CN105786897A - Context awareness ontology construction method for providing user interest information service based on context awareness - Google Patents

Abstract

The present invention relates to a method of predicting a user interest based on a context awareness technology. According to the present invention, there is provided a method of implementing a context awareness ontology for providing a user interest information service based on context awareness, wherein a problem of typical technologies, in which vast amounts of search results are unnecessarily provided with respect to a search request from a user and the information is required to be filtered to obtain necessary information, are solved. In order to accurately predict information required at a current situation of the user based on the context awareness technology and to provide the corresponding information to the user, a context ontology and a tour ontology are established to accurately predict the information necessary for the user, and relationships between core classes are defined so that information desired by the user is accurately predicted and provided through a semantic data mining algorithm.

Description

For providing the user based on context aware to pay close attention to the context aware body constructing method of information

Technical field

The method that the present invention relates to interest based on context aware (contextawareness) technological prediction user, particularly relates to and aims to provide based on the context aware technology Accurate Prediction information that user needs in current context and provide this information to the user based on context aware of user and pay close attention to the context aware body based on context aware of information service and realize method.

Additionally, the present invention relates to the user based on context aware and information service offer method is provided, wherein, for Accurate Prediction and provide user information needed, build the situation ontologies (contextontology) based on context aware and tourism body (tourontology) and define the relation between core classes, thus semantic data mining algorithm (SemanticDataMiningAlgorithm) Accurate Prediction can be passed through and provides user information needed.

Background technology

Recently, along with the arrival of information age, include that the network of the Internet exists various information, therefore, for accurately judging which type of information is useful information and only selects required information and provide in bulk information, the demand of Information Filtering Technology is continuously increased.

Additionally, recently, except desktop computer, development along with technology such as WiFi, 3G, 4G, GPS, sensors, for instance notebook computer (Notebook), panel computer (TabletPC), PDA (PersonalDigitalAssistant), PCS (PersonalCommunicationsService) and smart mobile phone (SmartPhone) etc. consider that the development of ambulant general fit calculation technology and mobile calculation technique is quite fast.

Meanwhile, along with described trend, information retrieval mode under described mobile environment also changes a lot, in other words, the existing information retrieval and the information transmitting methods that use desktop computer are the information retrieval being primarily adapted for use in the fixed space such as office or family, but now with the development of the technology manufacturing various mobile equipment, the mode of intelligence transmission is changed into the method considering that the information retrieval of mobility and portability maybe can transmit information.

Specifically, such as, according to [Kim2008] (with reference to list of references 1), it is the environment that the various equipment utilization web applications such as a large number of users and desktop computer, office equipment, household appliances interact by pervasive web application Environment Definition, and in described environment, most of users can use the portable sets such as mobile phone to obtain desired service.

Additionally, according to [Michalis2012] (with reference to list of references 2), the result of the information retrieval method being compared with wired mode and the information retrieval method that utilizes wireless mode shows, existing fixing information retrieval method is except user is except the time period of office work or time After Hours, on the whole in the trend reduced, and utilize the information retrieval method of wireless mode except the time that user sleeps, in the trend continued to increase.

As mentioned above, development along with wireless mobile apparatus and the information retrieval method utilizing wireless mobile apparatus, user can effectively retrieve information needed whenever and wherever possible, but when user search to information needed, than the accurate information that user needs, more unwanted information is provided, and this is the general problem existed in the existing method utilizing desktop computer.

Especially, as mentioned above, existing search method not only provides required concern information, accordingly, it would be desirable to filtered retrieval result one by one by user, relatively complicated, and, in order to know the position of affiliated information, also need to retrieve again map etc., till acquisition information needed, identical content need to be repeated retrieval by different classifications.

Therefore, for solving Problems existing in the search method in described existing mobile environment, need exploitation situation ontologies model, described situation ontologies model is in order to require to provide best and personalized information retrieval result for the retrieval of user whenever and wherever possible, based on context aware technology, the attention rate of prediction user, more accurately prediction user where on need which type of retrieval result, described context aware technology utilizes the current context of the perception users such as the current position of user or periphery situation, and provides the information being suitable for it.

nullAdditionally，In order to consider the mobility of the mobile phone users such as smart mobile phone、The restriction etc. of portability and display aspect，More accurately brief retrieval result is provided，To be suitable for the limited picture of mobile phone，Thus needing to develop the semantic data mining algorithm (SemanticDataMiningAlgorithm) of context data，The described semantic data mining algorithm utilizing context data is when user inputs required retrieval request，According to the query pattern that situation ontologies (contextontology) database lookup prestored is similar or consistent with the query pattern of user，And utilize user's query pattern searching system，Provide a user with and retrieve result more quickly and accurately，After the query pattern that the described user's of utilization query pattern searching system and user search are asked compares，Same or similar result is provided.

Especially, preferably, for overcoming the retrieval request for user to show the shortcoming of existing retrieval mode of substantial amounts of unnecessary retrieval result, and the retrieval request for user provides more accurately brief result, need to provide new information retrieval method, while described new information retrieval method provides the information of customization according to the retrieval result perception situation of user, utilize the GPS and map function that are built in the mobile equipment such as smart mobile phone, the information of the position of the current location to user and user's information needed is provided in the lump with retrieval result, so that user can find information needed rapidly and accurately, but also do not propose disclosure satisfy that device or the method for described requirement.

[list of references]

1、KimH.W.,NamJ.Y.,LeeH.,ShinD.M.,"ADeviceCoordinationMechanismwithContextOntologyforUbiquitousWebApplicationEnvironment",JournalofEngineering&Technology,Vol.17,2008.

2、MichalisK.,"AnIntroductiontoAndroid",Tutorial:hy439&hy539,2012.

3、UscholdM.F.,JasperR.J.,"AFrameworkforUnderstandingandClassifyingOntologyApplications",ProceedingsoftheIJCAI-99WorkshoponOntologiesandProblemSolvingMethods(KRR5),pp.1-12,1999.

4、FischerG.,OstwaldJ.L.,"KnowledgeManagement:Problems,Promises,Realities,andChallenges",IEEEIntelligentSystems,Vol.16,No.1,pp.60-72,2001.

5、HwangH.S.,LeeJ.Y.,"AStudyofaKnowledgeInferenceAlgorithmusinganAssociationMiningMethodbasedonOntologies",JournalofKoreaMultimediaSociety,Vol.11,No.11pp.1566-1574,2008.

6、MostefaouiS.K.,HirsbrunnerB.,"TowardsaContext-BasedServiceCompositionFramework",ProceedingsoftheInternationalConferenceonWebServices,pp.42-45,ICWS'03,LasVegas,Nevada,23-26June2003.

7、YilmazN.,AlptekinG.,"AnOntology-BasedDataMiningApproachforStrategicMarketingDecisions",The10thInternationalConferenceonCyberneticsandInformationTechnologies,SystemsandApplications:CITSA2013.

8、KimJ.H.,KimH.C.,"InformationRetrievalSystemforMobileDevices",JournaloftheKoreanSocietyofMarineEngineering,Vol.33,No.4,pp.569-577,2009.

9、ParkH.S.,ParkM.H.,ChoS.B.,"InformationRecommendationinMobileEnvironmentusingaMulti-CriteriaDecisionMaking",JournalofKIISE.ComputingPracticesandLetters,Vol.14,No.3,pp.306-310,2008.

10、SeoH.S.,LeeS.Y.,"UserBehaviorPatternInferenceModelforContext-AwarenessbasedPersonalizationRecommendedServices",Thesis(Master'sdegree),KongjuNationalUniv.GraduateSchool:ComputerScienceDepartment,2012.

11、JeongJ.E.,"OntologybasedPreprocessingSchemeforMiningDataStreamsfromSensorNetworks",IntelligenceInformationResearch,Vol.15,No.3,pp.67-80,2009.

12、KoS.K.,ChoiS.M.,EomH.S.,ChaJ.W.,ChoH.C.,KimL.H,HanY.S.,"ASmartMovieRecommendationSystem",HumanInterfaceandtheManagementof Information.InteractingwithInformationLectureNotesinComputerScience,Vol.6771,pp.558-566,2011.

13、SarkalehM.K.,MahdaviM.,BaniardalanM.,"DesigningaTourismRecommenderSystemBasedonLocation,MobileDeviceandUserFeaturesinMuseum",InternationalJournalofManagingInformationTechnology(IJMIT),Vol.4,No.2,May2012.

14、AbrahamT.,"KnowledgeDiscoveryinSpatio-TemporalDatabases",SchoolofComputerandInformationScience,UniversityofSouthofAustralia,Ph.Ddissertation,1999.

15、GuM.S.,HwangJ.H.,RyuK.H.,"Weigted-FP-TreeBasedXMLQueryPatternMining",6thInternationalConference,ADMA2010,Vol.6440,pp.417-428,2010.11.

16、Chen,L.Bhowmick,S.S.Chia,L.T.,"MiningMaximalFrequentlyChangingSubtreePatternsfromXMLDocuments",InProceedingsofthe6thInternationalConferenceonDataWarehousingandKnowledgeDiscovery(DaWaK),Zaragoza,Spain,2004.

17、Chen,L.Bhowmick,S.S.Chia,L.T.,"FRACTURE-Mining:MiningFrequentlyandConcurrentlyMutatingStructuresfromHistoricalXMLDocuments",ElsevierScienceJournal:Data&KnowledgeEngineeringVolume:59Issue:2pp.320-347,2006.

18、Han,J.Pei,J.Yin,Y.,"MiningFrequentPatternswithoutCandidateGeneration",Proceedingsofthe2000ACMSIGMODinternationalconferenceonManagementofdata,2000.

19、HunyadiD.,"PerformancecomparisonofaprioriandFP-growthalgorithmsingeneratingassociationrules",Proceedingsofthe5thEuropeanconferenceonEuropeancomputingconferencepp.376-381,2011.

20、http://www.android.com/

21、http://www.jetbrains.com/webstorm/

Look-ahead technique document

1, Korean Patent Laid the 10-2013-0097476th (2013.09.03.)

2, Korean registered patent gazette the 10-1009638th (2011.01.13.)

3, Korean registered patent gazette the 10-0868331st (2008.11.05.)

4, Korean registered patent gazette the 10-0688087th (2007.02.22.)

Summary of the invention

nullThe present invention proposes to solve existing issue as above，Thus it is an object of the invention to provide a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method，Wherein，For the problem solving the retrieval mode of prior art，Attention rate based on context aware technological prediction user，Build more Accurate Prediction user where on which type of needs retrieve situation ontologies (contextontology) model，And by the concern information according to described situation ontologies model prediction user，Retrieval request for user provides only relevant information of paying close attention to user，Thus more accurately brief retrieval result can be provided，The retrieval mode of wherein said prior art has the disadvantage that，Retrieval request for user provides substantial amounts of unnecessary retrieval result，Thus forcing user again to filter information needed.

In addition, further object is that, there is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, the defect of limited picture and few internal memory in order to overcome the mobile phones such as smart mobile phone, available semantic data mining algorithm (SemanticDataMiningAlgorithm) provides more accurately brief retrieval result, and described semantic data mining algorithm utilizes situation ontologies.

Another purpose again of the present invention is in that, there is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, in order to only be prone to find the content of information needed by primary retrieval, but also position can be found, while providing for the retrieval result of the retrieval request of user, utilize and be built in GPS and the map function of mobile equipment, the position of retrieval result needed for the current location of user and user is provided in the lump with retrieval result.

nullFor reaching described purpose，According to the present invention，A kind of context aware body is provided to realize method，Described context aware body (ContextAwarenessOntology) method of realization realizes retrieval service，Wherein，For the problem solving the searching system of prior art，Only the concern information of described user is provided as retrieval result based on context aware (ContextAwareness) technology，Thus more accurately brief retrieval result can be provided than existing searching system，The shortcoming of the searching system of described prior art is in that the retrieval request for user because unnecessary information being also served as retrieval result to user and providing，So that user again filters information needed and selects，Described context aware body realizes method and includes: user pays close attention to information database construction step，It collects the information of the region-of-interest for described user，Thus building data base；Body inference step, its relation to being stored in described user and pay close attention between each data of information database and attribute are configured, and pay close attention to described user and build described user in information database construction step and pay close attention to information database；And ontology database construction step, it builds ontology database according to relation and the attribute of each data arranged in described body inference step.

At this, the contextual information of described user is expressed as including user identifier U (Useridentification), spatial information L (Location), temporal information T (Time) and the time space contextual information O={U with the four-dimension with described user-dependent information on services S (Service) by described body, L, T, S}.

Additionally, described temporal information is the information divided by the certain time spacing pre-set, described spatial information is the general positional information of belonging euclidean (euclidean) coordinate (xi, yi) spatially, and described coordinate represents longitude and the latitude value in area that described user is presently in respectively.

Meanwhile, in described database sharing step, concern information as described user, the information to ground of going sightseeing is collected by each area, as most higher level's class designated root class (rootclass), each area is appointed as subclass (subclass), by to described regional go sightseeing ground title be appointed as the subclass to each area described, it is appointed as the subclass to described ground title of going sightseeing by the facility in each sightseeing ground or to the affiliated detailed description going sightseeing ground, described each department or described sightseeing are arranged to synonym or arrange synonym part of speech (equivalentclass) relation of consonant retrieval, thus building the sightseeing data base constituting hierarchy.

At this, described sightseeing data base includes multiple territory (field), (field) includes id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, long in the plurality of territory, the address on each ground of going sightseeing of described add_1 territory, described add_2 territory, described add_3 territory, described add_4 territory and described adde_5 domain representation, the storage of described tour_site territory has sightseeing ground title, described syn is that described lat territory and described long territory represent latitude required when arranging map and longitude respectively for processing synonym or the territory of consonant word.

