for a sightseeing spot, a synonym class relationship is set and the user is always accurate in case a search request is made with a similar word or synonym or a search is made in the form of an abbreviation. The information retrieval result is configured to be provided.

  More specifically, as shown in FIG. 4, “tour” indicating a tourist spot is designated as the top class, and Daejeon, Chungnam, Seoul Special City (under the tourist spot) Each region, such as Seoul), Busan Metropolitan City (Busan), Jeju Special Self-Governing Province (Jeju), etc. was designated as a subclass, and under each region, it was composed of subclasses constituting the region.

  In other words, the sub-classes of “East Ward”, “West Ward”, “Naka Ward”, “Yusong Ward”, “Teduk Ward” are included in the case of Daejeon Metropolitan City (Daejon). ”,“ Seogwipo City ”, etc., and Chungnam, for example,“ Conju City ”,“ Buyo County ”, etc., and the subclasses below specified detailed address values. . For example, in Daejeon, “Eunhen Cave” and “Sajon Cave” are sub-classified under “Naka Ward”, and “Jiang An Road” and “Mannyeong Cave” are located under West Ward. In the case of Jeju Special Self-Governing Province (Jeju), sub-classes such as “Sam Doi Cave”, “Guza Pass”, “Chunmun Cave” etc. under Jeju City, Sub-classes such as “Songsan Song” and “Anduk Face” were constructed.

  In addition, the subclass below it is composed of the names of tourist spots. In other words, in the case of Daejeon Metropolitan City, a subclass is constructed from Taejon Metropolitan Jung-gu Sajon-dong Oworld, etc., and in the case of Jeju Special Self-Governing Province, a subclass relationship is constructed from Jeju Manjungle, Jeju Special Self-Governing Province. did.

  Next, referring to FIG. 5, FIG. 5 is a diagram for explaining a process of setting attribute values and relationships of data values constituting each class for inference of ontology.

  As shown in FIG. 5, the relationship between each area (Place) and the sightseeing spot (Attraction) constituting the sightseeing ontology has an attribute relation of “hastraction” and “IsAttractionOf”, and the court in the sightseeing spot and the sightseeing spot. “Resources” such as buildings, relics, seaside, mountains, etc. have the attribute relationship of “hasResource” and “IsResourceOf”, and the user sees the court “Kyeongbokgung” in tourist spots, The act of strolling along the beach, the climbing of “Kerongsan”, and the act of shopping at a department store are “Activity”, and the attribute relationship between “hasActivity” and “IsActivityOf” is related to “Resource”. have.

  Here, each attribute relationship may have a relationship with multiple attributes between data constituting each class.

  That is, referring to FIG. 6, FIG. 6 is a diagram showing a multiple attribute relationship between data composing each class.

  As shown in FIG. 6, for example, an area called “Seoul” and an area called “Jongro” that has an inclusive relationship in Seoul have an attribute relationship of “hassubclass” and “IsSuperclassOf”. In addition, the area called “Jongno” includes two areas called “Kyongbokgung” and “Insadong”, so they have a subclass relationship at the same time.

  In addition, the area called “Jongno” and the tourist destination called “Kyung-bok-gun” have a relationship of “hastraction” and “isAttractionOf”, while the area called “Jongno” and the tourist spot called “Insa-dong” also have “hastraction”. "And" IsAttractionOf ".

  Therefore, such a relationship is called a multi-attribute relationship, and when there are a large number of tourist spots in the same area in each tourist spot, it is considered that such a multi-attribute relation exists.

  Here, “Kyeongbokgung” and “Insa-dong” belong to the same place, so they have an attribute relationship of “atSamePlace”. “Kyeongbokgung” Since there are palaces and gardens that can be seen, they have multiple Resource relationships. Therefore, it can be seen that “Kyonbok-kun” has the attribute relationship of “hasResource” and “IsResourceOf” twice with respect to “Palace” and “Garden”.

  That is, referring to FIG. 7, FIG. 7 is a diagram schematically illustrating a process of inferring an ontology using a protage, where FIG. 7 a illustrates a subclass relationship of each region, and FIG. The relationship of the subclass which comprises each sightseeing spot of Daejeon area is shown.

  In addition, recent users are searching for various types of useful information using various smartphone applications in a mobile environment. The GPS function is one of the conveniences of using such various applications. Can be mentioned.

  For this reason, the present invention provides a map search using a GPS module by using the advantage that GPS can be used, which is one of the characteristics of the mobile environment, when a search request is made in the mobile environment. Various search results can be provided.

  More specifically, since the current position of the user can be provided via the GPS module, the current position value of the user is measured as longitude and latitude using this, and the measured current position of the user is used. It is possible to search for sightseeing spot information at the point where the user is currently located, and provide a map for the current location and the searched sightseeing spot using a map service such as Google map, for example. It can be configured as follows.

  Next, in the aforementioned ontology inference process that links the attributes and relationships of each tourist destination and the relevant tourist destination to set the relationship between each class, information using semantic data mining techniques in a mobile environment using situation recognition data The process of building a tourism ontology necessary for search will be explained in detail.

  That is, in the embodiment of the present invention described below, a tourism database is constructed using information related to tourist spots collected through sites related to tourist spots such as portal sites for the construction of a tourism ontology. did. At this time, the database constituted a test_db database and a tour_db table with the address values of each sightseeing spot using MS SQL Server 2000 personal version.

  More specifically, referring to FIG. 8 and FIG. 9, FIG. 8 and FIG. 9 are diagrams respectively showing a part of the tourism database and the tourism database table constructed as described above.

  As shown in FIGS. 8 and 9, for example, tour_db is composed of fields such as id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, and long.

  Here, id is designated as a primary key. For example, in the case of Daejeon Metropolitan City, add_1 is composed of Daejeon Metropolitan City, add_2 is composed of Naka Ward, West Ward, East Ward, Yusong Ward, etc., and add_3 is Sajon-dong , Mannyeong-dong, Eun-Heng-dong, Jiang-an Road, etc., add_4 designates Puri Park Road in Jeongsan-dong, Jung-gu, Daejeon, and add_5 designates 327, 1st Road, Dongpo Temple, Bangpo-myeon.

  Also, tour_site specifies, for example, the name of a tourist spot such as “Kundokjeong” located at 19 Gwangdeok Road, Samdei-dong, Jeju City, Jeju Special Self-Governing Province, and syn is used when a user searches for a tourist spot. Is a field designated for synonym and initial word processing.

  Moreover, lat and long specified the latitude value and longitude value of each area. Here, the latitude and longitude are specified to set the position value of the user for showing the map using, for example, a Google map.

  Therefore, a tourist spot ontology can be constructed using the tourist spot database constructed as described above, and in this embodiment, protage — 4.0.2 is used for such ontology construction and ontology inference. Along with tour.tour, a tourism ontology using OWL, RDF, RDFS, etc. owl was constructed.

  More specifically, in order to construct a sightseeing spot ontology, the concept of tour_sites is set as the highest class, and add_1 field in the tour_db table is specified as the lower class. For example, Daejeon Metropolitan City, Jeju Special Self-Governing Province, Chungnam, Busan Metropolitan City, etc. are designated as add_1 class.

  That is, referring to FIG. 10, FIG. 10 is a diagram illustrating a configuration of a tourism ontology according to an embodiment of the present invention.

  As shown in FIG. 10, the upper class of the tour_sites ontology includes classes such as Daejeon Metropolitan City, Busan Metropolitan City, Jeju Special Self-Governing Province, Chungnam, etc., and the lower part thereof is the add_2 class in tour_db. Was designated as a subclass of the upper class indicating each region. For example, Naka-gu, East-ku, Nishi-ku, Te-dok-gu, etc. are designated under the Daejeon Metropolitan City.

  That is, referring to FIG. 11, FIG. 11 is a diagram illustrating an ontology structure in which the above-described add_2class is expanded.

