• Journal of KIISE:Databases
• /
• v.33 no.2
• /
• pp.188-200
• /
• 2006

Once a continuous query, which is commonly used in sensor networks, is issued, the query is executed many times with a certain interval and the results of those query executions are collected to the base station. Since this comes many communication messages continuously, it is important to reduce communication cost for collecting data to the base station. In sensor networks, in-network processing reduces the number of message transmissions by partially aggregating results of an aggregate query in intermediate nodes, or merging the results in one message, resulting in reduction of communication cost. In this paper, we propose a routing tree for sensor nodes that qualify the given query predicate, called the query specific routing tree(QSRT). The idea of the QSRT is to maximize in-network processing opportunity. A QSRT is created seperately for each query during dissemination of the query. It is constructed in such a way that during the collection of query results partial aggregation and packet merging of intermediate results can be fully utilized. Our experimental results show that our proposed method can reduce message transmissions more than 18% compared to the existing one.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.1
• /
• pp.126-132
• /
• 2011

In the application areas of LBS(Location Based Service) and ITS(Intelligent Transportation System), continuous k-nearest neighbor query(CkNN) which is defined as a query to find the nearest points of interest to all the points on a given path is widely used. It is necessary to acquire results quickly in the above applications and be applicable to spatial network databases. It is also able to cope successfully with frequent updates of POI objects. This paper proposes a new method to search nearest POIs for moving query objects on the spatial networks. The method produces a set of split points and their corresponding k-POIs as results with partial order among k-POIs. The results obtained from experiments with real dataset show that the proposed method outperforms the existing methods. The proposed method achieves very short processing time(15%) compared with the existing method.

• Journal of KIISE:Databases
• /
• v.31 no.5
• /
• pp.540-552
• /
• 2004

We propose XIR-Linear, a novel method for processing partial match queries on large-scale heterogeneous XML documents using information retrieval (IR) techniques. XPath queries are written in path expressions on a tree structure representing an XML document. An XPath query in its major form is a partial match query. The objective of XIR-Linear is to efficiently support this type of queries for large-scale documents of heterogeneous schemas. XIR-Linear has its basis on the schema-level methods using relational tables and drastically improves their efficiency and scalability using an inverted index technique. The method indexes the labels in label paths as key words in texts, and allows for finding the label paths that match the queries far more efficiently than string match used in conventional methods. We demonstrate the efficiency and scalability of XIR-Linear by comparing it with XRel and XParent using XML documents crawled from the Internet. The results show that XIR-Linear is more efficient than both XRel and XParent by several orders of magnitude for linear path expressions as the number of XML documents increases.

• The KIPS Transactions:PartD
• /
• v.8D no.4
• /
• pp.313-326
• /
• 2001

The XML that was proposed as a next-generation standard for describing Web documents is widely used in various Web-based applications. In addition, XML documents on the Web link each other by hyperlinks. The current works on XML focus on the XML storage system that can efficiently store, manage, and retrieve XML documents. However, the research on the query language that supports the XML links and on the XML retrieval systems to process the XML links, is little conducted until now. In this paper, we propose an extension of an XML query language for expressing the XML link query and its processing scheme. A link query is to retrieve contents from an XML document (a query document) and from the XML documents (referenced documents) that are referred to by the links in the query document. As far as retrieving from the referenced documents is concerned, the current practice is to manually generate queries to get the partial results, and to repeat such a procedure. The purpose of link query processing in this paper is to eliminate the manual work altogether in getting the complete query result. The performance analysis shows that our link query processing strategy outperforms the conventional approach including the manual tasks. The more links to the referenced documents and the more referenced documents there are in the site storing the query document, the more query processing time decreases.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.5
• /
• pp.1095-1101
• /
• 2012

As the amount of data increases explosively, a large scale database system is emerged to store, retrieve and manipulate it. There are several issues in this environments such as, consistency, availability and fault tolerance. In this paper, we address a efficient range-query method where data management services are separated from transaction management services in large-scale database systems. A study had been proposed using partitions to protect independence of two modules and to resolve the phantom problem, but this method was efficient only when range-query is specified by a key. So, we present a new method that can improve the efficiency when range-query is specified by a key attribute as well as other attributes. The presented method can guarantee the independence of separated modules and alleviate overheads for range-query using partial index.

• Journal of KIISE:Databases
• /
• v.30 no.1
• /
• pp.80-94
• /
• 2003

In this paper. we propose an enhanced concurrency control algorithm that minimizes the query delay efficiently. The factors that delay search operations and deteriorate the concurrency of index structures are node splits and MBR updates in multi dimensional index structures. In our algorithm, to reduce the query delay by split operations, we optimize exclusive latching time on a split node. It holds exclusive latches not during whole split time but only during physical node split time that occupies small part of whole split time. Also to avoid the query delay by MBR updates we introduce partial lock coupling(PLC) technique. The PLC technique increases concurrency by using lock coupling only in case of MBR shrinking operations that are less frequent than MBR expansion operations. For performance evaluation, we implement the proposed algorithm and one of the existing link technique-based algorithms on MIDAS-III that is a storage system of a BADA-III DBMS. We show through various experiments that our proposed algorithm outperforms the existing algorithm In terms of throughput and response time.