Especially, described body inference step is according to " Place " that be made up of the title in the area to go sightseeing, " Attraction " being made up of the sight-seeing resort title in each area, by the building in each sightseeing ground or remains, traces, " Resourc " and " Activity " that seashore and restaurant are constituted determines the attribute of each data, described " Activity " includes " Culturalactivity " that represent cultural activity, represent " Sceneryview " of the behavior viewing and admiring landscape, represent " Diningout " of the behavior dined out, thus representing the practical action that described user implements on each ground of going sightseeing.

At this, in described body inference step, the relation on attributes of described " Place " and described " Attraction " is appointed as " hasAttraction " and " IsAttractionOf ", the relation on attributes of described " Attraction " and described " Resource " is appointed as " hasResource " and " IsResourceOf ", and the relation on attributes of described " Activity " and described " Resource " is appointed as " hasActivity " and " IsActivityOf ".

It addition, in described body inference step, when there is multiple relation on attributes in the data constituting each class, be considered as that there is multiple attributes relation.

Meanwhile, in described body inference step, Protege_4.0.2version is utilized to perform.

Especially, described ontology database construction step performs following process: utilize OWL (network ontology language, WebOntologyLanguage), RDF (resource description framework, ResourceDescriptionFramework) and RDFS (resource description framework framework specification, ResourceDescriptionFrameworkSchemaSpecification) build described ontology database.

Additionally, according to the present invention, there is provided a kind of user based on context aware to pay close attention to information service and method is provided, it utilizes and realizes, by described described context aware body, the context aware body that method builds, the concern information of user is provided only as retrieval result, thus providing more accurately brief retrieval result than existing search method, therefore, it is suitable including through the mobile environment of smart mobile phone or panel computer.

Meanwhile, according to the present invention, a kind of user based on context aware is provided to pay close attention to information service system, it utilizes and realizes, by described described context aware body, the context aware body that method builds, the concern information of user is provided only as retrieval result, thus providing more accurately brief retrieval result than existing searching system, therefore, it is suitable including through the mobile environment of smart mobile phone or panel computer.

As mentioned above, according to the present invention, there is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, attention rate based on context aware technological prediction user, and build for more Accurate Prediction user at the situation ontologies model where which type of needing retrieve, and by the concern information according to described situation ontologies model prediction user, retrieval request for user provides only the information that user pays close attention to, thus more accurately brief retrieval result can be provided, therefore, it is possible to the problem solving the retrieval mode of prior art, the shortcoming of the retrieval mode of described prior art is in that, retrieval request for user provides substantial amounts of unnecessary retrieval result, so that user filters information needed again.

Additionally, according to the present invention, there is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, can pass through to utilize the semantic data mining algorithm (SemanticDataMiningAlgorithm) of situation ontologies to provide more accurately brief retrieval result such that it is able to overcome the limited picture of the mobile phones such as smart mobile phone and the defect of few internal memory.

Meanwhile, according to the present invention, there is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, while providing for the retrieval result of the retrieval request information of user, utilize and be built in GPS and the map function of mobile equipment, provide the position of the retrieval result needed for the current location of user and user in the lump with retrieving result, thus only can not only be prone to find the content of information needed by primary retrieval, additionally it is possible to find position.

Summary of drawings

Fig. 1 represents by the schematic diagram that the table feature to existing data digging method and pluses and minuses compare；

Fig. 2 represents specific sightseeing be appointed as geographic coverage and build the flow chart of process of situation ontologies；

Fig. 3 is for the context aware data construct sightseeing body utilizing the ground of going sightseeing being positioned at each area in order to provide the user with the service to ground of going sightseeing, carrying out the process schematic of inference so that this provides the user with the service to ground of going sightseeing；

Fig. 4 is the generalized schematic of the part-structure in the hierarchy of expression sightseeing body shown in Fig. 3；

Fig. 5 is the schematic diagram for the property value of data value constituting each class to carry out body inference to arrange and the process of relation are illustrated；

Fig. 6 is the schematic diagram representing the multiple attributes relation between the data constituting each class；

Fig. 7 utilizes the protege process generalized schematic carrying out body inference；

Fig. 8 is the partial schematic diagram representing the sightseeing data base built according to embodiments of the present invention；

Fig. 9 is the partial schematic diagram representing the sightseeing database table built according to embodiments of the present invention；

Figure 10 is the schematic diagram representing sightseeing body construction according to embodiments of the present invention；

Figure 11 is the schematic diagram of the body construction representing the add_2class launching the sightseeing body shown in Figure 10；

Figure 12 is the schematic diagram of the body construction representing the add_3class launching the sightseeing body shown in Figure 10；

Figure 13 is the schematic diagram of the body construction representing the add_4class launching the sightseeing body shown in Figure 10；

Figure 14 represents the schematic diagram of part of tour.owl body source code (ontologysource) utilizing OWL (WebOntologyLanguage), RDF and RDFS to build for utilizing protege_4.0.2 according to embodiments of the present invention；

Figure 15 is for representing the integrally-built generalized schematic for providing the context aware body that the user based on context aware pays close attention to information service to realize method according to embodiments of the present invention；

Figure 16 is the schematic diagram of expression XML query scheme-tree (Querypatterntree) respectively and the XML query subtree (QueryRootedsubtree) with root；

Figure 17 represents query pattern tree respectively and has the structural representation of the subtree that takes place frequently of root；

The generalized schematic of Figure 18 process with the process of each affairs (Transaction) for representing the prefix (prefix) of shared frequent pattern tree (fp tree) (FP-Tree) building process；

Figure 19 carries out, in the process of data mining, building the schematic diagram of the process of condition (conditional) frequent pattern tree (fp tree) (FP-Tree) of the m as any project (item) at execution FP-growth algorithm；

Figure 20 is the result of the process repeatedly performing search criterion pattern, by the schematic diagram that the condition pattern obtained is arranged by table；

Figure 21 is the exemplary plot that the Korean inquiry that user performs is made as XML form；

Figure 22 is the schematic diagram being denoted as the BookDTDTree based on the XQuery XML document made；

Figure 23 is the schematic diagram of the frequent item set (transactionfrequentitemset) representing each affairs in sightseeing ontology database；

Figure 24 is the partial schematic schematic diagram of frequent pattern tree (fp tree) (FP-tree) structure utilizing the sightseeing ontology database with the structure that takes place frequently by each department generation；

Figure 25 builds the situation schematic diagram of condition (Conditional) FP-tree for representing respectively with the Ma Luolu in the whole world (Global) FP-tree and sightseeing area for benchmark；

Figure 26 is the schematic diagram of the part of original code representing FP-growth algorithm；

Figure 27 is that to be suitable for by minimum support in FP-growth algorithm be the schematic diagram of 0.4 result carrying out data mining；

Figure 28 is the overall structure generalized schematic of the information retrieval system based on movement that the method for mode matching utilizing semantic data mining algorithm is applied；

The generalized schematic of performed a series of processing procedures when Figure 29 is the retrieval request receiving ground of going sightseeing from user；

Figure 30 is for illustrating when receiving retrieval request from user, by analyzing the requirement item of user and obtaining the query pattern of user, and by the mode configuration with the consistent pattern of query pattern analyzed and the ontology database that is stored in searching system is compared, retrieve the schematic diagram of the process of similar or identical mode configuration；

Figure 31 is the integrally-built generalized schematic that the user utilizing semantic data to excavate according to embodiments of the present invention pays close attention to information retrieval method；

Figure 32 is the schematic diagram of the part of original code of the initial picture for realizing retrieval window；

Figure 33 for realizing the source code schematic diagram of the event-action (eventaction) to index button by script function (Scriptfunction)；

Figure 34 is for performing the schematic diagram of the part of original code of the map button behavior (mapbuttonaction) for providing Map Services；

Figure 35 is the schematic diagram of the composition of the key frame of mobile message searching system according to embodiments of the present invention；

Figure 36 represents the schematic diagram to the composition of the map in area belonging to display when retrieving the ground zones values in window input wide area city, land for growing field crops and click map button and the composition according to the picture introducing relevant regional restaurant or area of going sightseeing to the input value display retrieving window input respectively；

Figure 37 realizes by the present position values of latitude and longitude automatic sensing user thus show that the map of the map of regional map that user is presently in and latitude and longitude realizes the schematic diagram of source code simultaneously；

Figure 38 is the schematic diagram of the result screen realized by source code as shown in figure 37；

Figure 39 is the schematic diagram that interpretation is table performing data mining process by being suitable for FP-growth algorithm in ontology database according to embodiments of the present invention；

Figure 40 will be suitable for FP-growth algorithm and perform the schematic diagram that interpretation is table of data mining process in existing relevant database；

Figure 41 is that the internal memory that the change by the minimum support marginal value along with ontology data and existing relevant database according to embodiments of the present invention changes makes the change procedure of consumption be expressed as the schematic diagram of chart；

Figure 42 is the schematic diagram that the number change of the frequent mode that the change by the minimum support value along with ontology data and existing relevant database according to embodiments of the present invention changes is expressed as chart；

Figure 43 is the schematic diagram that the change performing the time changed along with the change of ontology data and the minimum support of existing relevant database is expressed as chart according to embodiments of the present invention；

Figure 44 is the schematic diagram that the change performing the time changed along with the change of ontology data and the affairs of existing relevant database is expressed as chart according to embodiments of the present invention；

Figure 45 be according to embodiments of the present invention make the change of consumption compare and be expressed as the schematic diagram of chart the internal memory changed along with the change of ontology database and the affairs of existing relevant database；

Figure 46 is the schematic diagram that the change of the accuracy representing recall precision along with the change of the affairs as user's query pattern is expressed as chart；

Figure 47 is when the schematic diagram of the picture of the retrieval result shown when retrieving window input " chicken Longshan " in existing portal；

Figure 48 represents in mobile message searching system according to embodiments of the present invention, the schematic diagram of the picture of the retrieval result shown when retrieving window input " chicken Longshan "；

Figure 49 represents in mobile message searching system according to embodiments of the present invention, shows the schematic diagram of the picture of the map of the detailed description relevant to retrieval result and surrounding area during input search key.

Detailed description of the invention

Below, in conjunction with accompanying drawing to the present invention for providing the specific embodiment that the context aware body that the user based on context aware pays close attention to information service realizes method to illustrate.

At this, content described below is a kind of embodiment for implementing the present invention, should notice the present invention and be not limited to embodiment content as described below.

It addition, below to, in the explanation of the embodiment of the present invention, illustrating for simplifying, same or like with the content of prior art, or the part that skilled addressee readily understands that and implement, will omit its detailed description carried out.

Words sentence is talked about, as described below, the present invention relates to a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, for the problem solving the retrieval mode of prior art, attention rate based on context aware technological prediction user, build for more Accurate Prediction user at situation ontologies (contextontology) model where which type of needing retrieve, and by the concern information according to described situation ontologies model prediction user, retrieval request for user provides only relevant information of paying close attention to user, thus more accurately brief retrieval result can be provided, the retrieval mode of described prior art has the disadvantage that the retrieval request for user provides substantial amounts of unnecessary retrieval result, so that user filters information needed again.

Meanwhile, the present invention relates to a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, the defect of limited picture and few internal memory in order to overcome the mobile phones such as smart mobile phone, available semantic data mining algorithm (SemanticDataMiningAlgorithm) provides more accurately brief retrieval result, and described semantic data mining algorithm utilizes situation ontologies.

Especially, the present invention relates to a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method, wherein, in order to only be prone to find the content of information needed by primary retrieval, position can also be found, while providing for the retrieval result of the retrieval request information of user, utilize and be built in GPS and the map function of mobile equipment, provide the position of the retrieval result needed for the current location of user and user in the lump with retrieving result.

At this, to according to the present invention for before providing the specific embodiment that the context aware body that the user based on context aware pays close attention to information service realizes method to illustrate, the basic conception suitable in the body (Ontology) of the present invention, context aware (ContextAwareness) and data mining algorithm (DataMiningAlgorithm) being illustrated.

nullFirst，Body (Ontology) refers to and the information in particular range is carried out conceptualization，With logical structure classification or defined notion，Thus the method represented with hierarchy form，In the prior art，According to [Uschold1999]、[Fischer2001] and [Hwang2008] etc.，Body is the data presentation technique passing through the scattered data such as network and supplier for analytical integration，Standardization and clear and definite detail it is defined as relative to shared conceptualization，Generally speaking，Propose RDFS (the resource description framework framework specification as the key element for representing body to build body，ResourceDescriptionFrameworkSchemaSpecification)、RDF (resource description framework，ResourceDescriptionFramework)、OWL (network ontology language,OntologyWebLanguage) technology contents (with reference to list of references 3 to list of references 5).

In other words, RDFS is as the most basic language representing body, and definition is for using the basic vocabulary of RDF applicational language, and RDF is as the language that can represent the information of all kinds and the useful of metadata in data model, is made up of resource and sentence.

Specifically, RDF sentence is by subject (subject), narration (predicate (property)), purpose (object (value)) is constituted, and OWL (network ontology language, WebOntologyLanguage) it is according to the W3C standard body language proposed for definition body in the semantic network of making based on the Ontology Language by the initial stage as RDFS, for representing the hierarchy of body, the levels level structure between object class and classification can be represented, can any attribute defined by the user attribute of definition is distributed to classification, therefore, than RDFS more standard, and pass between objects fastens and can carry out logic and analogize, equally, by RDF sentence expression information.

At this, the embodiments of the invention that will illustrate below employ described RDFS, RDF, OWL etc. also for building sightseeing ontology database.

Then, the conceptual illustration of context aware (ContextAwareness) is as follows: contextual information can be the personal information as the current active of user, can be the technical information as the equipment in currently used, it is also possible to be the environmental information as temperature, position or time.

In other words, context aware individualizes and is the collection by described contextual information and exchange carries out perception, and provide a user with, by processing procedures such as explanation and inferences, the customization service being suitable for situation, for this, the various contextual information of periphery need to obtain suitable modeling, collection and distribution.