  Next, add_3 is specified in the tour_db table as a subclass. For such add_3 class, for example, Sajon-dong, Chimsan-dong, Eun-Heng-dong are specified in the middle district of Daejeon City. Appointed Jeju City under the self-governing province, and under Jeju City, Ideil Cave, Samdoi Cave, 1100 Road, Jeochon Pass, Guza Pass, Aewol Pass, Hean Cave, Han Rim Pass, Yongdam Road, Yongdam Ile Cave, Udo , Hankyeong, Smoke Wongil, Yongdam Sam-dong, Samyang-samu-dong, Iru-doi-dong, etc. were designated.

  That is, referring to FIG. 12, FIG. 12 is a diagram showing an ontology structure in which the add_3 class of the tourism ontology shown in FIG. 10 is expanded.

  More specifically, in FIG. 12, Namwon Pass and its subclass Iruju East Road, Daejeon Pass and its subclass Mara Road, Irju West Road 3000 Gil, belonging to the city of Seogwipo in the Jeju Special Self-Governing Province It shows the structure of Hyobze Coast Road and Donggyeong-dong and sub-class No. 214 Gil, Seventy-six Road.

  Subsequently, as its subclass, for example, Sajon Park Road is designated as add_4 class of tour_db table under the Daejeon Metropolitan City, Jung-gu, Sajon-dong, and the sub-classes thereof, ie, Daejeon Metropolitan City, Jung-gu, Sajon. Under the cave, Sajon Park Road, Oworld was designated as a tour_site class.

」という）」というｔｏｕｒ＿ｓｉｔｅクラスの場合は、同義語または初声の処理のために、「

」という同義語関係を指定するために同義語クラス（ｅｑｕｉｖａｌｅｎｔ ｃｌａｓｓ）を設定した。 In addition, "O world (in Korean"

In the case of a tour_site class called ")", for synonym or initial voice processing,

The synonym class (equivalent class) was set to specify the synonym relationship “

  Referring to FIG. 13, FIG. 13 is a diagram illustrating an ontology structure in which the add_4 class of the tourism ontology shown in FIG. 10 is expanded. The relationship between add_4 and tour_site such as “Miyananchi” is shown.

  Referring to FIG. 14, FIG. 14 shows tour. Constructed using OWL (Web Ontology Language), RDF, and RDFS using protage — 4.0.2. It is a figure which shows a part of ow ontology source.

  Therefore, as described above, it is possible to infer and construct a tourism ontology using situation recognition data using tourism data as situation data.

  That is, referring to FIG. 15, FIG. 15 is a diagram schematically illustrating an overall configuration of a situation recognition ontology implementation method for providing a user interest information service based on context awareness according to an embodiment of the present invention.

  More specifically, in the situation recognition ontology implementation method according to the embodiment of the present invention, first, information on a user's region of interest is collected to construct a user interest information database (step S151), and the constructed user interest information database is stored in the constructed user interest information database. After performing ontology inference that sets relationships and attributes between stored data (step S152), a situation recognition ontology database is constructed based on the relationships and attributes of each data set in the ontology inference step (step S153). ) It can be configured including a series of steps.

  Here, the specific process of constructing the ontology database by constructing the user interest information database described above and performing ontology inference is similar to the process of constructing the tourism ontology described above with reference to FIGS. It should be noted that the specific description of each step is omitted here for the sake of brevity.

  Next, a semantic data mining technique based on the FP-growth algorithm that finds a frequent pattern similar to or identical to a user query using the situation database constructed as described above will be described.

  In other words, semantic data mining refers to the process of finding hidden putters that can be logically structured for a variety of data and is useful for user queries, to provide quick and accurate information, In an embodiment of the present invention described below, data mining using a similar query pattern is used to efficiently transmit useful information suitable for a user by using an existing FP-Tree-based FP-growth algorithm. A series of processes for performing semantic data mining in order to provide a service suitable for a user's query using the process and situation recognition data (Context Awareness Data) will be described.

  Here, for example, as presented in [Gu 2010], in order to perform the data mining process, the following definitions are necessary (see Non-Patent Document 15).

[Definition 1] (XML Query Pattern Tree)
The XML query pattern tree is a rooted sub-tree, and XQPT = <X. V, X.V. E>.
Here, X. V is a vertex set (vertex set), X. E means each edge set (edge set), root is represented by Root (XQPT), and each edge is represented by a vertex (v1, v2), but v1 is a parent of v2 , Each vertex v has a label having a value of {*, //, tagSet}, and tagSet means a name set of all elements and attributes in the DTD.

[Definition 2] (XML Query Rooted Subtree)
Query pattern tree XQPT = <X. V, X.V. E> is given, an XML query subtree XQRST = <X.X with a root that is a subtree of XQPT. V ′, X. E ′> is Root (XQRST) = Root (XQPT) and X.E ′>. V'⊆X. V, X.V. E'⊆X. An XQPT subtree is satisfied when a condition having an inclusion relationship such as E is satisfied.

  That is, a useful subtree structure that frequently appears is searched for as an XML query subtree, and a similar structure is searched based on the XML query subtree. At this time, the searched tree is set as an XML query pattern tree.

  More specifically, referring to FIG. 16, (a) and (b) of FIG. 16 are diagrams respectively showing an example of an XML query pattern tree (Query pattern tree) and an XML query sub-tree (Query Rooted subtree) having a root. .

  As shown in FIG. 16, a query set is created in a query pattern tree and stored to obtain a query pattern tree database XQPT_DB = {XQPT1,..., XQPTn}, and a query pattern mining (query pattern). mining) finds a subtree having all frequent routes that satisfy the minimum support in the database XQPT_DB.

  That is, in XQP_DB, all cases including the sub-tree XQRST having the root are expressed by FRQ (XQRST) indicating the frequency of occurrence. Here, in order to obtain the support value of XQRST, the following formula SUP Using (XQRST) = FRQ (XQRST) / | XQP_DB |, the frequency of occurrence of XQRST can be obtained by dividing by the number of entire databases.

  For σ having a constant value (σ> 0), if XQRST satisfies the condition SUP (XQRST) ≧ σ, XQRST frequently appears in the database XQP_DB.

  For example, referring to FIG. 17, FIG. 17 is a diagram illustrating the structure of a frequent subtree having three query pattern trees and a root.

  In FIG. 17, since XQRST occurs in XQPT1 and XQPT2, the frequency of occurrence of a subtree having a root (Rooted Subtree) is FRQ (XQRST) = 2 and SUP (XQRST) = 2/3.

  As described above, the process of mining frequent query patterns is a process of searching for similar query patterns in order to increase the efficiency of the query, and in order to investigate whether the query patterns are similar to each other in this process, A query pattern tree matching process as shown in [Definition 3] below is required.

[Definition 3] (XML Query Pattern Tree Matching)
When a query pattern tree XQPT and a sub-tree XQRST having a root are given, XQRST is included in XQPT if the following condition is satisfied.

  1. The root node of XQRST and XQPT share the same label.

  2. If nodes wεXQRST and vεXQPT having an XML tree structure match each other, w. label ≦ v. Each sub-tree of w has a relationship included in the XQPT sub-tree.

  That is, when XQRST, which is a query pattern structure proposed by the user, has an inclusion relationship as described above in the relationship between the nodes between the XQPT structures, XQRST can be defined as included in XQPT. The query pattern tree becomes a frequent pattern tree.

  Hereinafter, it is assumed that the following query is actually executed by the user, and a process of searching for an answer to the query will be described as an example.

  Q1: Look for the title of the movie whose hero is “Jean Reno” among the movies whose genre is action and whose production year is 1994.

  In order to find an answer to the above-described query, data such as “genre”, “year”, and “actor”, values to be searched for, and these routes must be expressed based on such a query.

  That is, the query is expressed as // year = “1994” and // genre = “action” and // actor = “Jean Reno”, using such a route as one transaction, and similar with the user's query pattern The process of searching for the pattern is performed through a semantic data mining process based on the FP-growth algorithm.

