KR20090077258A - An information filtering system and method through context-aware service - Google Patents

Abstract

An information filtering system and method through context recognition is provided to filter a service suitable for a corresponding context through the surrounding environment and context information of a user, thereby providing a valuable service to the user. A weight/correlation operator(300) applies weight to collected context information collected according to the significance in the evaluation of context recognition, and measures the correlation between the context information. A context information modeler(400) performs the modeling of basic property and other properties which are classified by the weight/correlation operator and of which the correlation is measured by the weight/correlation operator by using OWL(Web Ontology Language). A service filter module(500) performs pattern matching among the weight and correlation of the modeled context information and user recommended service candidates to filter services more than a threshold value.

Description

Information filtering system and method through context awareness {AN INFORMATION FILTERING SYSTEM AND METHOD THROUGH CONTEXT-AWARE SERVICE}

The present invention relates to a system and method for modeling and filtering situation information so as to grasp newly generated data due to a ubiquitous environment, personal information related thereto, and even an individual's situation and provide an appropriate service to the situation.

If the situation is information indicating a user's request, the situation recognition system refers to a system that provides an appropriate service based on this information. The requirements of the context awareness system are initially Schilit's interface to facilitate the selection of objects near you as a component of the context awareness system, and the addition / deletion of new service components and changes in new contexts. They've been categorized into: automatic contextual configuration, contextual information and commands, and contextual triggered actions that specify the relationship in which the contextual system should work. Since then, Pascoe has classified the elements necessary for context awareness into contextual sensing, contextual adaptation, contextual resource discovery, and contextual augmentation.

Conventional techniques related to situational awareness include identifier-based server models (Auto-ID and uID). Auto-ID basically assigns a unique key to the various objects in the user's real life using the serial number of the electronic product code (EPC), and operates the Auto-ID center for the related standardization and technical infrastructure. At the center of the Auto-ID Center is SAVANT, which uses the XML Naming Language (ONS) and the Physical Markup Language (PML) to convert EPC data read through the EPC reader into the EPC domain name.

A technology related to situational awareness is the ubiquitous middleware model (Gaia, Aura), and Gaia on the Champagne Campus, Urbana, Illinois, USA. to be. Gaia provides five basic services for managing components: core management, event management, context management, context file management, presence services, and space storage. Gaia provides context management and context file management services for context awareness. In order to describe the situation sensed by the sensor, it is described in the form of a context similar to the English grammar format, converted to XML (Extensible Markup Language) format, and then in the form of a virtual directory according to the type of situation. Filesystem middleware stores and manages them.

The Aura project at Carnegie Mellon University in the United States models the user's environment as an abstract entity called Aura and consists of several supporting components such as Context Observer and Task Manager. Context Observer provides context awareness service continuously monitoring user and environment changes. Ubi-UCAM, presented by Woo Un-Taek of Gwangju Institute of Science and Technology, is a situation integrator for integrating situation-aware application services. It is delivered in the form of a manager, determines the service, and acts as a handler.

Typical situational awareness systems include Gaia at Illinois University, SOCAM at the National University of Singapore, and Scarlet at Illinois Institute of Technology.

 Gaia is a context-aware service project that a Context Provider collects contextual information from other sensors or other data sources and provides to applications. Context Synthesizer infers and abstracts context information collected from Context Provider into higher level context information and provides it to the application. Context Provider Lookup Service finds Context Provider that provides context information. Context History is a database that records previous context information. Context Consumer is an application service that uses context information.

Service-Oriented Context-aware Middleware (SOCAM) is a middleware project for context-aware mobile services that has been proposed to facilitate the development of context-aware services and systems. OWL (Web Ontology Language) was used for contextual information modeling in middleware. Components used in SOCAM are composed of Context Providers, Context Interpreter, Context Database, Context-Aware Services, and Service Locating Service.

Scarlet researched the transfer of context information between heterogeneous platforms. Scarlet uses SOAP and WSDL to maintain cross platform compatibility.

However, conventional systems have overlooked the feature that individual situation information has different weights for representing the user's situation. The weighting of situation information by individual situation is an important factor in situation analysis.

The present invention is designed to provide an appropriate service through accurate situation awareness,

A context collector configured to collect context information of the user when a service request of the user is received; A weight / association operator for analyzing the importance of context information; Context Modeler for modeling collected contextual information into basic and other attributes; It includes a service filter module (Service Filter Module) for filtering a service suitable for the user's context information,

The purpose of this service is to provide valuable services by filtering the services that are most suitable for the situation through the user's surrounding environment and situation information.

The present invention is to develop a system for recognizing a situation through context information and filtering the service accordingly. Therefore, the present invention classifies the situation in order to recognize the situation in the situation information, calculates the DOA weight of the classified situation information and analyzes and models the relationship between the situation information. In addition, it analyzes the context information and the threshold of DOA weights and DOA weights to filter out services appropriate to the situation.