Therefore, in order to provide context-aware services, the contextual information collected need to be combined with user profile and analyze and inference becomes to show the high-level information of an aspect of life, as it has been described above, storage has the important technology that the method that suitable information system are shared effectively is the success or not of decision context aware system.

Meanwhile, recently, increase along with the user using mobile equipment, the demand of the application program and service that are suitable for fast-changing environment is continuously increased, but in order to overcome the problem being difficult to so far build the context aware system meeting described requirement, as the method being used for semantic situation performance, situation inference, knowledge sharing, context classification, it is proposed that utilize the environmental model based on body (contextmodel) of OWL.

Specifically, in the prior art, for providing the context-aware services that can carry out various situation inference, propose SOCAM (service-oriented occasion sensing middleware, Service-OrientedContext-AwareMiddleware), in other words, such as according to [Mostefaoui2003], it is proposed that in conjunction with the CB-Sec (Services Composition based on situation of Context-aware computing (context-awarecomputing) and agency's (agent) technology, Context-BasedServiceComposition) framework (framework) (with reference to list of references 6).

Thus, it comprises history (History), thus overcome in major part context aware application program contextual parameter (contextparameter) only assert position (location) and ignore the time (temporal) in shortcoming, in existing context aware, include following 5 kinds of key elements: user context (usercontext), it includes the content etc. of the effect of user, identity, age, position, concern；Calculating situation (Computingcontext), it includes network and connects the printer with periphery or display, work station etc.；Time situation (Timecontext), it includes the times such as day, week, the moon, season；Physics situation (Physicalcontext), it includes weather, temperature etc.；And situation history (contexthistory), it is by user context, calculating situation and physics situation record certain time, thus can be effectively applied to constitute in service.

Then, data digging method is illustrated as follows: in order to be suitable for semantic data method for digging, existing text mining method can also be used, but in order to utilize the body being called sightseeing ontology database being applicable to the embodiment of the present invention being described below to perform data mining, at this, the data digging method of the XML document about the basic structure constituting body is illustrated.

In other words, it is represent by the schematic diagram that the table feature to existing data digging method and pluses and minuses compare with reference to Fig. 1, Fig. 1.

As it is shown in figure 1, in the prior art, have studied the data digging method to various XML data and the various methods to XML query mode excavation, but each method exists pluses and minuses.

Additionally, according to [Hwang2008], in present searching system, too much retrieval result is provided because huge data exist or is not provided that the situation of retrieval result, therefore, for making up the defect of described searching system, it is proposed that SWRL inference language will be shown as by the relevant knowledge of OWL body restriction condition and relevant method for digging institute inference, find new knowledge thereby through jess engine and support the algorithm of effective retrieval, in other words, build the individual according to territory (domain) and pay close attention to body, for individual's focused data, relevant and inference information process (with reference to list of references 5) is provided when illustrating based on relevant and body retrieval information.

Meanwhile, according to [Yilmaz2013], propose to utilize the data digging method based on body of Apriori algorithm, in other words, for determining the operating system of consumers, framework based on data digging method is proposed, after building body and carrying out questionnaire survey to user and measure preference, transmit data, analyze cluster and use dependency rule algorithm to analyze result, now, protege is used for building body, and Apriori algorithm performs following two steps: search the project set that takes place frequently, and described project set is divided into precedent (antecedent) and result (consequent) two kinds, thus searching the rule (with reference to list of references 7) that can meet credibility (confidence).

Then, the information retrieval system of mobile environment is illustrated as follows: recently, development along with prevailing technology, as the research to the context aware technology providing a user with useful service, the situation of perception user the context aware recommendation service of information that is provided with are studied, but the recommendation information based on situation is only recommended as the information of consistent concept level by existing context aware recommendation service, therefore, when the information recommended is not suitable for the current context of user, user need to directly search information needed in the information recommended, thus needing to expend a lot of effort of user and time.

Meanwhile, recently, development along with mobile internet environment, the Internal retrieval utilizing terminating machine comes to life, and in order to effectively input search key, it is proposed that the method considering the hardware capability aspects such as keyboard and the method considering to utilize the software function aspect of the key word etc. of the collection of user, storage.

At this, ontology model can use the purposes for the various contextual information for storing and manage the mobile equipment such as GPS, illuminance transducer, user profile, and the equipment of movement obtains the contextual information of the various kinds to the position of user, environment etc. according to the kind of the hardware sensor being included in equipment, and it is assigned as ontology model, understand the implication between contextual information, thus being processed into the information of required form or for situation inference.

It addition, mobile environment has the feature different from general cable environment, in other words, first, mobile environment by the restriction in time and space, therefore, user carries mobile equipment and retrieves information needed whenever and wherever possible；The second, mobile environment is very individualized, and most mobile equipment is person ownership, different according to the information needed for the interest of individual and region-of-interest, therefore, feature according to individual, even if content identical on surface, the content actually inquired about is likely different；3rd, because mobile equipment is all small-sized, so there is a lot of restrictions at input, output facet, therefore, in mobile environment, showing a lot of page than by little picture, the information needed for only extracting user displays more effectively；4th it is said that in general, the data transmission rate of mobile environment is much lower than cable environment, but price is higher；5th it is said that in general, there is the restriction of internal memory and disposal ability in mobile equipment, therefore, it is impossible to process substantial amounts of information at mobile equipment itself.

Therefore, it is necessary to consider that described mobile environment characteristic offer are suitable for the information retrieval system of mobile environment.

In other words, according to [Kim2009], consider the feature of described mobile environment, propose the mobile message retrieval model that document summarization and user profile (userprofile) are applicable to unit's retrieval (metasearch) model, described document summarization technology can alleviate the problem of the low transfer rate in little picture, and user profile can use personalized information retrieval model (with reference to list of references 8) in mobile environment.

Meanwhile, the preference for information recommendation service is different according to situation, and in order to provide information recommendation service, first has to know the contextual information of user.

In other words, according to [Park2008], suggestion considers the commending system of the preference of most users in mobile environment and is applicable to the recommendation in restaurant, and in order to the preference of individual user being monitored in mobile environment, employ Bayesian network, and restaurant recommends in most of the cases obtain multiple preference based on the preference of the most users of non-individual user, use the AHP (with reference to list of references 9) as multi-standard decision method.

Recently, along with major part user has mobile equipment, mobile equipment not only acts as the channeling that can contact personal information, and, because user takes all the time and takes with oneself, so may also operate as the effect of the instrument of the environmental information that can collect user, and the information so collected can effectively utilize at the information recommendation for user, therefore, the inference of mobile environment is also the important field of research relevant to information recommendation.

Meanwhile, in order to provide personalized recommendation service in environment sensing environment, need to quickly analyze the purpose of collected environmental information valid inference user, but there is the difference of data according to different situation environment in the information collected by mobile equipment, therefore, be not suitable for directly being suitable for existing inference algorithm, and need to be suitable for the effective algorithm of mobile environment.

In other words, according to [Seo2012], in order to reduce the loss because the omission or mistake etc. of information cause to greatest extent, utilize Naive Bayes Classifier that action model is classified, and employ method for mode matching (with reference to list of references 10) to effectively learn the tendency of user inference action purpose, additionally, according to [Jeong2009], describe the exploitation along with various sensors and the structure of sensor network, for collecting the environmental data of particular space and finding more useful information and knowledge and utilize the research (with reference to list of references 11) of data mining (Datamining) method.

nullMeanwhile，According to [Ko2011]，Propose the commending system based on type of relationship (Genrecorrelation)，To utilize collaborative filtering method (Collaborativefilteringmethod) to collect preference information from a large number of users，So that the automatic Prediction for the concern item of user is likely to (with reference to list of references 12)，And [Sarkaleh2012] proposes following commending system: because such as notebook computer、Mobile phone、PCS (PCS Personal Communications System，personalcommunicationsystem)、PDA (palm PC，Personaldigitalassistant) etc. the same mobile equipment has mobility、Independence、Individual、The features such as accessibility，Hence with described equipment, the study in correct time and place is individualized，And also make learning content rich and varied，And utilize GPS (system of geographic location，GeographicPositionSystem) function can obtain the information (with reference to list of references 13) in the various regions needed for user.

As it has been described above, in the prior art, it is proposed that the various methods relevant to the concern information recommendation system for providing only information that is necessary and that pay close attention to user, but in nearest mobile environment, there are the following problems for existing information retrieval method.

Specifically, when comparing existing data retrieval mode and being commonly used to the data retrieval mode of current moving area, can find that maximum difference is mobility, in other words, existing data retrieval method is because major part uses desktop computer, so being fixed on the three unities and achieving all retrievals and data and be suitable for, but when considering mobility and the activeness of society, the efficiency of described existing data retrieval mode is at a fairly low, and can as the mode of the most advanced technology lagging behind society.

Therefore, for the shortcoming overcoming described existing information retrieval mode, the demand of the information retrieval method being suitable for mobile environment is continuously increased, additionally, because existing information retrieval mode is left out mobility, so cannot build it can be considered that the system of current location of user, but the development along with recent mobile technology, can the context aware method that utilize the current location of GPS function search user be combined in information retrieval system, therefore, there is provided following service in the present invention: utilize the current location that user performs inquiry to transmit accurate information by described technology.

In other words, for instance, in order to provide a user with the service of information transmission more accurately based on context aware, utilizing context aware body to collect the sightseeing data message in each area, thus building sightseeing body, and the information service being suitable for user's current context can be provided.

Meanwhile, when existing information retrieval system inputting key word and obtain retrieval result, than the desired information accurately of user, more unrelated items occurs, thus the problem that unnecessary huge retrieval result is provided, therefore, in the present invention, context aware body of data is constructed in order to solve described problem, and service more quickly and accurately in order to the inquiry of user is provided, employ semantic data method for digging.

In other words, such as, build sightseeing body and store the query pattern that user often inquires about, when user inquires about, the matching process of the query pattern utilized and store and user's inquiry, provide a user with and inquire about identical or similar query pattern with user, thus providing only the information that user is actually needed.

Especially, existing information retrieval system provides only text message, therefore, the present invention provides Map Services (Mapservice), it utilizes the Map Services such as Google Maps (Googlemap) that such as smart mobile phone supports at interior mobile equipment, only can quickly and easily be found the particular geographic location information of information needed for user even by primary retrieval.

Then, for solving described problem of the prior art, the detailed process building the body utilizing context aware information is described in detail.

First, the basic conception of context aware body (ContextAwarenessOntology) is illustrated as follows: the mode position that time space Move Mode is the object by movement realizes individualized according to the position characteristic of client, and make suitable interior perhaps service provide the time space rule becoming possible, and the cause effect relation (causalrelationship) that described time space relation is to finding between the event relevant to time space object has and has very important significance, therefore, can pass through to consider that the action of user can be provided suitable service by the body in time and space simultaneously.

Thus, in the present invention, it is designed for use with being beneficial to structuring contextual information and the body of information of mutual relation and part situation can be represented, and the body according to the described embodiment of the present invention is the information that the contextual information with the user as time and spatial information and service etc. is relevant, it is expressed as O={U, L, T, S}.

At this, U (Useridentification) is user identifier, and L (Location) is spatial information, and T (Time) is temporal information, and S (Service) represents and user-dependent service regulation.

In other words, the time space contextual information of user U includes the four-dimensional information such as the identification information of user, temporal information, spatial information and service information.

Specifically, time space information for identifying the room and time of user indicates that the contextual information of user, and for distinguishing the benchmark of the service log according to different situations, and the positional information of user is made up of the vague generalization value of the coordinate spatially of the effective time that actual event is occurred.

Now, temporal information is the information divided by certain time spacing, and spatial information is the general positional information of the coordinate (xi, yi) on belonging Euclidean space.

It addition, the coordinate figure in space is carried out vague generalization by certain area (zone) by the vague generalization of position, thus identifying the position of user and object, at this, space coordinates represents longitude and the latitude value in the area being presently in respectively.

Such as, if user is currently located at Semen Ginkgo hole, wide area Shizhong District, land for growing field crops in the street, then the space coordinates of user is generalized to the zone locations value of the position in Semen Ginkgo hole, wide area Shizhong District, belonging land for growing field crops, thus the latitude value of the current location of available user and longitude.

Then, the process of the body inference process needed for building ontology database and structure sightseeing body is illustrated.

First, it is represent specific sightseeing be appointed as geographic coverage and build the flow chart of process of situation ontologies with reference to Fig. 2, Fig. 2.

As in figure 2 it is shown, the process building situation ontologies is, first, the territory (S21 step) of the body to build is determined for building body.

In other words, for instance, when providing the information to ground of going sightseeing, search the sightseeing ground information needed for user, thus collecting the information by the ground of going sightseeing belonging to each area.

Then, constituting the class (S22 step) to each key concept, now, each class is constituted with the hierarchy being made up of superclass (superclass) and subclass (subclass) etc..

In other words, such as, the root class (rootclass) being called tour_sites is specified as most higher level's class, in its lower section each area is appointed as subclass (subclass) and generates lower floor's class, and each area is arranged more accurate subclass relation, thus build the sightseeing body constituting hierarchical structure.

Then, each data of attribute definition utilizing the data value constituting each class have which kind of relation and which kind of feature (S23 step), in other words, such as, the characteristic in the area of each class will be constituted or be included in the famous-eat in affiliated area or illustrate that the content on sightseeing ground is appointed as the relation on attributes of each class.

Then, the characteristic specified attribute value utilizing this sightseeing ground to have the ground title of going sightseeing of each class belonging, it is called sightseeing ground example (instances) (S24 step) thus generating.

Then, perform concordance (consistency) and check, its check newly constructed sightseeing body whether with existing body on axiom or attribute etc. consistent (S25 step), the body so built is used for the information retrieval of user.

Therefore, by performing described process, when user retrieves required sightseeing ground in current location, in order to recommend sightseeing ground suitable on the position required by user, by classification sightseeing ground, different regions, thus building sightseeing body.