  Therefore, using the basic definition as described above, semantic query pattern mining for XML data builds FP-Tree, finds frequent XML query pattern structures, and finds useful subtrees that are similar or consistent with user queries. (See Non-Patent Document 16 and Non-Patent Document 17).

  Here, in order to mine a frequent query pattern as described above, in the present invention, an FP-growth algorithm that is more temporally and spatially more efficient than the priori algorithm, which is an algorithm for finding an existing frequent structure, is used. used. In order to perform data mining on a frequent tree pattern using the FP-growth algorithm as described above, a process of constructing an FP-Tree as described below is required (see Non-Patent Document 18).

  More specifically, the process of constructing the FP-Tree is as follows.

  First, an XML query pattern is defined as one transaction, and the frequency of each item is examined through an initial scan in a database composed of such transactions. Here, the frequency of occurrence means a support value of one element item set.

  Second, the items in each transaction are arranged in descending order according to the frequency of occurrence.

  Thirdly, items that are less than the minimum support specified by the user are removed from the items arranged in descending order according to the frequency of occurrence.

  Fourth, after an item that does not appear frequently in the above step is removed from the transaction database, an FP-Tree is constructed using the remaining items.

  Here, FP-Tree is a prefix tree, and each path is represented by a transaction sharing the same prefix in the tree, and each node represents one item.

  Therefore, it is possible to construct an FP-Tree by repeating the above-described process.

  That is, referring to FIG. 18, FIG. 18 is a diagram schematically illustrating a process of sharing a prefix of a FP-Tree construction process performed through the above-described process and a process of showing each transaction. .

  In FIG. 18, FP-Tree expresses a database in a compressed form, and each item is linked to a pointer that is a node link in a header table.

  As described above, the data mining process using the FP-growth algorithm that searches the XML query frequent pattern by constructing the FP-Tree selects items satisfying the minimum support level, and removes items less than the minimum support level. This is done by performing a process.

  Referring to FIG. 19, FIG. 19 is a diagram illustrating a process of constructing a conditional FP-Tree of m as an arbitrary item in a process of performing data mining by performing an FP-growth algorithm.

  As shown in FIG. 19, the process of building a conditional tree for each item and searching for a conditional pattern is repeated.

  Referring to FIG. 20, FIG. 20 is a table collectively showing the conditional patterns obtained as a result of repeating the process of searching for the conditional pattern as described above.

  Next, an example in which the above-described data mining technique is applied when a query is performed by the user will be described.

  First, it is assumed that an XML query is given based on the following XQuery Syntax.

  [Query]: Search for titles, authors, and prices in books written by someone named 'Kim'.

  Referring to FIG. 21, FIG. 21 is a diagram showing an example of a query made in Korean written by the user in XML format.

  Here, latname = “Kim” means that the title, author, and price of the book written by the person named “Kim” should be searched, and resultPattern indicates the schema pattern for the query result. , Predicates indicates filtering conditions necessary for executing the query, and documents indicates an associated XML file.

  Referring to FIG. 22, FIG. 22 is a diagram showing a Book DTD Tree, which is an XML document created based on the XQuery.

  In FIG. 22, “Book DTD Tree” indicates a form of a database including transactions.

  More specifically, in order to perform a data mining process to find frequent query patterns for queries coming from users, a database is first constructed with each query pattern tree generated by a user query as one transaction, The query pattern tree is checked for useful query patterns and stored, and a database consisting of transactions is constructed.

  Next, data mining is performed in the process of searching for a subtree having a root from the database storing the query pattern tree and searching for a similar pattern.

  That is, as described above, an FP-Growth technique for generating a FT-Tree-type structure based on a transaction made up of each query pattern and finding a frequent pattern tree based on the generated FP-Tree. A data mining process using the above is performed, and a frequent tree is mined through such a process to find a maximum frequent subtree.

  Here, the definition of the inclusive relation of the tree required in the process of searching for this is as follows.

[Definition 4] (Tree Subscription)
When S and S _1, which are sets of two subtrees, are given, S = S ₁ ∪ {t ₁ , t ₂ ,..., T _n }.
_{Here, ∀ i (1 ≦ i ≦} n), if _{∃t j ∈S 1, t j <} t i, S 1 is included in _S, expressed as _S 1 <S.

[Definition 5] (Maximum Frequent Subtree)
If the subtree set is a frequent subtree and not a relationship included by some other frequent subtree, the subtree set is a maximal frequent subtree, that is, the relationship {maximal FSP} ⊆ {FSP} is there.

  As described above, in the embodiment of the present invention, the FP-growth technique based on the existing association rule algorithm is used to increase the efficiency of the user query. As an extension of the FP-growth technique, in addition to a technique for finding a similar query pattern for XML data, a method of partitioning a frequent subtree from a non-frequent subtree and removing a non-frequent subtree is applied.

  In addition, a maximum pattern subtree is searched using the FP-tree algorithm, a query pattern input by the user is analyzed and compared with the maximum pattern subtree, and the user query pattern is classified. The user query pattern and the maximum pattern tree are checked to see if they match, and by repeating such a search, the sub-tree entered by the user query is compared with the maximum pattern sub-tree and the maximum Perform a mining process to find the pattern subtree.

  Next, a process of performing data mining that searches for a pattern tree for a user query using situation data will be described.

  That is, in the embodiment of the present invention, semantic data mining is performed using pattern tree data mining (Pattern Tree Data Mining) technique of FP-growth algorithm using situation data (Context Data).

  At this time, as the situation data, tourism data was collected using an internet portal site and an internet site having tourism information for each region. A tourism database was constructed using the collected tourism data, and a tourism ontology was constructed using the widely used Protage_4.0.2.

  The tourism ontology is expressed in RDF and RDFS using the ontology language OWL (Web Ontology Language). In addition to creating an ontology database called owl, a small area corresponding to that area was designated as a subclass under each area, and a sightseeing spot was designated as another subclass under a multi-level subclass.

  In addition, classes such as subclasses, superclasses, and synonym classes (equivalent classes) are specified using the relationship and attributes of tourist sites and each region, and at the relationship between various tourist sites in the same region in each region. An attribute specification such as name place was specified.

  In addition, when a user makes a search request, a synonym class is specified for synonym or initial word processing, a content including a description of a famous dining room corresponding to each sightseeing spot and the sightseeing spot is specified as an attribute, As for the longitude / latitude position attribute, the user can be provided with the situation information of each tourist spot using a Google map service.

  Therefore, the ontology database constructed as described above has a hierarchical structure using RDF and OWL, so that the present invention uses an XML data mining technique to perform a data mining process using a tourism ontology. In order to search for a query pattern that can be used as a basic data mining method and is related to the user, an ontology is used to search for a sightseeing spot that is the same as or similar to the user's query using the existing FP-growth algorithm. We applied a pattern tree data mining technique to find the same or similar patterns stored in.

  More specifically, the tour is designated as root, and using the sightseeing spot information constituting the tourism database, each area is configured by the next subclass relationship, and another subclass is configured thereunder. The hierarchy structure that configures each region and sightseeing spot is specified as one transaction.

  In addition, when a user tries to search for a tourist spot in a specific area, the address value for the tourist spot constituting the query that comes from the user also has a hierarchical structure like the structure stored in the ontology database. Therefore, the data mining process was performed in the process of searching for a transaction in which the structure of the user query and the transaction having the hierarchical structure stored in the ontology database are similar or identical to each other.

  Referring to FIG. 23, FIG. 23 is a diagram showing a frequent item set for each transaction in the tourism ontology database.

  In FIG. 23, a transaction is configured for each sightseeing spot, a table is configured with each transaction as a structure forming a hierarchical structure of a specific sightseeing spot, and a transaction that frequently appears with transaction id is configured. It shows the contents of each sightseeing spot.