The information filtering system through the situation recognition according to the present invention having the above configuration and operation can recognize the situation by analyzing the context information to which the service user belongs, and thus can be used to filter the service appropriate to the situation.

The present invention provides a context collector (Context Collector) for collecting the user's context information when the user's service request; A weight operator that performs DOA estimation to analyze the importance of contextual information and analyzes the degree of association between basic and other attributes; Context Modeler for modeling collected contextual information into basic and other attributes; It consists of a service filter module (Service Filter Module) for filtering a service suitable for the user's context information,

According to the present invention, in a ubiquitous environment based on a wireless network environment and a sensor network, a user's situation is changed from a web environment, which is an environment where an individual uses a service, and a system environment that is used by a change in a system environment, such as a position shift or a time change, is used. By applying the features that the user's desired service changes from time to time to the personalized service, it is possible to find and provide the best service for the user who wants to use the service. In other words, in order to enable the provision of personalized service, modeling individual situation information and weighting the modeled situation information according to the influence of situation discrimination, and through this, dynamic information filtering through situation recognition to find and provide a suitable service. System and method.

Hereinafter, exemplary embodiments of the XML digital signature system and method according to the present invention will be described in detail with reference to the accompanying drawings.

1 is an overall configuration diagram of an information filtering system through situational awareness according to the present invention. As shown in Figure 1, the configuration of the information filtering system through the situation recognition, the situation information collector 200 for collecting the situation information of the user 100 when the user's service request comes; Performing a DOA (Degree Of Acceptance) inference to select and weight the situation information affecting the recognition of the situation from the collected situation information, and a weight / association operator 300 for the association between the basic attribute and other attributes; A situation information modeler (400) for classifying situation information weighted by the weight / association operator into a basic attribute and other attributes; A service filter module 500 for filtering a service suitable for the context information of the user by performing pattern matching on the context information model and the user service list, and the service information 600 provided to the user.

The user 100 requests a request for using the service to the information filtering system through situational awareness. However, such a request is not only performed by a user's explicit request, but may also be periodically transmitted by an information device owned by the user. Examples of the user's information device refer to various types of devices that can handle information such as sensors in addition to mobile phones, personal digital assistants (PDAs), and global positioning systems (GPSs), which are widely used.

The contextual information collector 200 serves as a simple collection and transmission that receives various kinds of contextual information transmitted from the user's information device and transmits the contextual information to the weight / association operator. The contextual information delivered through the contextual information collector is modeled after weighting through the weight / association operator.

The weight / association operator 300 classifies information for determining each situation by assigning weights according to the importance of the situation information and analyzes the relationship between the situation information. First, the weight / association operator performs an operation for classifying context information into basic attributes and other attributes. 2 is an algorithm for distinguishing a basic attribute from other attributes. If enough calculation results are obtained from this, situation information corresponding to the basic user attributes can be classified. The rest of the contextual information is classified into subclass attributes for the corresponding basic attributes by comparing their association with each base attribute, and given priority according to the degree of association. Weight according to the classification of the situation information is shown in FIG. At this time, the distinction between basic attribute and other attributes is directly set by experts in the field where the situation recognition system is used. Therefore, basic attributes change according to the situation.

The contextual information modeler 400 models the contextual information classified into basic attributes and other attributes by a weight / association operator using OWL (Web Ontology Language). At this time, basic attribute means context information which is used for situation recognition. On the other hand, the other attribute is information that can be used depending on the degree of association with the basic attribute, and serves as an auxiliary context information of the basic attribute. The context information corresponding to the basic attributes contains contextual information that must be considered according to the characteristics of each context. However, there is no guarantee that the contextual information corresponding to the basic attributes is sufficient to recognize the situation. In this case, the dependence on other properties is further increased. In order to provide accurate services to the user at the moment when the user requests a service or the user determines that the service is needed, various situation information of the user is added in addition to the basic properties. Because it is necessary. 4 is a contextual information representation of the contextual information modeler.

The service filter module 500 performs pattern matching between the user's context information and the service, digitizes the degree of association with the service or information requested by the user by DOA estimation, and compares the threshold with the threshold. Here, the threshold means statistically the level of service or information that users are satisfied with. 5 is a diagram illustrating a process of finding an appropriate service by comparing a DOA estimation of context information with a threshold value of a service desired by a user. The threshold is set in two parts. It is divided into basic attributes and other attributes. First, pattern matching is performed for the basic attributes and the highest weight is used, so the threshold is also set based on the basic attributes. After that, when a situation arises where a service or information should be provided at the request of a user, the DOA of the basic attribute is first estimated. Next, the threshold of the service or information satisfying the estimated DOA is determined, and the DOA of the second weighted other attribute is estimated to determine the priority to be provided to the user. FIG. 5 recalculates the weighted miscellaneous attributes and the result calculated from the basic attributes after the DOA estimation. This operation is repeated as long as the priority is not repeated. Finally, the service or information is provided to the user when the priority is determined. However, if no service or information has a threshold that satisfies the user context information DOA estimation result, the service or information provided to the user is considered to be unreliable. The service or information provided to the user then estimates the DOA of the second highest weighted attribute that is directly related to the base attribute and finds the service or information with the appropriate threshold and provides it to the user.