Then, the process of all kinds of attributes and the relation building described sightseeing body and inference sightseeing data base it is illustrated as.

Generally speaking, user wishes the impact not being subject to oneself location or surrounding enviroment, any position or any environment obtain best service, now, service is the unit describing and realizing hierarchical structuring after the information to available function as equipment, position, execution condition etc. virtualizes.

Additionally, if needing the service regulation set stored is absent from the service that composite users requires, then claim new service or combine existing service and be provided, and in order to provide a user with such service, need to intrinsic each place and the sightseeing being positioned at each place and the service available content such as the restaurant on each sightseeing ground, shopping center, seashore, boulevard carry out inference, to specify intrinsic all kinds of specified attribute.

Additionally, if the existing service regulation set stored being absent from be suitable for the service that user requires, then need generate new service or combine existing service and provide, in order to provide a user with described service, to providing each place in the body and be positioned at the sightseeing ground in each place, and the content that restaurant shopping center, seashore, the mountain-climbing alive etc. that are positioned at each ground of going sightseeing service, the attribute of each class intrinsic need to be specified by inference.

In order to the inference process of described sightseeing body is described, first, it is an illustration for the schematic diagram of following process with reference to Fig. 3, Fig. 3: in order to provide a user with the service to ground of going sightseeing, utilizes the context aware data construct sightseeing body on the ground of going sightseeing being positioned at each area and inference.

In figure 3, " Place " represents the area name to go sightseeing, in other words, such as, road (Jejudo) as autonomous especially in Jizhou, special city, Soul (Seoul), wide area city, land for growing field crops (Daejon), wide area city of Pusan (Busan), South Chungchong, North Chungchong etc., it is made up of area name.

Additionally, " Attraction " represents the sightseeing ground title in each department, in other words, such as, it is made up of the sight name as Rijksmuseum (Nationalmuseum), Jing Fuguang (Palace), sea The Cloud Terrace (Beach), chicken Longshan, shopping center (Shoppingmall) etc..

Meanwhile, " Resource " represents the building in each sightseeing ground, remains, traces and seashore etc., in other words, such as, it is made up of scape good fortune palace (Ancientremains) in utero or the local historical building (Oldbuilding) as folk customs village, sea The Cloud Terrace or the seashore on cattle island, Jizhou, the restaurant (Restaurant) etc. in department store.

Especially, " Activity " represent user each go sightseeing ground implement practical action, in other words, such as " Culturalactivity " represents the cultural activity accessing scape good fortune palace in utero, " Sceneryview " represents that Niu Dao and the Han Shan etc. at sea The Cloud Terrace or Jizhou Island views and admires the behavior of beautiful scenery with going sightseeing, and " Diningout " represents the behavior etc. that the restaurant within department store dines out.

It addition, in embodiments of the present invention, use protege_4.0.2 for carrying out ontological construction and inference, and, utilize OWL (WebOntologyLanguage) and RDF, RDFS etc. to build the tour.owl as sightseeing body.

Then, reference Fig. 4, Fig. 4 is the generalized schematic of the part-structure in the hierarchical structure of the sightseeing body shown in Fig. 3.

As shown in Figure 4, body according to embodiments of the present invention, in order to build sightseeing body, subclass (subclass) is specified by each area, and specify the class (class) of the less group belonging to described area, ground title of going sightseeing is specified in its lower section as subclass, and in order to restaurant in each sightseeing ground or described sightseeing ground are described, it is intended that another subclass.

It addition, in order to each area or sightseeing are carried out synonym or consonant retrieval, be provided with synonym part of speech (equivalentclass) relation representing that synonym processes.

In other words, when user's input is to the key word on ground of going sightseeing, tackle retrieval request or the situation with the retrieval of abbreviation form of carrying out with similar word or synonym, synonym part of speech (equivalentclass) relation is set, information retrieval result accurately is provided all the time to user.

Specifically, as shown in Figure 4, " Tour " that represent ground of going sightseeing is specified as highest class, below sightseeing ground, each areas such as wide area city, land for growing field crops, South Chungchong, special city, Soul, wide area city of Pusan, Jizhou especially autonomous road are appointed as subclass, are made up of the subclass constituting described area below each area.

In other words, by East below wide area city, land for growing field crops, West, middle district, scholar city, great virtue district is constituted, when autonomous road especially, Jizhou, subclass is constituted by Jizhou city and XIGUI Pu city, when South Chungchong, constitute the subclass of public state and Fuyu County etc., subclass below specifies detailed address value, such as, when wide area city, land for growing field crops, in middle district below by Semen Ginkgo hole, Sha Ting hole etc. is configured to subclass, when West, by Chang'an road and Wan Nian hole etc., when autonomous road especially, Jizhou, below the city of Jizhou, constitute three apprentice 2 holes, Jiu Zuo city, the subclass in Chinese hole etc., and below the city of XIGUI Pu, by Cheng Shanyi, pacify to obtain the compositions such as face.

Meanwhile, subclass below is made up of ground title of going sightseeing, in other words, when wide area city, land for growing field crops, it is made up of subclass middle district Sha Ting hole O-World etc., and when autonomous road especially, Jizhou, the Jizhou especially lofty or bottomless cave of autonomous road Jiu Zuo city of Jizhou city etc. constitutes subclass relation.

Then, with reference to Fig. 5, Fig. 5 it is the schematic diagram that the property value of data value constituting each class to carry out body inference to arrange and the process of relation are illustrated.

nullAs shown in Figure 5，The relation in each area (Place) and ground (Attraction) of going sightseeing of constituting sightseeing body has " hasAttraction " and the relation on attributes of " IsAttractionOf "，Go sightseeing ground and sightseeing ground in such as the buildings such as palace or remains、Seashore、" Resource " equally such as mountain has " hasResource " and the relation on attributes of " IsResourceOf "，And user is in go sight-seeing scape Fu Gong palace behavior with going sightseeing、In the behavior taken a walk in The Cloud Terrace seashore, sea、Behavior in the mountain-climbing of chicken Longshan and the behavior etc. to department store shopping are " Activity "，With the relation of " Resource ", there is " hasActivity " and the relation on attributes of " IsActivityOf ".

At this, each relation on attributes there is also the situation of multiple attributes and the relation having between the data constituting each class.

In other words, with reference to Fig. 6, Fig. 6 it is the schematic diagram representing the multiple attributes relation between the data constituting each class.

As shown in Figure 6, such as " Soul (Seoul) " area and with Soul in there is " Zhong Lu (Jongro) " area of inclusion relation there is " hasSubclass " and the relation on attributes of " IsSuperclassOf ", and comprise " Jing Fugong (Kyongbokgung) " and " Ren Si hole (Insadong) " two areas in " Zhong Lu " area, thus there is subclass relation simultaneously.

Additionally, while the relation on " Zhong Lu " area and ground " Jing Fugong " of going sightseeing has the relation of " hasAttraction " and " isAttractionOf ", " Zhong Lu " area and ground " Ren Si hole " of going sightseeing also have " hasAttraction " and the relation of " isAttractionOf ".

Therefore, described relation is referred to as multiple attributes relation, in each sightseeing ground, when having multiple sightseeing ground in identical area, is considered as having described multiple attributes relation.

At this, sightseeing ground " Jing Fugong " and " Ren Si hole " belongs to identical area, therefore, there is the relation on attributes of " atSamePlace ", and go sightseeing in " Jing Fugong ", because having the palace and garden that are available for visitor's visit in sightseeing ground, so having multiple Resource relation, therefore, it is known that " palace (Palace) " and " gardens (Garden) " is respectively provided with the relation on attributes of " hasResource " and " isResourceOf " of twice by " Jing Fugong ".

In other words, being the generalized schematic utilizing process that protege carries out body inference with reference to Fig. 7, Fig. 7, wherein, Fig. 7 a represents the subclass relation in each area, and Fig. 7 b represents the subclass relation on each ground of going sightseeing constituting Datian area.

Meanwhile, nearest user utilizes the various application programs of smart mobile phone to carry out the retrieval of various useful information in mobile environment, as the one in the facility utilizing described various application program, can be exemplified as GPS function.

Therefore, in the present invention, when carrying out retrieval request in mobile environment, the available advantage as a kind of the used GPS in the feature of mobile environment, provide the map retrieval utilizing GPS module in the lump, thus providing various retrieval result.

Specifically, because GPS module can be passed through provide the current location of user, so by the described function present position values with longitude and latitude determination user, and utilize the current location of the user of mensuration can retrieve the sightseeing ground information can gone sightseeing in the place residing for active user, additionally, such as, the Map Services such as Google Maps (Googlemap) are utilized current location can be provided and the map on the ground of going sightseeing retrieved.

Then, after connecting the attribute on each go sightseeing ground and affiliated ground of going sightseeing and relation and the described body inference process of relation between each class being set, to utilize context aware data construct utilize mobile environment semantic data method for digging information retrieval needed for the process of sightseeing body be described in detail.

In other words, in the embodiments of the invention that will illustrate below, in order to build sightseeing body, utilize by portal the interior information architecture sightseeing data base being correlated with sightseeing that is that collect to relevant website of going sightseeing, now, data base manipulation MSSQLServer2000personalversion has the address value on each ground of going sightseeing, and constitutes test_db data base and tour_db table.

Specifically, with reference to the schematic diagram that Fig. 8 and Fig. 9, Fig. 8 and Fig. 9 are the part representing the sightseeing data base of described structure and sightseeing database table respectively.

As can be seen from figures 8 and 9, for instance tour_db is made up of the territories such as id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, long (field).

At this, id is appointed as major key (primarykey), such as when wide area city, land for growing field crops, add_1 is made up of wide area city, land for growing field crops, add_2 is made up of middle district, West, East, scholar city etc., add_3 is made up of Sha Ting hole, Wan Niandong, Semen Ginkgo hole, Chang'an road etc., and add_4 specifies the root park road of wide area Shizhong District, land for growing field crops anvil cavern, and add_5 specifies ground, 327 kinds of Mian Donghe temple 1 tunnel, anti-Pu, South Chungchong Gong Zhou city.

Additionally, in tour_site, such as, sightseeing specifying the sights moral booth as being positioned at 2 Dong Guande road 19 on foot, autonomous especially Jizhou city three, road, Jizhou ground title, and syn is when user carries out retrieving with going sightseeing, in order to carry out synonym and the territory arranging the process of consonant word and specifying.

Meanwhile, lat and long specifies latitude and the longitude in each area, and at this, latitude and longitude are such as specified to arrange customer location value, and described customer location value is used for utilizing Google Maps (googlemap) to show map.

Therefore, available database sharing sightseeing ground, the sightseeing ground body built by described process, in the present embodiment, protege_4.0.2 is used in order to carry out ontological construction and body inference, and utilize OWL (network ontology language, WebOntologyLanguage) and RDF, RDFS etc. to build the tour.owl as sightseeing body.

Specifically, in order to build sightseeing ground body, the concept of tour-sites is set to highest class, the add_1 territory (field) being present in tour_db table is specified, for instance wide area city, land for growing field crops, Jizhou especially autonomous road, South Chungchong, wide area city of Pusan etc. are appointed as add_1 class as its lower floor's class.

In other words, with reference to shown in Figure 10, Figure 10 is the schematic diagram of the sightseeing body construction of the embodiment of the present invention.

As shown in Figure 10, upper class at tour_sites body includes the autonomous especially class such as road, South Chungchong in wide area city, land for growing field crops, wide area city of Pusan, Jizhou, in its lower floor, will be present in the add_2 class of tour_db and be appointed as the subclass of the upper class representing each area, such as in wide area city, land for growing field crops is specified below, district, East, West, great virtue are gone, and when autonomous road especially, Jizhou, Jizhou city, XIGUI Pu city etc. are appointed as lower floor's class.

In other words, with reference to Figure 11, Figure 11 it is the schematic diagram representing the body construction launching described add_2class.

Then, in tour_db table, add_3 is specified as subclass, and at described add_3 apoplexy due to endogenous wind, Sha Ting hole, anvil cavern, Semen Ginkgo hole is such as specified in Shizhong District, land for growing field crops, and when autonomous road especially, Jizhou, in Jizhou, autonomous especially road is specified below Jizhou city, and two apprentice 1 holes, three apprentice 2 holes, 1100 tunnels, Chao Tianyi, Jiu Zuoyi, Ya Yue city, seashore hole, Han Linyi, dragon's pool road, dragon's pool 1 hole, cattle island face, writing brush capital face, arboretum road, dragon's pool 3 hole, three Yang3Dongs, apprentice 2 holes etc. are specified below in Jizhou city.

In other words, with reference to the schematic diagram of the body construction that Figure 12, Figure 12 are the add_3class launching the sightseeing body shown in Figure 10.

Specifically, in fig. 12, represent the XIGUI Pu city belonging to area, autonomous especially road, Jizhou Nan Yuan city and as the Zou East Road of subclass below, great Jing city and as subclass below Ma Luolu, Zou West Road 3000 tunnel, hold a memorial ceremony for seashore road, eastern baking hole and the structure such as 70 No. 214 tunnels, li as subclass below.

Then, as its lower floor's class, such as below husky booth park road is appointed as the add_4 class of tour_dbtable in wide area city, land for growing field crops, Zhong Qu, Sha Ting hole, and in its lower floor's class, namely in wide area city, land for growing field crops, Zhong Qu, Sha Ting hole, husky booth park road below O-World is appointed as tour_site class.

It addition, when as the tour_site class of O-World, process to carry out synonym or consonant, in order to specifySynonym relation, synonym part of speech (equivalentclass) is set.

Meanwhile, with reference to the schematic diagram of the body construction that Figure 13, Figure 13 are the add_4class launching the sightseeing body shown in Figure 10, the relation of add_4 and the tour_site of the palace The South Pool etc. represented in Yi Ling mountain, prefecture Fuyu County of South Chungchong Fuyu County in tumulus, the southeast.