  Referring to FIG. 24, FIG. 24 is a diagram schematically showing a part of an FP-tree structure using a tourism ontology database having a frequent appearance structure generated in each region as described above.

  In FIG. 24, an item (item) is configured on the left side using each region name, and a link of the node is illustrated on the right side using items constituting each transaction as one node.

  In addition, the frequency of the items composing each node is expressed by a number next to it.

  That is, for example, in the tourism ontology database, in the case of Jeju Special Self-Governing Province, the frequency of frequent occurrences is 5 times, so it is indicated by the number 5, and in the case of Sogipo City, it occurs 3 times. The FP-tree structure is constructed by continuing such notation for different regional names.

  Further, it is possible to construct a conditional FP-tree for a specific item based on such FP-tree.

  That is, referring to FIG. 25, (a) in FIG. 25 is a global FP-tree, and (b) in FIG. 25 is a conditional FP-tree based on “Mala Road” in a tourist area. It is a figure which shows the case where it each constructed | assembles.

  As shown in FIG. 25, the FP-growth algorithm performs a process of repeatedly generating a conditional FP-tree for a specific item based on the constructed FP-tree, and provides conditional processing for each such item. The process of repeatedly generating the FP-tree and searching for a frequent FP-tree structure is a data mining process.

  More specifically, for example, when a query is entered by a user to search for “Jeung An Mountain Recreation Forest in West District, Sejong-gu,” the process of investigating the tourism ontology database to find a query similar to the user query If there is the same structure as such a user query, it is possible to provide services to the user accurately and quickly, but if you find a structure that only matches up to "Jeong An Road, Nishi-gu, Daejeon Metropolitan City" Are not accurate, but are not accurate, and can provide results for similar structures.

  Referring to FIG. 26, FIG. 26 is a diagram showing a part of an FP-growth algorithm source that performs the above-described process (see Non-Patent Document 19).

  Next, a process in which data mining is performed by applying the FP-growth algorithm as described above will be described in more detail.

  First, in order to apply the ontology database to data mining, a preprocessing process for encoding data is performed.

  For this purpose, the present inventor considered the process of converting data corresponding to each region into an appropriate code and applying a FP-growth algorithm to generate a pattern.

  Here, since the FP-growth algorithm also applied the minimum confidence and the minimum support in the same manner as the priori to which the association rule is applied, such a threshold value is applied. When we examined the results of applying data mining.

  In more detail, referring to FIG. 27, FIG. 27 is a diagram illustrating a result of performing data mining by applying a minimum support level of 0.4 to the FP-growth algorithm.

  As shown in FIG. 27, the number of transactions stored in the database is 99, the maximum memory usage is 0.96 mb, and 27 frequent item sets are generated. The time taken to perform data mining is 15 ms.

  Also, it can be seen that a pattern is generated for each stage while applying the support value based on the minimum support value 0.4 while performing the five steps L1 to L5. The value in each pattern means a value obtained by transforming the value of each region in the database.

  Therefore, in FIG. 27, it can be confirmed that the frequent occurrence pattern for the sightseeing spot in “Seogwipo City” of “Jeju Special Self-Governing Province” occurs most frequently.

  The process of performing data mining using situation data based on the FP-growth algorithm has been described above. Next, a general process in which information is searched by performing a semantic data mining process in a mobile environment will be described in detail.

  In other words, the present invention provides a relationship between similar words or words having the same meaning, such as class relationship setting and attribute setting between regions, in order to accurately and concisely provide useful information requested by the user at the current position. A tourism ontology was used to express the settings. In the above-described embodiment, the tourist spot ontology is constructed in order to accurately transmit the tourist spot information of the area desired by the user.

  In addition, the sightseeing spot ontology is classified into each region and expresses information about the sightseeing spot using an ontology using a hierarchical structure. For example, in a state where the user is currently located in the Daejeon City Yusong District, When the query “Search for a sightseeing spot located” is performed, the user searches for the Yusong District area included in Daejeon Metropolitan City via the ontology, and the user is provided with the sightseeing spots included in the Yusong District. As a search result, along with this, provide a map of the current location and the tourist destination desired by the user via a map function attached in the mobile device, such as Google map, etc. It is possible to provide a search service for a sightseeing spot in the region closest to the user position desired by the user integration.

  In addition, as described above, the FP-growth algorithm can be applied to perform a data mining process on the data in the tourism ontology database to provide useful information to the user. The semantic information retrieval process performed by applying the mining algorithm will be described in detail.

  First, referring to FIG. 28, FIG. 28 is a diagram schematically showing an overall configuration of a mobile-based information search system utilizing a pattern matching technique using a semantic data mining algorithm.

  In the system shown in FIG. 28, the operations performed by each module and the overall flow are described as follows.

  First, an information search request is made at the current location of the user.

  That is, when a search request is received from a user using a mobile device, information on the current location of the user is transmitted along with the user search request using WiFi, 3G, 4G, GPS, sensor (Sensor), etc. Context information is collected.

  Second, build a context database.

  Here, in this embodiment, in order to collect sightseeing spot information, the sightseeing spot information introduced in each Internet portal site and each area was collected, and various sightseeing spots in each area and characteristics of the sightseeing spots were collected. Longitudinal and latitude values for each area were collected for address, surrounding cafeteria, and Google map settings.

  In addition, the data collected in this way is subjected to a purification process and a pretreatment process, and stored in a context database.

  Third, an ontology inference process is performed on the data stored in the situation database through the process described above, and an ontology database is constructed by setting attributes and relationships of each data value.

  Here, in the ontology inference process, the subclass of each region, the superclass relationship, the synonym class for the synonym or initial word processing (equivalent class) relationship, between each region and the tourist destination An attribute such as “at the same place of” that expresses a relationship between “is attraction of” attribute and a sightseeing spot that belongs to the same area is set.

  Fourth, in order to improve search efficiency using a database stored in the ontology, a semantic data mining process is performed to search for a query pattern tree that is similar to or matches the query requested by the user.

  Here, the process of searching for a pattern tree using search information and situation information requested by a user is performed by a data mining process using an existing FP-Growth algorithm.

  Fifth, a service is provided that provides a result obtained by performing the above-described process as a result of information requested by the user through a context-aware mobile service interface (Context Awareness Mobile Service Interface).

  Therefore, as described above, it is possible to realize an overall framework in which information retrieval is performed by performing a semantic data mining process in a mobile environment, and subsequently, a semantic data mining technique is applied in a mobile environment to apply a user's A general process for searching for desired information will be described in detail.

  That is, referring to FIG. 29, FIG. 29 is a diagram schematically showing a series of processing steps performed when a search request for a sightseeing spot is received from a user.

  More specifically, as shown in FIG. 29, first, when a user requests an information search for a sightseeing spot to be searched by the user through a service request module, a mobile search system using an ontology. Analyzes the search requirements entered by the user and constructs a query pattern.

  Second, the user's current information is collected through the context processing module, that is, as described above, including the information content queried by the user, WiFi, 3G, 4G, GPS, sensor ( The current position and the user's requirements in the situation where the user requests the search via the Sensor) is checked.

  Third, the tourism content DB is a context database described above, and it collects tourist site information on tourist site introduction sites in each region including portal sites. This is a tourist destination database.

  Here, the sightseeing spot database is preprocessed to refine the data and stored in the relational database, and then the ontology inference process is performed to construct the sightseeing spot ontology database.

  Therefore, in order to search for a pattern that matches the user query pattern requested by the user using the query pattern stored in the situation database stored in the ontology form in this way, the relationship of the hierarchical structure to the tourist destination A similar or identical query pattern is searched by searching the tourism ontology database configured by the above, and at this time, as described above, a similar or identical query pattern is searched using the semantic data mining algorithm using the FP-growth algorithm. .

  Fourth, a query pattern is searched from the ontology database for information requested by the user via a search module (Retrieval Module), and a search result desired by the user is provided to the user together with GPS information.