The service information 600 provided to the user is a service filtered by the service filter module and is delivered to the user 100 as a list of services suitable for the user to use in the current situation.

6 is a flowchart of information filtering through situation recognition according to the present invention. As shown in Figure 5, the present invention uses a system for modeling the situation information and filtering by the weight of the information,

Transmitting the situation information by the user and the user's information device 100 (S100); A situation information collection step (S110) of receiving and collecting the situation information transmitted by the user; A weighting / correlation operation (S120) for weighting the contextual information according to the influence of the contextual awareness and measuring the degree of association between the contextual information; A situation information modeling step of classifying and classifying the classified and weighted situation information into a basic attribute and other attributes (S130); A service filtering request step (S140) for filtering a service by comparing a weight DOA of the context information with a threshold of the service; It includes a service providing step (S150) for providing the user with a service 600 suitable for the user's situation that is the result processed by the service filter module.

The association operation processing performed in the weighting / association operation step S120 gives weights according to the importance of the situation information and filters the situation information according to each situation. When the weighting technique is applied, it is divided into basic attributes and other attributes. Basic attributes are factors that are always considered when filtering situation information. Pattern filtering on basic attributes is first performed during filtering operation. If enough calculation results are found here, the sub-class attributes directly related to the context information corresponding to the user basic attributes are recalculated, and the priority is given in the order of highest relevance. If the result of the operation on the base attribute is not sufficient, pattern matching is performed on all classes that are directly or indirectly related to the base attribute, and the user is provided in the order of high relevance. At this time, basic attributes and other attributes are directly set by experts in the field where situational awareness systems are used.

In the service filtering performed by the service filtering request step (S140), pattern matching between the user's context information and the service is performed to quantify the degree of association with the service or information requested by the user through DOA estimation and compare with a threshold. In the processing for this, the pattern matching is performed first and the weight is the highest, so the threshold value is set around the basic attribute. After that, when a situation arises where a service or information should be provided at the request of a user, the DOA of the basic attribute is first estimated. Next, the threshold of the service or information satisfying the estimated DOA is determined, and the DOA of the second weighted other attribute is estimated to determine the priority to be provided to the user. The other attributes also recalculate the result computed from the base attribute after DOA estimation. This operation is repeated as long as the priority is not repeated. Finally, the service or information is provided to the user when the priority is determined. However, if there is no service or information with a threshold that satisfies the user context information DOA estimation result, the service or information provided to the user is considered to be unreliable. The service or information provided to the user then estimates the DOA of the second highest weighted attribute that is directly related to the base attribute and finds the service or information with the appropriate threshold and provides it to the user.

1 is an overall configuration diagram of an information filtering system through situational awareness according to the present invention

2 is a basic attribute and other attribute classification algorithm according to the present invention

3 is a modeling structure in which the contextual information according to the present invention is divided into basic attributes and other attributes.

4 is a list of weights of basic attribute and other attribute context information according to the present invention.

5 is a DOA estimate and service threshold processing diagram according to the present invention.

6 is a flowchart of information filtering through situation recognition according to the present invention.

Claims (3)

In the situation information recognition system, A situation information collector 200 for collecting situation information transmitted by the user and the user's information device 100; A weighting / association operator 300 which weights the contextual information collected from the contextual information collector according to importance in evaluating contextual awareness and measures the degree of association between the contextual information; A situation information modeler 400 classifying the basic attributes and other attributes classified by the weight / association operator 300 and measuring the degree of association using OWL; A service filter module 500 for filtering a service having a threshold value or more by performing pattern matching between weights and association degrees of situation information modeled by the situation information modeler 400 and user recommendation service candidates; A service list 600 filtered by the service filter module 500; Information filtering system through situation recognition, characterized in that consisting of The method of claim 1, The weight / association operator 200, Information filtering system through situation recognition, characterized in that the situation information data collected from the user and the user's information device 100 is weighted by applying DOA (Degree Of Acceptance) technique The method of claim 1, The service filter module 500, System for filtering information to be provided by comparing the service threshold of the service candidates to be provided to the user and the DOA of the contextual information modeled by the contextual information modeler 400, the information filtering system through the situational awareness

Images