Especially, the protege_4.0.2 partial schematic diagram representing the tour.owl body source code utilizing OWL (WebOntologyLanguage), RDF and RDFS to build is utilized with reference to Figure 14, Figure 14.

Therefore, as it has been described above, sightseeing data are utilized as context data, so that utilize the inference of the sightseeing body of context aware data and build possible.

In other words, with reference to shown in Figure 15, Figure 15 is the integrally-built generalized schematic for providing the context aware body that the user based on context aware pays close attention to information to realize method according to embodiments of the present invention.

Specifically, context aware body according to embodiments of the present invention realizes method and includes series of steps: first collects the information architecture user to user's region-of-interest and pays close attention to information database (S151 step), perform to arrange and be stored in (S152 step) after the body inference that constructed user pays close attention to the relation between each data of information database and attribute, relation according to each data set in body inference step and attribute, build context aware ontology database (S153 step).

At this, the described user of structure pays close attention to information database and builds the detailed process of ontology database by carrying out body inference, can refer to Fig. 2 to Figure 14, can realize together with the process at structure sightseeing body described above, at this, should notice for the purpose of simplifying the description, eliminate illustrating each step.

Then, context data storehouse configured as mentioned above is utilized to illustrate based on semantic data excavation (SemanticDataMining) method finding to inquire about the FP-growth algorithm of similar or identical frequent mode with user.

In other words, semantic data is excavated and is referred to be likely to so that the structure of the logic to various data, and user's inquiry is provided with and information fast and accurately, find the process of the pattern hidden, in embodiments of the invention illustrated below, it is suitable for and useful information to utilize the existing FP-growth algorithm based on FP-Tree effectively to transmit to user, illustrate to utilize the data mining process of similar query pattern and context aware data (ContextAwarenessData) to provide the service being suitable for perform semantic data and excavate a series of processes of (SemanticDataMining) inquiry of user.

At this, for instance what propose in [Gu2010] is the same, for performing data mining process, it is necessary to following several definition (with reference to list of references 15).

[definition 1] (XMLQueryPatternTree)

XML query scheme-tree is the subtree (rootedtree) with root, by XQPT=<X.V, X.E>constitute.

At this, X.V is vertex set (vertexset), X.E is limit collection (edgeset), root represents with Root (XQPT), each limit is using (the v1 as summit, v2) represent, v1 is the father and mother (parent) of v2, each vertex v possesses and has { *, //, the label (label) of the value of tagSet}, tagSet represents the name set of all key elements (element) and the attribute (attribute) being present in DTD.

[definition 2] (XMLQueryRootedSubtree)

When providing query pattern tree XQPT=<X.V, X.E>time, as the XML query subtree XQRST=<X.V', X.E' having root of subtree of XQPT>, satisfied such as Root (XQRST)=Root (XQPT) andDuring the same inclusion relation, become the subtree of XQPT.

In other words, the structure searching the useful subtree taken place frequently is called XML query subtree, and searches similar structure based on this, and now, the tree found is referred to as XML query scheme-tree.

Specifically, with reference to the exemplary plot that Figure 16, Figure 16 a and 16b is expression XML query scheme-tree (Querypatterntree) respectively and the XML query subtree (QueryRootedsubtree) having root.

As shown in figure 16, the XQPT_DB={XQPT as query pattern tree data base is obtained after query set (queryset) being made as query pattern tree and it is stored₁,...XQPT_n, and query pattern excavates (querypatternmining) and searches all subtrees with the root that takes place frequently meeting minimum support (minimumsupport) in data base XQPT_DB.

In other words, in XQP_DB, include the FRQ (XQRST) representing degree of taking place frequently of all situations with the subtree XQRST of root to represent, at this, for trying to achieve support (support) value of XQRST, utilize following mathematical expression SUP (XQRST)=FRQ (XQRST)/| XQP_DB |, the degree of taking place frequently of XQRST is divided into the quantity of all database be calculated.

It addition, for the σ (σ > 0) with certain value, if XQRST meets the condition of SUP (XQRST) >=σ, then mean that XQRST takes place frequently in data base XQP_DB.

It is represent three kinds of query pattern trees respectively and there is the schematic diagram of structure of the subtree that takes place frequently of root for example, referring to Figure 17, Figure 17.

In fig. 17, XQRST occurs in XQPT1 and XQPT2, and therefore, the degree of taking place frequently with the subtree (RootedSubtree) of root is FRQ (XQRST)=2, and SUP (XQRST)=2/3.

As mentioned above, the process excavating the query pattern that takes place frequently is the process searching similar query pattern for improving search efficiency, in the process, for finding whether query pattern is similar or identical, it is necessary to the query pattern tree matching process shown in following [definition 3].

[definition 3] (XMLQueryPatternTreeMatching)

When having query pattern tree XQPT and having the subtree XQRST of root, if meeting following condition, then XQRST is contained in XQPT.

1, the root node (rootnode) of XQRST and XQPT shares identical label (label).

If certain node w ∈ XQRST and v ∈ XQPT of 2 XML tree structure is consistent, then will have the relation of w.label≤v.label, and each subtree of w will have the relation of the subtree being contained in XQPT.

In other words, as the XQRST of the proposed query pattern structure of user between XQPT structure, when relation between each node exists inclusion relation as above, then may be defined as XQRST and be contained among XQPT, and the query pattern tree of user becomes frequent pattern tree (fp tree).

Hereinafter it is assumed that user performs after inquiring about as follows, illustrate to search the process of the answer to inquiry.

Q1: please find out type (genre) for action (action), makes in the film that year (year) is 1994, and leading role (actor) is the title (title) of the film of " JeanReno ".

For finding the answer of described inquiry, the data of genre, year, actor and the value to search and their path need to be denoted as based on described inquiry.

In other words, inquiry is represented with //year=" 1994 " and//genre=" action " and//actor=" JeanReno ", and described path is used as affairs, by the semantic data mining process based on FP-growt algorithm, perform the operation utilizing the query pattern of user to search similar pattern.

Therefore, utilize basic definition as above, semantic query mode excavation to XML data is to be performed by following process: builds frequent pattern tree (fp tree) (FP-Tree) and searches the XML query mode configuration taken place frequently, thus finding the useful subtree (with reference to list of references 16 and list of references 17) similar or consistent with the inquiry of user.

At this, in order to be taken place frequently query pattern by described process model mining, in the present invention, employ the Apriori algorithm more effective FP-growth algorithm in time and space of algorithm than the structure that takes place frequently as existing lookup, as mentioned above, in order to utilize FP-growth algorithm to perform the data mining to the tree schema that takes place frequently, it is necessary to the process (with reference to list of references 18) of following structure frequent pattern tree (fp tree) (FP-Tree).

Specifically, the process building frequent pattern tree (fp tree) (FP-Tree) is first, XML query pattern to be decided to be affairs, and is investigated the degree of taking place frequently of each project (item) in the data base being made up of described affairs by preliminary sweep.

At this, degree of taking place frequently represents the support angle value of a key element (element) project set (itemssets).

The second, by the project in each affairs by degree of taking place frequently descending.

3rd, in the project by degree of taking place frequently descending, the project fewer than the minimum support (minimumsupport) that user specifies is removed.

4th, after the project not taken place frequently being removed from transaction database in described step, remaining project build frequent pattern tree (fp tree) (FP-Tree) is utilized.

At this, frequent pattern tree (fp tree) (FP-Tree) is set for prefix (prefix), and each path (path) utilizes the affairs sharing identical prefix (prefix) in tree to represent, and each node represents a project.

Therefore, can pass through to repeat described process and build frequent pattern tree (fp tree) (FP-Tree).

In other words, with reference to the generalized schematic that Figure 18, Figure 18 are the process sharing prefix (prefix) process and each affairs of expression in the process of the structure frequent pattern tree (fp tree) (FP-Tree) performed by described process.

In figure 18, frequent pattern tree (fp tree) (FP-Tree) is data base to be represented in a compressed format, is connected to each project and the pointer (pointer) as node link in header (Header) table.

As mentioned above, utilization realizes by performing following process by building the data mining process of the FP-growth algorithm that frequent pattern tree (fp tree) (FP-Tree) searches XML query frequent mode: selects to meet the project of minimum support, and removes the project less than minimum support.

It addition, be represent to carry out, in the process of data mining, building the schematic diagram of the process of condition (conditional) frequent pattern tree (fp tree) (FP-Tree) as Arbitrary Term purpose m at execution FP-growth algorithm with reference to Figure 19, Figure 19.

As shown in figure 19, each project is repeated the process building condition tree search criterion pattern.

Meanwhile, it is the result being denoted as repeatedly performing the process of described search criterion pattern with reference to Figure 20, Figure 20, is arranged the schematic diagram of the condition pattern obtained by table.

Then, illustrate, when user inquires about, to be suitable for the example of described data digging method.

First, it is assumed that carried out XML query based on following XQuerySyntax.

[inquiry]: in having the book that the people of name of " Kim " writes, retrieval title, writer and price.

It addition, be the exemplary plot that the Korean inquiry that user performs is made as XML form with reference to Figure 21, Figure 21.

At this, lastname=" Kim " represents the content of title, author, price of retrieving its book in the book that the people being called " Kim " writes, resultPattern represents the pattern normal form (schemapattern) to Query Result, predicates represents the filtercondition required when performing inquiry, and documents represents relevant XML file.

Meanwhile, reference Figure 22, Figure 22 is the schematic diagram being denoted as the BookDTDTree based on the XML document made by described Xquery.

In fig. 22, BookDTDTree represents the form of the data base being made up of affairs.

Specifically, the mining process that data in order to perform the query pattern that takes place frequently to the inquiry from user find, first, build the data base by being become affairs by the inquiry of user each query pattern tree produced, it is made whether to check for useful query pattern and store to each query pattern tree, thus building the data base being made up of affairs.

Then, search, by having in storage in the data base of query pattern tree, the root tree having, and perform data mining by searching the process of similar pattern.

In other words, as mentioned above, the structure of frequent pattern tree (fp tree) (FP-Tree) form is generated based on the affairs being made up of each query pattern, and the data mining process of the FP-Growth method for searching frequent pattern tree (fp tree) is performed to utilize based on FP-Tree generated as discussed above, taken place frequently tree by described process model mining, thus searching the maximum subtree that takes place frequently (maximalfrequentsubtree).

At this, the definition of the inclusion relation of tree required in the process that lookup is described is as follows:

[definition 4] (TreeSubsumption)

When there being two S and S as subtree collection₁Time, it is S=S₁∪{t₁, t₂,...,t_nRelation.

At this, ift_j<t_i, then S₁It is contained in S, is expressed as S₁<S。

[definition 5] (MaximalFrequentSubtree)

If the set of subtree is subtree and the relation that do not comprised by other any subtrees that take place frequently of taking place frequently, then described subtree collection is the subtree (MaximalFrequentSubtree) taken place frequently most, in other words, existsRelation.

As mentioned above, in an embodiment of the present invention, in order to improve the efficiency to user's inquiry, utilize the FP-growth method based on existing association rule algorithm, and as the extension of FP-growth method, searching on the basis of method of the similar query pattern to XML data, add and distinguish subtree and the non-mode taking place frequently subtree and removing the non-subtree that takes place frequently of taking place frequently.

Additionally, by utilizing FP-tree algorithm to search max model subtree (maximalpatternsubtree), query pattern from user is analyzed and by comparing with max model subtree, user's query pattern is classified, whether consistent with max model subtree by described Classification Count user's query pattern, subtree and the max model subtree of user's inquiry is compared, thus performing to search the mining process of max model subtree by repeatedly performing described investigation.

Then, the data mining process utilizing context data to search the scheme-tree to user's inquiry is carried out process.

In other words, in an embodiment of the present invention, utilize context data (ContextData), perform semantic data by scheme-tree data mining (PatternTreeDataMining) method of FP-growth algorithm and excavate.

Now, context data is to utilize portal website and have internet site's collection sightseeing data of each sightseeing information, and utilizes the data construct sightseeing data base collected, and utilizes widely used Protege_4.0.2 to build sightseeing body.

Additionally, in sightseeing body, the OWL (WebOntologyLanguage) as Ontology Language is utilized to be represented by RDF and RDFS, thus generating the ontology database being called tour.owl, based on each area, the small-scale area belonging to this area is appointed as the subclass under it, and, under the subtree of multilamellar, sightseeing is appointed as another subclass.

Meanwhile, the relation and the attribute that utilize go sightseeing ground and each department specify the classes such as subclass, superclass, synonym part of speech (equivalentclass), and in each area, in the relation belonging to regional multiple ground of going sightseeing, it is intended that be similar to the attribute of atthesameplace.

Especially, when user carries out retrieval request, process to carry out synonym and consonant word, specify synonym part of speech, by the famous restaurant on each ground of going sightseeing belonging or comprise the described content of explanation going sightseeing ground is appointed as attribute, and longitude and latitude position attribution utilizes Google Maps (Googlemap) service that each contextual information going sightseeing ground is supplied to user.

Therefore, the ontology database of described structure has the hierarchical structure utilizing RDF and OWL, accordingly, in the present invention, in order to utilize sightseeing body to perform data mining process, XML data method for digging be may utilize and excavate mode into master data, and, in order to search the query pattern of incidence relation to user, in order to utilize existing FP-growth algorithm to search the same or similar ground of going sightseeing with the inquiry of user, it is suitable for the scheme-tree data digging method that lookup is stored in the same or similar pattern of body.

Specifically, tour is appointed as root, utilizes the sightseeing constituting sightseeing data base ground information that each area is configured to next subclass relation, relation is set under it, to constitute another subclass, thus the hierarchical structure constituting each area and ground of going sightseeing is appointed as affairs.