  Next, a data mining process for searching for similar or identical pattern trees will be described.

  First, a database is constructed by collecting information related to sightseeing spots using an Internet site, etc., and a user query pattern is searched when a user queries an ontology database constructed using such data. Let it be a database (Pattern Database). Since various patterns are stored in such a pattern database, when a user performs a query, a process of comparing with an existing pattern to find a matching pattern, a similar pattern, and a pattern that does not match Data mining is performed by

  Therefore, when searching for a pattern that matches or is similar to the user's query, it is possible to provide the user with useful information desired by the user accurately and quickly.

  Referring to FIG. 30, when a search request is received from a user, the user's requirements are analyzed to analyze the user's query pattern, and a pattern matching the analyzed query pattern is stored in the ontology database of the search system. It is a figure explaining the process in which a similar or identical pattern structure is searched by comparing with the stored pattern structure.

  Here, in FIG. 30, the pattern database means a tourism ontology database.

  In other words, when a query request for a sightseeing spot is made at the user's current location, a preprocessing process and an ontology inference process are performed, and the stored tourism ontology database stores the query pattern and pattern database currently requested by the user. Whether or not the query pattern matches is checked, and if it matches or is similar, it is shown as a search result, and if it does not match, a pattern is further searched.

  More specifically, when a user requests a search for a sightseeing spot desired by the user in the search window of the search system and a search request is received from the user, the user query pattern is analyzed to generate a user query pattern. .

  Next, it checks whether the user query pattern is a new pattern. If the user's pattern is a new pattern, it checks whether it matches the existing pattern. A matching pattern stored in a database (pattern database) is presented as a search result.

  If it is not checked whether the new pattern matches the pattern in the pattern database, this process is repeated, and if there is a similar pattern, the similar pattern is presented as a search result, and the user The new pattern queried by is stored as a new pattern in the pattern database.

  Therefore, as described above, according to the embodiment of the present invention, the above-described series of processes is repeated, and the query data matching or similar to the user's query is presented as a search result. It is possible to realize a search method for searching for user interest information using a semantic data mining algorithm based on the FP-growth algorithm from a tourism ontology constructed as data, and thereby a user interest information recommendation service based on context awareness Can be provided.

  That is, referring to FIG. 31, FIG. 31 is a diagram schematically illustrating an overall configuration of a user interest information search service providing method based on context awareness using semantic data mining according to an embodiment of the present invention.

  More specifically, as shown in FIG. 31, a user interest information search service providing method based on context awareness using semantic data mining according to an embodiment of the present invention includes a mobile device including a user's own smartphone or tablet PC. When a search request for interest information is transmitted at a current location using a terminal, the mobile terminal's WiFi, mobile communication network (3G, 4G, etc.), GPS, and various sensors installed in the terminal are used together with the user's search request. Then, information on the current location of the user is received and collected as status information of the user (Context Information) (step S311).

  Next, a refinement process and a pre-processing process are performed on the collected information to construct a context database (step S312), and each data stored in the constructed situation database is subjected to ontology inference. Then, the ontology database is constructed by setting the attributes and relationships of the respective data values (step S313).

  Then, using the ontology database constructed through the above-described process, semantic data mining that searches for a query pattern tree that is similar to or matches the query requested by the user based on the FP-Growth algorithm as described above. Processing is performed (step S314), and the result obtained through the semantic data mining processing step is displayed on the user's mobile terminal as a search result for the search request requested by the user (step S315).

  Here, more specific contents of the processing performed at each stage described above can be realized as described above with reference to the prior art documents and FIGS. It should be noted that detailed description of each step is omitted here.

  Next, the environment necessary for realizing the mobile information retrieval system using the situation recognition ontology as described above and the implementation process will be described in detail.

  That is, in the above-described embodiment, the present inventor collected and collected sightseeing spot information data using a site that introduces sightseeing spots in each region including a portal site in order to collect sightseeing data. A preprocessing process for refining data into information was performed, and for example, a database was constructed using MS SQL Server 2000 Personal version in order to construct a relational database in an operating system environment such as Windows (registered trademark).

  In the database, the fields of id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, and long are configured to store data. Here, each field such as add_1, add_2, add_3, add_4, and add_5 indicates the address constituting each sightseeing spot, tour_site stores the name of the sightseeing spot, and syn is a synonym or initial speech processing. Lat and long mean the latitude and longitude necessary for setting a map (Google map), respectively.

  In addition, the present inventor constructed a tourism ontology by setting relationships and attributes between data using such a relational database. At this time, ontology is constructed and inferred using Protage_4.0.2 version for construction of ontology database, and ontology is expressed using OWL and RDF. The owl ontology was constructed.

  In addition, the present inventor constructed an information retrieval system in a mobile environment by performing semantic data mining using the FP-growth algorithm using the ontology database constructed as described above. Here, the implementation environment for realizing the mobile information search system is, for example, implementation of Android 4.3 version and Google map (Google map) as presented in [Android 2014] (see Non-Reference Document 20). For this purpose, Google APIs 4.3 was used. HTML5, CSS3, JavaScript (registered trademark), etc. were used to realize the mobile information retrieval system, and Node JS server utilized in a large-capacity database for use as a database and server.

  Since Android programming is based on Java, an implementation environment is configured using Jdk 1.7.0_05. As an editor for building a mobile environment information search system, for example, WebStorm-7.0.3 version as presented in [WebStorm 2014] is used (see Non-Reference Document 21).

  Here, the characteristics of Android programming, which is a mobile platform applied to the present embodiment, will be described. The characteristics of Android, which is a platform for mobile devices, are as follows. First, Android is an open source mobile platform created and released by Google. It is a combination of Linux (registered trademark) operating system, user interface and application program optimized for mobile environment. Unlike the closed mobile phone environment, it is possible to develop and operate various applications with minimal restrictions.

  In addition, the detailed contents of the development process using Android can be easily understood by those skilled in the art with reference to prior art documents such as [Michallis 2012] (see Non-Reference Document 2). Here, in order to simplify the description, detailed description thereof is omitted.

  That is, as described above, the inventor builds a tourism database using information on an Internet portal site or a site that guides tourism information of each region, and the id, add_1, add_2, add_3, Fields such as add_4, add_5, tour_site, syn, lat, and long are configured. Here, for example, if add_1 is composed of regional names such as “Taejeon Metropolitan City”, “Jeju Special Self-Governing Province”, “Busan Metropolitan City”, “Chungnam”, add_2 is the Daejeon Metropolitan City In this case, it is composed of the names of wards such as “Nishi-ku,” “East-ku,” “Naka-ku,” “Yusong-gu,” “Tedok-gu,” etc., and add_3 is “Mannyong-dong” located in West-gu The name of the cave is designated as “Eunhen-dong” located in Jung-gu, Daejeon, and add_4 is located on “Dangsan-daoji” located in Mangyeong-dong, West-gu, Daejeon. Designate a detailed address, such as “Jungjan-ri”, and add_5 is “169 address” in Mangyeong-dong, Dungsan-ro, West-gu, Daejeon. The lot number, such as the position "address 52" can be configured to specify.

」という）」の初声語たる「

」、テジョン文化芸術団地（韓国語で「

」という）の初声語たる「

」などのように入力する初声語処理のために指定でき、ｌａｔとｌｏｎｇは地図（グーグルマップ）を表示するために各地域の経度値および緯度値を指定するように構成できる。 Also, tour_site is the name of a tourist destination such as “Taejeon Culture and Arts Complex” located at 169, Dungsan-ro, Mangyeong-dong, West-gu, Daejeon, and “Gapsa” located at 52, Junjan-ri, Kelong-myeon, Chungnam-jang City. Syn, for example, synonym processing used when “Taejeon” is designated as “Hambat” or “Jeju” is designated as “Tamura”, or “Gapsa ( in Korean"

")")

"Daejeon Cultural Arts Complex (" Korean "

))

"Lat" and "long" can be configured to designate a longitude value and a latitude value of each region in order to display a map (Google map).