• Journal of KIISE:Databases
• /
• v.32 no.4
• /
• pp.452-472
• /
• 2005

We propose XIR-Branching, a novel method for processing partial match queries on heterogeneous XML documents using information retrieval(IR) techniques and novel instance join techniques. A partial match query is defined as the one having the descendent-or-self axis '//' in its path expression. In its general form, a partial match query has branch predicates forming branching paths. The objective of XIR-Branching is to efficiently support this type of queries for large-scale documents of heterogeneous schemas. XIR-Branching has its basis on the conventional schema-level methods using relational tables(e.g., XRel, XParent, XIR-Linear[21]) and significantly improves their efficiency and scalability using two techniques: an inverted index technique and a novel prefix match join. The former supports linear path expressions as the method used in XIR-Linear[21]. The latter supports branching path expressions, and allows for finding the result nodes more efficiently than containment joins used in the conventional methods. XIR-Linear shows the efficiency for linear path expressions, but does not handle branching path expressions. However, we have to handle branching path expressions for querying more in detail and general. The paper presents a novel method for handling branching path expressions. XIR-Branching reduces a candidate set for a query as a schema-level method and then, efficiently finds a final result set by using a novel prefix match join as an instance-level method. We compare the efficiency and scalability of XIR-Branching with those of XRel and XParent using XML documents crawled from the Internet. The results show that XIR-Branching is more efficient than both XRel and XParent by several orders of magnitude for linear path expressions, and by several factors for branching path expressions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.7
• /
• pp.3172-3193
• /
• 2018

Recently, several studies have shown that linear representation based approaches are very effective and efficient for image classification. One of these linear-representation-based approaches is the Collaborative representation (CR) method. The existing algorithms based on CR have two major problems that degrade their classification performance. First problem arises due to the limited number of available training samples. The large variations, caused by illumintion and expression changes, among query and training samples leads to poor classification performance. Second problem occurs when an image is partially noised (contiguous occlusion), as some part of the given image become corrupt the classification performance also degrades. We aim to extend the collaborative representation framework under limited training samples face recognition problem. Our proposed solution will generate virtual samples and intra-class variations from training data to model the variations effectively between query and training samples. For robust classification, the image patches have been utilized to compute representation to address partial occlusion as it leads to more accurate classification results. The proposed method computes representation based on local regions in the images as opposed to CR, which computes representation based on global solution involving entire images. Furthermore, the proposed solution also integrates the locality structure into CR, using Euclidian distance between the query and training samples. Intuitively, if the query sample can be represented by selecting its nearest neighbours, lie on a same linear subspace then the resulting representation will be more discriminate and accurately classify the query sample. Hence our proposed framework model the limited sample face recognition problem into sufficient training samples problem using virtual samples and intra-class variations, generated from training samples that will result in improved classification accuracy as evident from experimental results. Moreover, it compute representation based on local image patches for robust classification and is expected to greatly increase the classification performance for face recognition task.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.669-672
• /
• 2009

The techniques extract shape descriptors from 3D models and use these descriptors for indices for comparing shape similarities. Most similarity search techniques focus on comparisons of each individual 3D model from databases. However, our similarity search technique can compare not only each individual 3D model, but also partial shape similarities. The partial shape matching technique extends the user's query request by finding similar parts of 3D models and finding 3D models which contain similar parts. We have implemented an experimental partial shape-matching search system for 3D pagoda models, and preliminary experiments show that the system successfully retrieves similar 3D model parts efficiently.

• The KIPS Transactions:PartD
• /
• v.12D no.5 s.101
• /
• pp.689-698
• /
• 2005

Various systems are used to provide the location based services. But, the existing systems have some problems which have difficulties in dealing with faster services for above million people. In order to solve it, a multi-level DBMS which supports both fast data processing and large data management support should be used. The multi-level DBMS with snapshots has all the data existing in disk database and the data which are required to be processed for fast processing are managed in main memory database as snapshots. To optimize performance of this system for location based services, the query classification component which classifies the queries for efficient snapshot usage is needed. In this paper, the query classification component in multi-level DBMS for location based services is designed and implemented. The proposed component classifies queries into three types: (1) memory query, (2) disk query, (3) hybrid query, and increases the rate of snapshot usage. In addition, it applies division mechanisms which divide aspatial and spatial filter condition for partial snapshot usage. Hence, the proposed component enhances system performance by maximizing the usage of snapshot as a result of the efficient query classification.