Additionally, when user to retrieve the sightseeing ground of particular locality, because the address value to the ground of going sightseeing constituting the inquiry from user, also the hierarchical structure there is the structure as being stored in ontology database, therefore, by searching the process execution data mining process of the affairs similar or consistent with the affairs of the hierarchical structure being stored in ontology database of the structure with user's inquiry.

Meanwhile, with reference to the schematic diagram that Figure 23, Figure 23 are the project set that takes place frequently (transactionfrequentitemset) representing each affairs in sightseeing ontology database

In fig 23, constitute affairs by each sightseeing and the structure of the hierarchical structure constituting specific one ground of going sightseeing is constituted as each affairs table, and representing the content on each ground of going sightseeing of the affairs constituting affairs id and take place frequently.

Especially, reference Figure 24, Figure 24 is the partial schematic schematic diagram of the FP-tree structure of the sightseeing ontology database by utilizing as discussed above the structure that takes place frequently by each area generation.

In fig. 24, utilize in left side each area name to constitute project (item), and on right side, the project constituting each affairs is used as a node, thus representing the link of node.

It addition, numerically constitute the degree of taking place frequently of the project of each node on its side.

In other words, such as in sightseeing ontology database, when autonomous road especially, Jizhou, because situation about taking place frequently occurs 5 times, so representing with 5, and when XIGUI Pu city, because occurring 3 times, so representing that there is the degree of taking place frequently crying 3, to each other area names, proceed labelling as above, thus building FP-tree structure.

Meanwhile, the condition FP-tree to certain specific project can be built based on described FP-tree.

In other words, it is the whole world (Global) FP-tree with reference to Figure 25, Figure 25 a, and the schematic diagram that Figure 25 b is the situation building condition (Conditional) FP-tree with the Ma Luolu going sightseeing in area for benchmark.

As shown in figure 25, FP-growth algorithm performs based on constructed FP-tree, certain specific project to be repeatedly generated the process of condition FP-tree, and namely the process searching FP-tree structure by repeating the process generating the condition FP-tree to each project described is data mining process.

Specifically, such as when user sends the inquiry of retrieval " mountain, Changtai, Chang'an road, West, wide area city, land for growing field crops mastery woods ", in order to search the inquiry similar with the inquiry of user, perform the process of investigation sightseeing ontology database, when there is the structure identical with the inquiry of described user, perform service quickly and accurately can to user, but time, when searching only consistent with " wide area city, land for growing field crops West Chang Anlu " structure, because being similar but not being consistent structure, so providing the result being not as accurate but similar structure.

Meanwhile, with reference to Figure 26, Figure 26 it is the part of original code schematic diagram (with reference to list of references 19) representing the FP-growth algorithm performing process as described above.

Then, the process utilizing FP-growth algorithm as above to perform data mining it is described in more detail.

First, in order to ontology database is applicable to data mining, perform the preprocessing process that data are encoded.

For this, the data in each area belonging are transformed to suitable code and are suitable for FP-growth algorithm by the present inventor, observe, with this, the process generating which type of pattern.

At this, the same with the Apriori of applicable associations rule, FP-growth algorithm is also suitable Minimum support4 (minimumconfidence) and minimum support (minimumsupport), therefore, observation is suitable for what the result that during marginal value as above (threshold), applicable data excavates is.

Specifically, it is represent that being suitable for by minimum support in FP-growth algorithm is the schematic diagram of 0.4 result carrying out data mining with reference to Figure 27, Figure 27.

As shown in figure 27, the quantity of the affairs being stored in data base is 99, and maximum memory makes consumption be 0.96mb, and frequent item set (frequentitemset) generates 27, and performing the time that data mining consumes is 15ms.

Additionally, can be observed when performing five steps of L1 to L5, while being suitable for support angle value with minimum support (minimumsupport) value 0.4 respectively for benchmark, by each step generation mode, at this, the value in each pattern is the value meaning the value in the area in data base is deformed.

Therefore, in figure 27, it can be seen that the frequent mode on the ground of going sightseeing being pointed to the XIGUI Pu city in autonomous especially road, Jizhou the most often occurs.

Above, utilize the process that context data performs data mining to be illustrated to based on FP-growth algorithm, then, the overall process realizing information retrieval in mobile environment by performing semantic data mining process is described in detail.

In other words, the present invention is in order to accurately provide user's useful information needed for current location compactly, arrange in being arranged on attribute including each interlocal class relation, relation in order to show the word representing similar word or identical meanings arranges and utilizes sightseeing body, and in embodiment as above, construct sightseeing ground body for sightseeing ground information regional needed for accurately transmitting user.

Additionally, go sightseeing in ground body, utilize hierarchical structure point different regions by go sightseeing ground Informational Expression be body, such as when user sends the inquiry on " the sightseeing ground that retrieval is positioned at area, scholar city, land for growing field crops city " when being currently located at area, scholar city, land for growing field crops city, the area, scholar city in wide area city, land for growing field crops it is contained in by body retrieval, and show as retrieval result with being included in sightseeing in scholar city to user, meanwhile, such as, as Google Maps (Googlemap), by being built in the map on the ground of going sightseeing required for the map function of mobile equipment provides current location and user, and, by integrate a series of process provide be pointed to user needed for customer location nearest area go sightseeing ground retrieval service.

Especially, as it has been described above, the data in sightseeing ontology database are performed data mining process by available FP-growth algorithm, provide a user with useful information, below, the semantic information retrieval process performed being suitable for described semantic data mining algorithm is described in detail.

First, with reference to Figure 28, Figure 28 it is the integrally-built generalized schematic representing the information retrieval system based on movement utilizing the method for mode matching based on semantic data mining algorithm.

In system as shown in figure 28, the operation performed by modules and overall flow are described as follows:

The first, user performs information retrieval requests in current location.

In other words, when user uses mobile equipment to initiate retrieval request, utilize WiFi, 3G, 4G, GPS, Sensor etc. by requiring to transmit together with item with user search to the information of user current location, thus collecting the contextual information (ContextInformation) of user.

The second, context data storehouse (contextdatabase) is built.

At this, in the present embodiment, in order to collect sightseeing ground information, collect the sightseeing ground information of each the Internet portal website and each area introduction, in order to each of each area go sightseeing and the setting of the feature on this sightseeing ground and address and periphery restaurant and Google Maps (Googlemap), collect longitude and the latitude value in each area.

It addition, by gathered as explained above to data after extractive process and preprocessing process, be stored in context data storehouse (contextdatabase).

3rd, the data being stored in context data storehouse by described process are performed body inference process, attribute and the relation of each data value are set, thus building ontology database.

At this, in body inference process, in order to the subclass (subclass) in each area, superclass (superclass) relation and synonym or consonant word process, synonym part of speech (equivalentclass) relation is set, " isattractionof " relation on attributes between each area and sightseeing ground, performance belong to the attribute such as " atthesameplaceof " of relation on ground of going sightseeing in identical area.

4th, in order to utilize the data base being stored in body to improve recall precision, perform to search the semantic data mining process of the query pattern tree similar or consistent with the inquiry from user.

At this, the process utilizing the requested retrieval information of user and contextual information search pattern tree is to utilize existing FP-Growth algorithm to perform data mining process.

5th, provided passing through to perform the acquired result of described process as to the result of the information required by user by context aware Information Mobile Service interface (ContextAwarenessMobileServiceInterface).

Therefore, the overall process utilizing the retrieval of semantic data method for digging to provide user's information needed in mobile environment by described process, it may be achieved realize the general frame of information retrieval in mobile environment by performing semantic data mining process, then, is described in detail.

In other words, the generalized schematic of performed a series of processing procedures when being represent the retrieval request receiving ground of going sightseeing from user with reference to Figure 29, Figure 29.

Specifically, as shown in figure 29, if the first user information retrieval to ground of going sightseeing that I to retrieve by service request module (ServiceRequestModule) request, then the mobile search systematic analysis of body is utilized to require item from the retrieval of user and constitute query pattern.

The second, the current information of user is collected by situation processing module (ContextProcessModule), in other words, as mentioned above, obtaining the information content of user's inquiry interior by WiFi, 3G, 4G, GPS, sensor (Sensor) etc., user requires the requirement item of the current location under the situation of retrieval and user.

3rd, sightseeing content (TourContents) DB can be referred to as described context data storehouse (ContextDatabase), is that the portal sightseeing in interior each department is introduced the ground information of going sightseeing of website and carried out mobile phone the sightseeing ground data base built.

At this, go sightseeing in ground data base, after performing preprocessing process refinement data and being stored in relevant database, build, by performing body inference process, ground ontology database of going sightseeing.

Therefore, as mentioned above, in order to utilize the query pattern being stored in the context data storehouse being stored as bulk form to search the pattern that the user query pattern requested with user is consistent, similar or identical query pattern is found by the sightseeing ontology database that constitutes of hierarchical structure relation on ground of going sightseeing by retrieving, now, as it has been described above, by utilizing the semantic data mining algorithm based on FP-growth algorithm to search similar or identical query pattern.

4th, in ontology database, retrieve, by retrieval module (RetrievalModule), the query pattern that user asks the information of retrieval, and provide a user with the retrieval result needed for user together with GPS information.

Then, illustrate to search similar or model identical tree data mining process.

First, utilize internet sites etc. to collect and go sightseeing relevant information build data base, and utilize the ontology database of described data construct to be referred to as to search the pattern database (PatternDatabase) of user's query pattern when user inquires about, and contain various pattern at described pattern database (PatternDatabase) internal memory, therefore, when user performs to inquire about, perform the comparison procedure with present mode, perform data mining thereby through the process searching consistent pattern or similar pattern and inconsistent pattern.

Therefore, when find inquire about identical or similar pattern with user time, the useful information needed for user can be provided quickly and accurately to user.

In other words, with reference to Figure 30, Figure 30 is the schematic diagram that following process is illustrated: from user receive retrieval request time, the query pattern of user is analyzed by analyzing the requirement item of user, and by the mode configuration of the consistent pattern of the query pattern of comparison and analysis and the ontology database being stored in searching system, retrieve similar or identical mode configuration.

At this, in fig. 30, pattern database refers to sightseeing ontology database.

In other words, if the user while the inquiry request to ground of going sightseeing is initiated in current location, then in the sightseeing ontology database that preprocessing process and body inference process store, whether consistent the query pattern of investigation active user's request is with the existing query pattern being stored in pattern database, retrieval result is shown when one makes peace similar, and search pattern again in the case of inconsistencies.

Specifically, if the user while the retrieval window of searching system carries out the retrieval request to my desired ground of going sightseeing, thus obtaining the retrieval request from user, then the analysis process passing through to perform the query pattern to user generates user's query pattern.

Then, check that whether user's query pattern is new pattern, when user model is new model, whether consistent with present mode investigate, if there being consistent pattern, then will be stored in the consistent pattern of pattern database (patterndatabase) provides as retrieval result.

If the pattern that cannot investigate new pattern and be stored in pattern database is whether consistent, then again perform described process, when there being similar pattern, similar pattern is provided as retrieval result, and the new pattern of user's inquiry is stored in pattern database in the new modes.

Therefore, as mentioned above, according to embodiments of the invention, by repeatedly performing described a series of process, query pattern identical or similar with the inquiry of user is provided as retrieval result, therefore, can realize from sightseeing body sightseeing data utilized as context data and build, the user based on context aware utilizes the semantic data mining algorithm based on FP-growth algorithm to search user and pays close attention to the search method of information, it addition, can be provided accordingly to pay close attention to information recommendation service.

In other words, the integrally-built generalized schematic of information retrieval service offer method is paid close attention to reference to Figure 31, Figure 31 for the user based on context aware that semantic data is excavated that utilizes according to embodiments of the present invention.

Specifically, as shown in figure 31, the user based on context aware that semantic data is excavated that utilizes according to embodiments of the present invention pays close attention to information retrieval service offer method, first, if user utilizes the mobile equipment including smart mobile phone or panel computer of oneself to send the retrieval request to the information of concern in current location, then utilize the WiFi of terminating machine, mobile radio communication (3G, 4G etc.), GPS and be arranged at the various sensors of affiliated terminating machine and receive the retrieval request of described user and the information of the current location to described user simultaneously, and it can be used as the contextual information (ContextInformation) of described user to collect (S311 step).

Then, to collected information and executing extractive process and preprocessing process and build context data storehouse (contextdatabase) (S312 step), and each data being stored in constructed context data storehouse are arranged by body inference attribute and the relation of each data value, thus building ontology database (S313 step).

Then, utilize the ontology database built by described process, based on described FP-Growth algorithm, the semantic data excavation performing to search the similar or consistent query pattern of inquiry required with user processes (S314 step), and using the result being processed step acquisition by semantic data excavation as the mobile terminal (S315 step) that the retrieval result of the retrieval request that user requires is shown in user.

At this, the content more specifically of the process performed in each step described can refer to the content illustrated by prior art literature and Figure 16 to Figure 30, therefore, illustrates for simplifying, and omits the detailed description to each step at this.

Then, describe in detail in order to the environment needed for realizing utilizing the mobile message searching system of context aware body as above and its realize process.

In other words, in the described embodiment, the present inventor is for collecting sightseeing data, the portal website on the interior ground of going sightseeing introducing each area is utilized to collect ground information data of going sightseeing, and perform the preprocessing process that collected information is refined, such as in order to build relevant database in the operating system environments such as Windows, MSSQLServer2000Personalversion is utilized to build data base.

Additionally, in data base, respectively constitute the territory (field) of id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, long and store data, at this, each domain representation as add_1, add_2, add_3, add_4, adde_5 constitutes the address on each ground of going sightseeing, tour_site storage sightseeing ground title, syn is the domain name processed for synonym or consonant word, and lat and long represents latitude required when arranging map (Googlemap) and longitude respectively.

Meanwhile, the present inventor is by utilizing relation between described each data of relational data lab setting and attribute and building sightseeing body (Ontology), now, in order to build ontology database, utilize Protege_4.0.2version to perform ontological construction and inference, and utilize OWL (WebOntologyLanguage) and RDF to show ontological construction tour.owl body.