  In the embodiment of the present invention described above, the tourism database stored in this way can be used to construct a tourism ontology database, and for ontology inference and ontology construction using the attributes and relationships of each data in the database. Using Protege_4.0.2 version and using OWL and RDF to tour. We built an ontology database called owl.

  Here, the ontology inference process specifies attributes and relationships of data values indicating each region, that is, specifies superclass and subclass relationships and multiple inclusion relationships for each region, and is synonymous or If you specify a synonym class for initial speech processing (equivalent class) and are included in the same area, set an attribute relationship such as “at the same place of” In the case of a sightseeing spot included in the area, for example, a tourist spot called “Ushijima” on the Udo side of Jeju Special Self-Governing Province sets an attribute relationship such as “is attraction of” in relation to the area and the tourist spot. It can be configured as follows.

  In the above-described embodiment of the present invention, semantic data based on the FP-Growth algorithm is used to provide the user with more accurate and quick information retrieval results using the ontology database constructed in the above process. A mining process (Semantic Data Mining) was performed.

  Here, in the data mining method described above, one pattern tree is formed by considering the hierarchical structure relations of the upper class, lower class, and sightseeing spot constituting each region of the ontology database as one transaction. Based on the pattern tree stored in this way, when a user query is entered, the user query pattern is compared with the existing pattern tree stored to find a matching or similar pattern tree. Done by the process.

  Therefore, the data mining process performed as described above can be performed to provide a search result that matches or resembles the search term requested by the user.

  Next, a process of constructing a data search system that performs a search for a user's query through a semantic data mining process based on the FP-Growth algorithm as described above in a mobile environment will be described.

  Here, in the embodiments of the present invention described below, the present invention will be described by taking as an example a case in which a search system is realized by using a program such as HTML5, CSS3, or JavaScript for realizing an information search system in a mobile environment. However, it should be noted that the present invention is not necessarily limited to such cases and can be implemented in any number of other languages or platforms as required.

  In other words, when implementing a system using Android programming, it can be executed only on a device having the Android operating system. However, as described above, when programming using HTML5 and CSS3, only the Android operating system is used. In addition, there is an advantage that it can be executed even in an iOS operating system.

  In the embodiment of the present invention described below, JavaScript is used for processing of event actions (event actions) such as buttons and functions, and the program used as an editor at this time is WebStorm_7.0.3 version. (See Non-Patent Document 21).

  In addition, Android 4.3 version was used to execute the constructed system (see Non-Patent Document 20), and Google APIs 4.3 for Google map was used to realize each region's map (Google map). .

  More specifically, referring to FIG. 32, FIG. 32 is a diagram showing a part of a source for realizing the initial screen of the search window.

  That is, the source shown in FIG. 32 indicates a process in which, when a tourist destination desired by the user is input to the search window, the result is displayed as a list on the next page.

  Referring to FIG. 33, FIG. 33 is a diagram illustrating a source in which an event action (Event action) for a search button (search button) is realized by a script function (Script function).

  That is, the source shown in FIG. 33 changes the page when the user enters a search term in the search window and clicks the button next to it or clicks the three buttons at the bottom of the main screen. The button action in which the result value is displayed on the screen by clicking the button is shown.

  Referring to FIG. 34, FIG. 34 is a diagram illustrating a part of a source for performing a map button action for providing a map (Google map) service.

  That is, the source shown in FIG. 34 is an operation showing a Google map using the event action and function of btn-map, and an operation showing a map corresponding to each position using the latitude value and longitude value of the region. Indicates the button action to be performed.

  Referring to FIG. 35, FIG. 35 is a diagram illustrating a configuration example of a main screen of the mobile information search system according to the embodiment of the present invention.

  As shown in FIG. 35, when an area name or a sightseeing spot name is entered in the search window on the left main screen and the search button is clicked, a search result corresponding to the corresponding area or sightseeing spot is output, that is, for example, “Tejon”. If the region value is input to the search window, the name of the sightseeing spot related to “Tejon” is output as shown in the right screen.

  Here, if another value that is related to the subclass of Daejeon Metropolitan City is input as the address value, the search result presents a finer result value.

  Further, as shown in the right screen of FIG. 35, a plurality of buttons are displayed at the lower end (footer) portion of all pages except the main screen of the search system, that is, for example, the first button Is a button that returns to the home, the second button is a map button that reads the value entered in the search window and calls the map of the related area, and the third button is the name of the cafeteria in the area that belongs to the related sightseeing spot Or it is a button which shows the page of the content to introduce with respect to a sightseeing spot.

  More specifically, referring to FIG. 36, FIG. 36 shows a configuration in which a map of a corresponding area is displayed when a region value of Daejeon Metropolitan City is entered in the search window and a map button is clicked, and is input to the search window. It is a figure which respectively shows the structure by which the screen which introduces the restaurant or sightseeing area of a related area according to the input value displayed is displayed.

  Here, the map button described above uses the latitude value and the longitude value set as the field values of lat and long in the database to set the map (Google map) in the explanation about the tourism ontology described above, and displays the map of the related area. Can be configured to bring as a result value.

  37 and 38, FIG. 37 automatically recognizes the user's current position value as a latitude value and a longitude value, and shows a map showing a map of the region where the user is currently located along with the latitude value and the longitude value. FIG. 38 is a diagram showing a result screen realized through the source as shown in FIG. 37. FIG.

  As shown in FIG. 37, the map realization source creates a myLocation function in order to realize a map corresponding to the current position of the user, and navigator. geolocation. The current position value of the user can be obtained using the getCurrentPosition function.

  In addition, the result screen shown in FIG. 38 uses the values of latitude 36.30 and longitude 127.36 as the current position value of the user to obtain the position value of Jeonrim-dong, Nishi-gu, Daejeon Metropolitan City, and displays a map of the surrounding area. The screen to be displayed is shown.

  Therefore, as described above, by realizing the operation of transmitting information at the user's current position using the user's current position value and the operation performed when the user inputs a search term, the mobile environment It is possible to realize an information retrieval system using the semantic data mining technique.

  Next, experimental results obtained by actually performing information retrieval using the mobile information retrieval system implemented as described above and performance evaluation based on the experimental results will be described.

  That is, the inventor constructed a tourism ontology database as described above, and conducted an experiment by applying a data mining algorithm using data in the corresponding database.

  Here, the experiment conducted in the embodiment of the present invention described below is for tourism related to each region with the goal of obtaining a search result applying a semantic data mining technique for searching for a tourist destination for each region. In addition to analyzing frequent patterns using the land as one transaction, we also experimented with the accuracy of information retrieval in order to quickly provide search results for tourist destinations desired by the user.

  First, the experimental results of applying the semantic data mining algorithm will be described. The inventor applied the FP-growth algorithm as an algorithm for searching for user query patterns associated with each other in order to search for sightseeing spots associated with each region. The performance of the ontology database and the existing relational database were compared and evaluated.

  In other words, the process of finding a frequent pattern between two data by applying a semantic data mining algorithm based on the FP-growth algorithm, the execution time, the memory usage, and the like were compared. In order to perform such experimental evaluation, a pre-processing process for tourism ontology data was performed.

  More specifically, firstly, a tourist database is constructed by collecting information related to each tourist spot posted on various portal sites and sites introducing tourist spots, and secondly, data stored in the tourist database. Sightseeing ontology tour. owl was constructed. Here, such an ontology was constructed using protege_4.0.2, and its inference and expression used OWL, RDF, and RDFS.

  Third, the XML document was parsed using a DOM parser (Parser) against the tourism ontology database. Fourth, a mapping table is created to extract the tags of the XML document and perform the FP_growth algorithm, and each tag is coded to perform a mining process.

  Also, in the case of existing relational database data, tourism data was extracted, a pre-processing process was performed in a process similar to the ontology data, a mapping table was created, and a data mining process was performed by encoding.

  That is, referring to FIG. 39 and FIG. 40, FIG. 39 and FIG. 40 are experimental results of performing a data mining process by applying the FP-growth algorithm to the ontology database and the existing relational database according to the embodiment of the present invention, respectively. FIG.

  Here, in the case of existing relational data, unlike ontology data, it is possible to conduct experiments by applying a minimum support value of only up to 0.6. It was impossible, and the experiment was conducted only up to that point.

  More specifically, referring to FIG. 41, FIG. 41 illustrates a process of changing memory usage according to a change in minimum support threshold value of ontology data and an existing relational database according to an embodiment of the present invention. It is a graph.

  That is, the present inventor limited the number of transactions to 100 and investigated the memory usage, the number of frequent patterns, the execution time, etc. due to the change in the minimum support value. First, as a result of considering how the memory usage changes with a change in the minimum support value between the two data, as shown in FIG. 41, the memory usage between the two data is similar. I understand that.

  Therefore, from the experimental results shown in FIG. 41, when the number of transactions is small, the ontology database data according to the embodiment of the present invention uses a little less memory than the existing relational database data. , Can be seen to be relatively similar.

  Next, referring to FIG. 42, FIG. 42 similarly restricts the number of transactions to 100, and frequent occurrences due to changes in the minimum support value of the ontology data and the existing relational database according to the embodiment of the present invention. It is a graph which shows the change of the number of patterns.

  Here, such an experiment was performed on the premise that the frequent pattern structure has a different form between the ontology database and the general database depending on what structure data is used as the experimental data. As shown in FIG. 42, considering the change process of the number of frequent patterns between ontology data and general data, it is confirmed that more frequent structures are generated than the general data when using the ontology database. be able to.

  This is because an ontology database is a database having a more hierarchical structure than a general database, and thus more frequent structures can be generated.

  In other words, in the process of building an ontology database as described above, each class relationship is structured to be connected to each other via ontology inference, so that it has more frequent structures than general data. As a result, according to the present invention, when searching for frequent patterns and providing search results, more frequent structures can be presented, so that more excellent search results can be presented.

  Next, referring to FIG. 43, FIG. 43 is a graph showing a change in run time due to a change in minimum support of ontology data and an existing relational database according to an embodiment of the present invention.

  Similarly, in this experiment, the experiment was performed with the number of transactions limited to 100. As shown in FIG. 43, it was confirmed that the ontology database takes a relatively short execution time compared to existing relational data.

  Here, in the case of relational data, the singularity increases rapidly when the minimum support value is small and when the minimum support value is the largest, but in contrast to this, in the case of an ontology database The change process of the execution time is relatively similar with the change of the minimum support value.

  Next, referring to FIG. 44, FIG. 44 is a graph showing a change in execution time due to a change in transaction between ontology data and an existing relational database according to the embodiment of the present invention.

  That is, in FIG. 44, the minimum support value is fixed to 0.4, and the change in the execution time of the ontology data and the existing relational database according to the embodiment of the present invention is shown while changing the number of transactions. While the number of transactions increases, it can be confirmed that there is a difference in the execution time between the two data. The data of the ontology database according to the embodiment of the present invention is compared with the data of the existing relational database. It can be seen that it takes less time to execute.

  Therefore, it can be understood that the information search using the ontology database according to the embodiment of the present invention takes less time to search when a user's search request is entered than the information search using the existing relational database.

  Next, referring to FIG. 45, FIG. 45 is a graph showing a comparison of changes in memory usage due to changes in transactions between the ontology database and the existing relational database according to the embodiment of the present invention.

  Here, the result shown in FIG. 45 is a result of an experiment conducted while keeping the minimum support level constant at 0.4 and changing the number of transactions. As shown in FIG. 45, the amount of memory used due to the change in the number of transactions between two data is that the ontology database data according to the embodiment of the present invention is relatively small compared to the existing relational database data. I understand.

  In addition, when the number of transactions is larger than when the number of transactions is smaller than when the number of transactions is limited to 100, when using the ontology database according to the embodiment of the present invention, an existing relational database is used. It can be seen that the memory usage is more efficient than using it.

  Therefore, in view of the shortage of mobile device capacity, which is one of the biggest problems in information retrieval in the mobile environment, it can be seen that it is more efficient to use the ontology database according to the embodiment of the present invention.

  Next, the results of experiments conducted on the accuracy of information retrieval of the ontology database and the existing relational database according to the embodiment of the present invention will be described.

  First, referring to FIG. 46, FIG. 46 is a graph showing a change in accuracy indicating search efficiency in accordance with a change in a transaction which is a user query pattern.

  That is, the inventor measures the accuracy (Accuracy) indicating the accuracy of information retrieval with respect to experimental data using the FP-growth algorithm used as the semantic data mining algorithm applied to the above-described embodiment of the present invention. did.

  As a result of the measurement, as shown in FIG. 46, it can be confirmed that the accuracy of the ontology database and the existing relational database according to the embodiment of the present invention increases as the number of transactions increases.

  However, when comparing the accuracy of an existing portal site or a tourist site-related site search system with the mobile information search system using the semantic data mining technique constructed according to the embodiment of the present invention as described above, It can be seen that the accuracy of the system is extremely low.

  This is because the existing system provides a search result not related to the search content desired by the user when the search word desired by the user is input to the search window.

  Next, in order to consider the accuracy of information retrieval by comparing the difference between the mobile information retrieval system to which the semantic data mining technique according to the embodiment of the present invention is applied and the existing retrieval system, The result of comparing the result screens when the tourist information is input to the search window in the mobile information search system according to the embodiment and the existing portal site will be described.

  That is, referring to FIG. 47, FIG. 47 is a diagram showing a screen for displaying a search result when “mount Kerong” is input to the search window in the existing portal site.

  More specifically, as shown in FIG. 47, the search result of the existing system is the contents related to the actual tourist spot in the contents displayed as the search result value when the tourist spot “Mount Kelong” is entered in the search window. May also be displayed as search results, but the search results provide too many results compared to the smaller smartphone screen size than the PC, There is an inconvenience that the user has to check the search result while moving the scroll of the screen downward.

  Therefore, in order to eliminate such inconvenience, there is a need for a search system that provides only accurate search results desired by the user as search results. Therefore, in the present invention, in order to solve the problems of the existing search system as described above, a mobile information search system using a semantic data mining technique has been proposed and realized.

  More specifically, referring to FIG. 48, FIG. 48 is a view showing a screen for displaying a search result when “Mount Kelon” is input to the search window in the mobile information search system according to the embodiment of the present invention.

  In other words, the existing search system has a drawback in that the contents not related to “Keron” are provided on the result screen, and the information desired by the user must be checked and selected one by one. 48, the mobile information search system according to the embodiment of the present invention, as shown in FIG. 48, only an accurate value is obtained when the same “mount Kerong” as the value entered in the search term is entered in the search window of the existing search system. As the result value.

  Referring to FIG. 49, FIG. 49 shows a screen that provides both a description related to a search result and a map of the surrounding area as a result value when a search word is input in the mobile information search system according to the embodiment of the present invention. FIG.

  As shown in FIG. 49, when the mobile information search system according to the embodiment of the present invention is used, a search result for an input search word is provided, and for example, a more specific explanation about a related tourist spot and a surrounding area are provided. Since a map of the surrounding area including related information such as a delicious restaurant is provided together as a result value, it can be seen that the accuracy of the search result is more accurate than the existing search system.

  Therefore, as described above, a mobile information retrieval system to which a semantic data mining technique is applied using a situation recognition ontology in a mobile environment can be realized.

  That is, according to the present invention, first, in order to store the situation data, the tourist information is collected through the tourist site introduction sites in each region including the portal site, and the collected information is preprocessed. The situation recognition ontology for providing user interest information service that builds a tourism ontology database by performing an ontology inference process that sets the relationship and attributes of each sightseeing spot using the tourism data thus stored A method can be provided.