Especially, the present inventor utilizes the ontology database of described structure to perform to excavate based on the semantic data of FP-growth algorithm and build the information retrieval system of mobile environment, at this, realize in environment for what realize mobile message searching system, such as utilize [Android2014] (with reference to list of references 20) proposed Android4.3 version and be used for realizing the GoogleAPIs4.3 of Google Maps (Googlemap), in order to realize mobile message searching system, make use of HTML5, CSS3, JavaScript etc., and in order to be utilized as data base and server, make use of the NodeJSserver for high-capacity database.

Additionally, because Android programs based on Java, so utilizing Jdk1.7.0_05 to constitute realize environment, the editing machine for building the information retrieval system of mobile environment make use of the WebStrom_7.0.3version (with reference to list of references 21) being such as set forth in [WebStorm2014].

At this, if the feature that the Android as the mobile platform (MobilePlatform) suitable in the present embodiment is programmed illustrates, then have a characteristic that first as the Android of platform being used for the equipment that moves, Android be Google make, disclosed open source mobile platform, be suitable for mobile environment based on the operating system of Linux, user cross section, application program set；Second is different from the mobile phone environment closed so that exploitation and the operation of the various application programs of restrained minimization are likely to.

Meanwhile, the detailed content to the development process utilizing Android, the prior art literature that those skilled in the art can pass through such as [Michalis2012] (with reference to list of references 2) is easy to understand, therefore, does not repeat them here.

In other words, as mentioned above, the present inventor utilizes the Internet portal website or introduces the information architecture sightseeing data base of website of sightseeing information in each area, and in affiliated sightseeing data base, constitute the territories such as id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, long, at this.Such as add_1 is by " wide area city, land for growing field crops ", " autonomous especially road, Jizhou ", " wide area city of Pusan ", area names such as " South Chungchongs " is constituted, add_2 when wide area city, land for growing field crops by " West ", " East ", " middle district ", " scholar city ", district's titles such as " great virtue districts " is constituted, add_3 is as being positioned at " Wan Niandong " of West, wide area city, land for growing field crops, " the Semen Ginkgo hole " etc. that be positioned at wide area Shizhong District, land for growing field crops equally specifies hole title, add_4 is as being positioned at " mountain, the village main road " in 10000 years holes, West, wide area city, land for growing field crops, " by inner " of being arranged in chicken dragon face, South Chungchong Gong Zhou city etc. equally specifies better address, and add_5 specifies " No. 169 " of mountain, the village main road such as 10000 years holes, West, wide area city, land for growing field crops, it is arranged in chicken dragon face, South Chungchong Gong Zhou city by lot numbers such as inner " No. 52 ".

Additionally, tour_site specifies " the culture and arts garden, land for growing field crops " of such as mountain, village main road 169 as being arranged in 10000 years holes, West, wide area city, land for growing field crops, is positioned at chicken dragon face, South Chungchong Gong Zhou city by the ground of the sightseeing " the first temple " of inner No. 52 title, syn is such as in order to the synonym that uses when " land for growing field crops " being appointed as " Korea Spro's bar (hanbat) " or " Jizhou " is appointed as " indulging in sieve (tamra) " processes, or in order to when inputting search key input as the consonant word at " first temple "Consonant word as land for growing field crops culture and arts gardenDeng consonant word process and specify, and lat and long specifies longitude and the dimension values in each area to represent map (Googlemap).

Meanwhile, in described embodiments of the invention, the sightseeing data base of storage as described above is for the structure of ontology database of going sightseeing, and in order to utilize the attribute of each data in data base and relation to carry out body inference and ontological construction and to make use of Protege_4.0.2version, and OWL (WebOntologyLanguage) and RDF is utilized to build the ontology database of tour.owl.

At this, body inference process specifies attribute and the relation of the data value representing each area, in other words, specify the superclass (superclass) to each area and subclass (subclass) relation and multiple inclusion relation, specify the synonym part of speech (equivalentclass) processed for synonym and consonant word, when the sightseeing ground being contained in identical area, relation on attributes as " atthesameplaceof " is set, when the sightseeing ground being contained in area, such as it is positioned at cattle Dao Mianniu island, autonomous Jizhou city, road especially, Jizhou with going sightseeing, can fasten, with closing with going sightseeing, the relation on attributes arranged as " isattractionof " in area.

Especially, in described embodiments of the invention, in order to utilize the ontology database built by described process to provide a user with information retrieval structure more quickly and accurately, perform to excavate (SemanticDataMining) process based on the semantic data of FP-Growth algorithm.

At this, the hierarchical structure relation of the higher level's class and subordinate's class and ground of going sightseeing that constitute each area of ontology database is considered as an affairs (transaction) and is considered as constituting a scheme-tree by described data digging method, and based on the scheme-tree of so storage, when the inquiry receiving user, the process looking into and searching identical or similar scheme-tree by comparing the scheme-tree of user's query pattern and existing storage carries out.

Therefore, the data mining process that can pass through to perform described process provides the identical or similar retrieval result of search key required with user.

Then, illustrate to be constructed by the described semantic data mining process based on FP-Growth algorithm in mobile environment, carry out the process of the data retrieval system of retrieval to user's inquiry.

At this, in the embodiments of the invention that will illustrate, to utilize the programs such as HTML5, CSS3, JavaScript to realize searching system describe the present invention to realize the information retrieval system of mobile environment, but the present invention is not limited to this, any other language or platform can be utilized as required to realize.

In other words, if utilizing Android (Android) programming realization system, then can only operate in the equipment with Android system, but have the advantage that as mentioned above, if utilizing HTML5 and CSS3 to be programmed, then being possible not only to operate in Android system, also may operate in iOS system.

Additionally, in the embodiments of the invention that will illustrate, event-action (Eventaction) as button and function etc. utilize JavaScript in processing, and now, the program as editing machine is WebStrom_7.0.3version (with reference to list of references 21).

Meanwhile, in order to perform the system built, make use of Android4.3 version (with reference to list of references 20), and in order to realize the map (Googlemap) in each area, make use of GoogleAPIs4.3forGooglemap.

Specifically, with reference to the schematic diagram that Figure 32, Figure 32 are for realizing retrieving the part of original code of the initial picture of window.

In other words, the source code representation shown in Figure 32, after user is by the required sightseeing ground of retrieval window input, shows its result at lower one page with catalogue form.

It addition, be the source code schematic diagram realizing the event-action (Eventaction) to index button (searchbutton) by script function (Scriptfunction) with reference to Figure 33, Figure 33.

In other words, source code representation as shown in figure 33 performs the button actions of following operation: to retrieval window input search key and click the button on its side or when click is positioned at three buttons of key frame lower end user, conversion page, shows end value thereby through button click in picture.

Meanwhile, with reference to Figure 34, Figure 34 for representing that execution is used for providing the part of original code schematic diagram of the map map button action that (Googlemap) services.

In other words, source code as shown in figure 34 is the action of event-action and the function display Google Maps utilizing btn-map, represents that the latitude value utilizing area and longitude show the button actions of the action of the map of each position belonging.

Especially, reference Figure 35, Figure 35 is the composition exemplary plot of the key frame of mobile message searching system according to embodiments of the present invention.

As shown in figure 35, if the retrieval window of the key frame in left side inputs area name or sightseeing ground title and clicks index button, the then retrieval result on the belonging affiliated area of output or ground of going sightseeing, in other words, if such as to the ground zones values of retrieval window input " land for growing field crops ", then as right panel, export the sightseeing ground title relevant to land for growing field crops.

At this, if address value inputs other values for having subclass relation with wide area city, land for growing field crops, then retrieval end value will show more detailed result.

Additionally, as the right panel of Figure 35, lower end (footer) part at all pages of the key frame except searching system shows multiple button, in other words, such as, first button is the button that back to homepage, second button be read input to retrieval window and value call the map button of map in relevant area, the 3rd button is restaurant's title in the area for showing belonging relevant ground of going sightseeing or the content page button of introduction to ground of going sightseeing.

Specifically, it is represent the schematic diagram to the composition shown by the map in affiliated area when retrieving the ground zones values in window input wide area city, land for growing field crops and click map button and the composition according to the picture to the relevant regional restaurant of input value display introduction or area of going sightseeing of retrieving window input respectively with reference to Figure 36, Figure 36.

At this, described map button, in order to arrange map (Googlemap) in the explanation to described sightseeing body, utilizes the latitude and longitude that arrange with the thresholding of lat and long in data base, with, the map in relevant area is called as end value.

Additionally, with reference to Figure 37 and Figure 38, Figure 37 represents the present position values with latitude and longitude automatic sensing user, thus realizing together representing the map in area that user is presently in and the source code schematic diagram of the map of latitude and longitude to user, and Figure 38 is the schematic diagram representing the result screen realized by source code as shown in figure 37.

As shown in figure 37, map realizes source code and may be configured as follows: create myLocation function to realize the map of belonging user current location, and uses navigator.geolocation.getCurrentPosition function to try to achieve the present position values of user wherein.

Meanwhile, in the result screen shown in Figure 38, as the present position values of user, utilize the value of latitude 36.30 and longitude 127.36 to try to achieve the positional value in hole, Zhenglin, West, wide area city, land for growing field crops, and show the map of its surrounding area.

Therefore, the action utilizing the information of the current location of the present position values transmission user of user and the action realized when user inputs search key is realized respectively, thus the information retrieval system of semantic data method for digging can be realized mobile environment is suitable for by described process.

Then, experimental result and corresponding performance evaluation to the mobile message searching system actual execution information retrieval that utilization is realized by described process illustrate.

In other words, the present inventor builds sightseeing ontology database by described process, and utilizes the data in affiliated data base applicable data mining algorithm to perform experiment.

At this, the embodiment of the present invention being described below perform experiment, it is suitable for the retrieval result of semantic data method for digging for retrieving ground of going sightseeing as purpose obtaining by each area, the sightseeing relevant to each area is considered as affairs and analyzes frequent mode, additionally, in order to quickly provide retrieval result to the sightseeing ground needed for user, the accuracy of information retrieval is also carried out experiment.

First, the experimental result being suitable for semantic data mining algorithm is described as follows: the present inventor is in order to retrieve the sightseeing ground relevant to each area, utilize FP-growth algorithm as the algorithm searching the user's query pattern that is mutually related, and the data being utilized respectively ontology database and existing relevant database compare and evaluate performance.

In other words, the semantic data mining algorithm based on FP-growth algorithm is utilized to perform the process to the frequent mode found between two data and execution time and experimental evaluation that internal memory makes consumption etc. compare, in order to carry out described experimental evaluation, perform the preprocessing process to sightseeing ontology data.

Specifically, first, collect various portal and introduce go sightseeing ground website about each go sightseeing ground information, thus building sightseeing data base；The second, utilizing the data being stored in sightseeing data base, build the tour.owl as sightseeing body, at this, described body utilizes protege_4.0.2 to build, and inference and performance are to make use of OWL, RDF, RDFS.

The 3rd, sightseeing ontology database utilize DOM parser (Parser) perform the parsing to XML document；4th, perform FP-growth algorithm to extract the label (tag) of XML document, create map table (mappingtable), and to each being encoded of label, thus performing mining process.

It addition, when the data of existing relevant database, also extract sightseeing data and perform preprocessing process by the process similar with described ontology data, thus creating map table and carrying out numeralization and perform data mining process.

In other words, it is represent by table ontology database according to embodiments of the present invention and existing relevant database being suitable for FP-growth algorithm respectively and performing the schematic diagram that the experimental result of data mining process arranges with reference to Figure 39 and Figure 40, Figure 39 and Figure 40.

At this, when existing relevant database, different from ontology data, minimum support (minimumsupport) value is only suitable for 0.6, so that experiment is likely to, if becoming higher, then cannot be carried out experiment, thus experiment only carry out described place till.

Specifically, it is the schematic diagram that the internal memory changed by the change of the graph representation ontology data along with the embodiment of the present invention and minimum support critical (minimumsupportthreshold) value of existing relevant database makes the change procedure of consumption with reference to Figure 41, Figure 41.

In other words, the quantity of affairs is limited to 100 and makes consumption, frequent mode quantity, execution time etc. investigate the internal memory changed along with the change of minimum support value by the present inventor, first, change according to the minimum support value between two data is observed internal memory and is made how consumption changes, result shows, as shown in figure 41, the internal memory between two data makes consumption similar.

Therefore, by experimental result as shown in figure 41 it can be seen that when affairs are few, than the data of existing relevant database, the internal memory of the data of ontology database according to embodiments of the present invention makes consumption somewhat a little less, but ratio is relatively similar.

Then, it is that also the quantity of affairs is limited to 100 and the schematic diagram of the number change of frequent mode changed along with the change of the ontology data of the embodiment of the present invention and the minimum support value of existing relevant database by graph representation with reference to Figure 42, Figure 42.

At this, described experiment is according to using the data of which kind of structure as experimental data, and between ontology database and general database frequent mode structure have multi-form premised under carry out, as shown in figure 42, if the pattern that the takes place frequently number change process observed between ontology database and general database, then can confirm that when use ontology data, than general database take place frequently structure occur more.

Described result is when ontology database, because having more structurized hierarchical structure than general database, so more structure that takes place frequently can be produced.

In other words, in the process of component ontology database as above, because the relation being made each class by body inference process carries out structure connection mutually, so can have, than general data, the structure that more takes place frequently, accordingly, according to the present invention, when passing through to search frequent mode offer retrieval result, because the structure that more takes place frequently can be provided, better retrieve result so can provide.

Then, with reference to the schematic diagram of the change that Figure 43, Figure 43 are the execution time (runtime) changed by the change of the graph representation ontology data along with the embodiment of the present invention and the minimum support of existing relevant database.

In described experiment, equally also transactions being limited to 100 and test, as shown in figure 43, than existing relational data, the execution time of ontology database is shorter.