  In addition, according to the present invention, a data mining technique based on an existing FP-growth algorithm is used to provide a user with a quicker and more accurate information search result using the tourism ontology database constructed as described above. By applying, it is possible to compare the user query pattern entered into the user with the query pattern stored in the tourism ontology database, and to quickly provide a search result for a query having a similar or identical structure. It is possible to provide a user interest information search service providing method using semantic data mining, which is configured as follows.

  In addition, according to the present invention, the situation recognition ontology and semantic data mining as described above are used to present a more concise and accurate search result suitable for the mobile environment, and at the same time the user's current position or the user's desired A mobile information search service configured to be able to perform a map service for the position of the information search result together can be provided.

  In addition, according to the present invention, it is possible to input various search terms while performing a search service by enabling processing of synonyms, similar words or initial words by attribute setting and relationship setting in the ontology inference process. By applying the semantic data mining technique, it is possible to solve the problem that it is not possible to overcome the disadvantage that the screen of the smartphone terminal is small by providing unnecessary search results with the existing search service.

  As described above, the context awareness ontology based on context awareness and the user interest information service using the same can be realized according to the embodiment of the present invention.

  Here, in the above-described embodiment of the present invention, in order to predict and provide interest information related to a sightseeing spot to a user, a situation awareness ontology regarding the sightseeing spot is constructed, and such a situation recognition ontology is constructed. For example, the present invention provides an interest information service configured to be suitable for a mobile environment by performing a semantic data mining process on the basis of a user and providing a more accurate and concise result to a user. However, the present invention is not limited to such a case. That is, it should be noted that the present invention can be applied in various ways to the interest information service for any other fields as necessary, in addition to the above-described interest information service for the sightseeing spot.

  Therefore, as described above, the situation recognition ontology realizing method for providing the user interest information service based on the context awareness according to the present invention can be realized.

  In addition, as described above, according to the present invention, by realizing a method for realizing a situation recognition ontology for providing a user interest information service based on context awareness according to the present invention, according to the present invention, user interest based on context awareness technology is achieved. By constructing a context ontology model for predicting the degree and more accurately predicting what kind of search the user wants at which position, and predicting user interest information based on such context ontology model, the user To provide a user interest information service based on context awareness that is configured to provide a more accurate and concise search result by presenting only information related to user interest in response to a search request A situation awareness ontology implementation method is provided. Therefore, it is possible to solve the problem of the search method of the prior art that has a disadvantage that it provides a huge amount of search results in response to a search request of the user and the user needs to further filter only necessary information. it can.

  In addition, according to the present invention, it is possible to provide context-awareness that is configured to be able to provide a search result summarized more accurately using a semantic data mining algorithm using a context ontology (Semantic Data Mining Algorithm). By providing a method for realizing a situation recognition ontology for providing a user interest information service based on the above, it is possible to overcome the shortcomings of a limited screen and a small memory of a mobile terminal such as a smartphone.

  In addition, according to the present invention, the search result for the user search request information is provided, and at the same time, the current location of the user and the location of the search result requested by the user are searched using the GPS and the map function built in the mobile device. By providing a context awareness ontology implementation method for providing a user-aware information service based on context awareness, which is configured to be able to be provided along with the results, the user can obtain the content of the desired information with only one search. You can easily find not only the position but also the position.

  As described above, the details of the method for realizing the situation recognition ontology for providing the user interest information service based on the context awareness according to the present invention has been described using the embodiment of the present invention as described above. The present invention is not limited to the contents described in the above-described embodiments. It goes without saying that the present invention can be modified, changed, combined, and replaced by a person having ordinary knowledge in the technical field to which the present invention belongs, according to design needs and various other factors. is there.

Claims (11)

In a method of realizing a context awareness ontology implemented by a computer ,
A user interest information database construction stage in which the computing device of the computer collects information on the region of interest of the user and constructs a database;
An ontology inference stage in which the computing device of the computer sets a relationship and an attribute between each data stored in the user interest information database constructed in the user interest information database construction stage;
An ontology database construction stage in which the computing device of the computer constructs an ontology database based on the relationship and attributes of each data set in the ontology inference stage ;
A pattern search stage in which the computing device of the computer searches for a pattern tree for a query input by the user using the constructed ontology database;
The computer computing device includes a result providing step of providing the obtained pattern tree to the user via a service interface.
A method for realizing a situation recognition ontology , wherein the computing device of the computer executes the pattern search step by data mining using an FP-growth algorithm .   The ontology includes four types of user status information including a user identifier U (User identification), spatial information L (Location), time information T (Time), and service information S (Service) associated with the user. The method for realizing a situation recognition ontology according to claim 1, wherein the situation recognition ontology is configured to be expressed by space-time situation information O = {U, L, T, S} having dimensions. The time information is information divided at predetermined time intervals,
The spatial information is generalized position information corresponding to coordinates (xi, yi) in the Euclidean space,
The method according to claim 2, wherein the coordinates mean a longitude value and a latitude value of an area where the user is currently located. Computing device of the computer, the user interest information database construction stage,
Collecting information on tourist spots as the user's interest information for each region,
Specify the root class as the top class, specify each region as a subclass,
Designate the name of the tourist destination for each said region as a subclass for each said region,
Designate a detailed description of the facilities in each tourist destination or the relevant tourist destination as a subclass for the name of the tourist destination,
A synonym or initial class search for a synonym or initial voice search is set for each of the areas or the tourist spots, and a process for constructing a hierarchical tourism database is executed. The method for realizing a situation recognition ontology according to claim 3, wherein: The tourism database is
It includes a plurality of fields including id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, long.
The add_1 field, the add_2 field, the add_3 field, the add_4 field, and the add_5 field indicate the address of each tourist spot,
The tour_site field stores the name of the tourist destination,
The syn field is a field for synonym or initial word processing,
The method of realizing a situation recognition ontology according to claim 4, wherein the lat field and the long field are configured to mean latitude and longitude necessary for setting a map, respectively. Computing device of the computer, the ontology reasoning steps,
“Place” consisting of the name of the area where you are going to visit
“Attraction” consisting of the names of tourist attractions in each region;
“Resource” composed of buildings, relics, ruins, beaches and canteens in each tourist destination; and “Cultural activity” meaning cultural action, “Scenery view” meaning act of enjoying scenery, and "Dinin" means the act of eating out
g out ”means“ activities that the user actually performs at each sightseeing spot ”
based on "ty"
The method for realizing a situation recognition ontology according to claim 5, wherein the method is executed to determine an attribute of each data. Computing device of the computer, the ontology reasoning steps,
Specify the attribute relationship between the “Place” and the “Attraction” as “hasAttraction” and “IsAttractionOf”
The attribute relationship between the “Attraction” and the “Resource” is designated as “hasResource” and “IsResourceOf”,
7. The situation recognition ontology implementation method according to claim 6, wherein the attribute relationship between the "Activity" and the "Resource" is executed so as to specify "hasActivity" and "IsActivityOf". Computing device of the computer, the ontology reasoning steps,
The method for realizing a situation recognition ontology according to claim 7, wherein when a plurality of attribute relationships exist in the data constituting each class, the situation recognition ontology is executed so as to be regarded as having a multiple attribute relationship. Computing device of the computer, the ontology reasoning step, and executes using Protege_4.0.2 verion, situation recognition ontology implemented method of claim 1. Computing device of the computer, the ontology database construction stage,
The ontology database is structured to be executed using the Web Ontology Language (OWL), the Resource Description Framework (RDF), and the Resource Description Framework Specification (RDFS). Realization method of situation recognition ontology. Billing with either situational awareness ontology constructed by situation recognition ontology implemented method according to one of claim 1 to 10, the user interest information service providing method based on context awareness.

Images