At this, it should be noted that and can confirm that, when relational data, the execution time when supporting angle value little and time maximum sharply increases, but in contrast to this, when ontology database, the change procedure performing the time needs the similar execution time together with the change of minimum support value.

Then, with reference to Figure 44, Figure 44 it is the schematic diagram representing the change performing the time that the change of affairs of the ontology data along with the embodiment of the present invention and existing relevant database changes with graph mode.

In other words, as shown in figure 44, minimum support is fixed on 0.4, and while changing the quantity of affairs, representing the change performing the time of ontology data according to embodiments of the present invention and existing relevant database, result shows the increase along with transactions, and the execution time between two data, difference occurred, and than the data of existing relevant database, the execution time of the data of the ontology database of the embodiment of the present invention is shorter.

Therefore, than the situation utilizing existing relevant database, utilizing the information retrieval of the ontology database of the embodiment of the present invention, when receiving the retrieval request of user, retrieval required time is shorter.

Then, with reference to Figure 45, Figure 45 it is the schematic diagram comparing the change that the internal memory that the change of the affairs representing the ontology database along with the embodiment of the present invention and existing relevant database changes makes consumption with graph mode.

At this, result as shown in figure 45 be minimum support is fixed on 0.4 and changes transactions while carry out the result tested, and as shown in figure 45, the internal memory that the number change along with affairs between two data changes makes consumption, than the situation of the data utilizing existing relevant database, utilize the ontology database of the embodiment of the present invention data situation in internal memory make consumption less.

Additionally, compared with the situation that transactions is limited to 100, than the situation that the quantity of affairs is few, when the quantity of affairs is many, than the situation of the data utilizing existing relevant database, the internal memory in the situation of the data of utilization ontology database according to embodiments of the present invention makes consumption more effective.

Thus, taking into account can as the off-capacity of the mobile equipment of the maximum problem of the information retrieval in mobile environment, according to embodiments of the present invention utilize ontology database situation more effective.

Then, the result that the accuracy of ontology database according to embodiments of the present invention and the information retrieval of existing relevant database is tested is described.

First, with reference to the schematic diagram that Figure 46, Figure 46 are the change representing the accuracy representing recall precision along with the change of the affairs as user's query pattern using graph mode.

In other words, the present inventor utilizes the FP-growth algorithm as the semantic data mining algorithm suitable in described embodiments of the invention, measures the accuracy (Accuracy) of the accuracy representing the information retrieval to experimental data.

Measurement result shows, as shown in figure 46, ontology database and the accuracy of existing relevant database according to embodiments of the present invention increase along with the increase of transactions.

But, if compared to existing portal or to the searching system of relevant website of going sightseeing and the accuracy of the mobile message searching system being suitable for semantic data method for digging built according to embodiments of the invention by described process, then can find that the accuracy of existing system is substantially on the low side.

This is because existing system is when user's search key required to retrieval window input, it is provided that the retrieval result unrelated with the retrieval content desired by user.

Then, in order to the mobile message searching system being suitable for semantic data method for digging of the embodiments of the invention according to described explanation and the difference of existing searching system being compared the accuracy of rear observed information retrieval, the result that result screen time in mobile message searching system according to embodiments of the present invention and existing portal respectively to retrieval window input sightseeing ground is compared is described.

In other words, the schematic diagram of the picture of retrieval result is shown when being, in existing portal, window is extremely retrieved in " chicken Longshan " input with reference to Figure 47, Figure 47.

Specifically, as shown in figure 47, the retrieval result of existing system is, when inputting the sightseeing ground in chicken Longshan to retrieval window, as although the content that retrieval end value shows includes going sightseeing to reality relevant content, but completely irrelevant content also serves as retrieval, and result shows, meanwhile, than computer picture, too much retrieval result is provided at little a lot of smart mobile phone picture, therefore, while user's rolling picture always, retrieval result is confirmed, very inconvenient.

Therefore, for solving described defect, it is necessary to the searching system of an accurate retrieval result advantageously provided desired by user in retrieval result, for this, in the present invention, the problem for solving described existing searching system, it is proposed to and realize utilizing the mobile message searching system of semantic data method for digging.

Specifically, it is in mobile message searching system according to embodiments of the present invention with reference to Figure 48, Figure 48, shows the schematic diagram of the picture of retrieval result when retrieving window input " chicken Longshan ".

In other words, in existing searching system, the content unrelated with chicken Longshan provides to result screen, thus there is the defect that need to be confirmed and select to check information needed by user one by one, but as shown in figure 48, in mobile message searching system according to embodiments of the present invention, when chicken Longshan that input is identical with the value of the retrieval window input search key to existing searching system, only will be worth accurately and offer will be provided as a result.

It addition, when being be illustrated respectively in mobile message searching system according to embodiments of the present invention to input search key with reference to Figure 49, Figure 49, by the retrieval relevant explanation of result and the schematic diagram of picture that surrounding area map is provided in the lump as end value.

As shown in figure 49, if, with mobile message searching system according to embodiments of the present invention, retrieval result to the search key inputted then can not only be provided, and, such as it is worth the map that the surrounding area including the relevant information such as the more detailed description to relevant ground of going sightseeing and periphery restaurant is provided in the lump as a result, therefore, the accuracy than its retrieval result of existing searching system is higher.

Therefore, by described process, context aware body can be utilized in mobile environment to realize the mobile message searching system based on semantic data method for digging.

In other words, according to the present invention, a kind of context aware body can be provided to realize method, described context aware body realizes method for providing user to pay close attention to information service, wherein, first, in order to store context data, sightseeing information is collected by introducing website with including the sightseeing in each area of portal, collected information is carried out pretreatment and is stored in data base, and the sightseeing data base of utilization so storage performs to arrange the body inference process of relation and the attribute etc. on each ground of going sightseeing, thus building sightseeing ontology database.

Additionally, according to the present invention, a kind of user utilizing semantic data to excavate can be provided to pay close attention to information retrieval service method is provided, wherein, in order to utilize the sightseeing ontology database built by described process to provide information retrieval result exactly to user is more rapid, it is suitable for the data digging method based on existing FP-growth algorithm, thereby through the user's query pattern compared from user and the query pattern being stored in sightseeing ontology database, to similar or mutually isostructural inquiry, retrieval result can be provided rapidly.

Meanwhile, according to the present invention, a kind of mobile message retrieval service can be provided, its utilize described context aware body and semantic data to excavate to provide be suitable for mobile environment more succinctly retrieve result accurately while, the Map Services of the position of information retrieval result needed for the current location to user or user can be performed in the lump.

Additionally, according to the present invention, in body inference process, arranging to arrange with relation by attribute makes the process of synonym or similar word, consonant word be likely to, thus various search key can be inputted in carrying out the process retrieved, and can by being suitable for the problem that semantic data method for digging solves the defect that the picture of smart terminal cannot be overcome in existing retrieval service because providing unnecessary retrieval result little.

So far, the context aware body based on context aware according to embodiments of the present invention can be realized by described process and utilize its user to pay close attention to information service.

At this, described embodiments of the invention have the situation paying close attention to information service constituted as follows and describe the present invention for example with offer: in order to predict and provide user the concern information to ground of going sightseeing, build the context aware body (ContextAwarenessOntology) about ground of going sightseeing, and perform semantic data mining process (SemanticDataMining) based on described context aware body, provide a user with more accurately succinct result, thus it is that example is illustrated that information service is paid close attention in the offer being more suitable for mobile environment, but the present invention is not limited to this, in other words, the present invention is not only suitable for the concern information to ground of going sightseeing as above, and, the service of the concern information to other any regions can be applicable to as required.

Therefore, by described process realize according to the present invention for providing the context aware body that the user based on context aware pays close attention to information to realize method.

nullAdditionally，As mentioned above，By described process realize according to the present invention for providing the context aware body that the user based on context aware pays close attention to information to realize method，There is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information service to realize method according to the present invention，Wherein，By building the content paid close attention to based on context aware technological prediction user，And more Accurate Prediction user at the situation ontologies model which type of where needs retrieve，And the concern information according to described situation ontologies model prediction user，Retrieval request for user provides only the information that user pays close attention to，Thus more accurately brief retrieval result can be provided，Thus solving to provide substantial amounts of unnecessary retrieval result for the retrieval request of user，Thus forcing user again to filter the problem being present in existing search method of information needed.

Meanwhile, according to the present invention, there is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information to realize method, it utilizes the semantic data mining algorithm (SemanticDataMiningAlgorithm) based on situation ontologies to provide more accurately brief retrieval result, thus the defect of the limited picture of the mobile equipment as smart mobile phone and few internal memory can be overcome.

Especially, according to the present invention, there is provided a kind of for providing the context aware body that the user based on context aware pays close attention to information to realize method, it is while providing for the retrieval result of the retrieval request information of user, utilize the GPS and the map function that are built in mobile equipment, with the position that retrieval result provides the retrieval information needed for the current location of user and user in the lump, thus only passing through primary retrieval can not only be prone to find the content of information needed, additionally it is possible to be prone to find position.

By embodiments of the invention as above to according to the present invention for providing the detailed content that the context aware body that the user based on context aware pays close attention to information realizes method to be illustrated, but the present invention is not limited to be recorded in the content of described embodiment, by the technical staff with general knowledge of the technical field belonging to the present invention, can as required and other a variety of causes can to carry out various correction, change, combination and replacement etc. be natural.

Claims (12)

1. a context aware body realizes method, it also provides for unnecessary retrieval result in order to solve the retrieval request for user to described user, thus forcing user again to filter and selecting the problem being present in existing searching system of information needed, only the concern information of described user is provided as retrieval result based on context aware technology, thus can provide more accurately brief retrieval result than existing searching system, described context aware body realizes method and includes:

User pays close attention to information database construction step, and it collects the information of the region-of-interest for described user, thus building data base；

Body inference step, it arranges the relation and attribute that are stored between each data that described user pays close attention in information database, and described user pays close attention to information database and pays close attention to structure in information database construction step described user；And ontology database construction step, it builds ontology database according to relation and the attribute of each data arranged in described body inference step.

2. context aware body according to claim 1 realizes method, it is characterised in that:

The contextual information of described user is expressed as by described body include user identifier U, spatial information L, temporal information T and time space contextual information O={U, L, T, the S} with the four-dimension with described user-dependent information on services S.

3. context aware body according to claim 2 realizes method, it is characterised in that:

Described temporal information is the information divided by certain time spacing set in advance, described spatial information is the coordinate (xi on belonging Euclidean space, yi) general positional information, and described coordinate represents longitude and the latitude value in area that described user is presently in respectively.

4. context aware body according to claim 3 realizes method, it is characterised in that:

In described database sharing step, concern information as described user, the information to ground of going sightseeing is collected by each area, as most higher level's class designated root class, each area is appointed as subclass, by to described regional go sightseeing ground title be appointed as the subclass to each area described, it is appointed as the subclass to described ground title of going sightseeing by the facility in each sightseeing ground or to the affiliated detailed description going sightseeing ground, each regional or described sightseeing described is arranged to synonym or synonym part of speech (equivalentclass) relation of consonant retrieval, thus building the sightseeing data base constituting hierarchy.

5. context aware body according to claim 4 realizes method, it is characterised in that:

Described sightseeing data base includes multiple territories (field) of id, add_1, add_2, add_3, add_4, add_5, tour_site, syn, lat, long, the address on each ground of going sightseeing of described add_1 territory, described add_2 territory, described add_3 territory, described add_4 territory and described adde_5 domain representation, the storage of described tour_site territory has sightseeing ground title, described syn territory is the territory for synonym or the process of consonant word, and described lat territory and described long territory represent latitude required when arranging map and longitude respectively.

6. context aware body according to claim 5 realizes method, it is characterised in that:

nullIn described body inference step，According to " Place "、“Attraction”、“Resource”、" Activity " determines the attribute of each data，Described " Place " is made up of the title in the area to go sightseeing，Described " Attraction " is made up of the sight-seeing resort title in each area，Described " Resource " is by the building in each sightseeing ground or remains、Traces、Seashore and restaurant are constituted，Described " Activity " includes " Culturalactivity " that represent cultural activity、Represent " Sceneryview " of the behavior viewing and admiring landscape、Represent " Diningout " of the behavior dined out，Thus representing the described user actual the implemented action on each ground of going sightseeing.

7. context aware body according to claim 6 realizes method, it is characterised in that:

In described body inference step, the relation on attributes of described " Place " and described " Attraction " is appointed as " hasAttraction " and " IsAttractionOf ", the relation on attributes of described " Attraction " and described " Resource " is appointed as " hasResource " and " IsResourceOf ", and the relation on attributes of described " Activity " and described " Resource " is appointed as " hasActivity " and " IsActivityOf ".

8. context aware body according to claim 7 realizes method, it is characterised in that:

In described body inference step, when there is multiple relation on attributes in the data constituting each class, it is considered as that there is multiple attributes relation.

9. context aware body according to claim 1 realizes method, it is characterised in that:

In described body inference step, Protege_4.0.2version is utilized to perform.

10. context aware body according to claim 1 realizes method, it is characterised in that:

In described ontology database construction step, OWL, RDF and RDFS is utilized to perform to build the process of described ontology database.

11. the user based on context aware pays close attention to information service provides method, it is characterised in that:

Utilize the context aware body realizing method by the context aware body described in any one of claim 1 to 10 and build, the concern information of user is provided only as retrieval result, thus providing more accurately brief retrieval result than existing search method, therefore, it is more suitable for including the mobile environment of smart mobile phone or panel computer.

12. the user based on context aware pays close attention to information service system, it is characterised in that:

Utilize the context aware body realizing method by the context aware body described in any one of claim 1 to 10 and build, the concern information of user is provided only as retrieval result, thus providing more accurately brief retrieval result than existing searching system, therefore, it is more suitable for including the mobile environment of smart mobile phone or panel computer.