Hyoseop Shin

Adaptive multi-stage distance join processing

A spatial distance join is a relatively new type of operation introduced for spatial and multimedia database applications. Additional requirements for ranking and stopping cardinality are often combined with the spatial distance join in on-line query processing or internet search environments. These requirements pose new challenges as well as opportunities for more efficient processing of spatial distance join queries. In this paper, we first present an efficient k-distance join algorithm that uses spatial indexes such as R-trees. Bi-directional node expansion and plane-sweeping techniques are used for fast pruning of distant pairs, and the plane-sweeping is further optimized by novel strategies for selecting a sweeping axis and direction. Furthermore, we propose adaptive multi-stage algorithms for k-distance join and incremental distance join operations. Our performance study shows that the proposed adaptive multi-stage algorithms outperform previous work by up to an order of magnitude for both k-distance join and incremental distance join queries, under various operational conditions.

Adaptive and incremental processing for distance join queries

A spatial distance join is a relatively new type of operation introduced for spatial and multimedia database applications. Additional requirements for ranking and stopping cardinality are often combined with the spatial distance join in online query processing or Internet search environments. These requirements pose new challenges as well as opportunities for more efficient processing of spatial distance join queries. In this paper, we first present an efficient k-distance join algorithm that uses spatial indexes such as R-trees. Bidirectional node expansion and plane-sweeping techniques are used for fast pruning of distant pairs, and the plane-sweeping is further optimized by novel strategies for selecting a sweeping axis and direction. Furthermore, we propose adaptive multistage algorithms for k-distance join and incremental distance join operations. Our performance study shows that the proposed adaptive multistage algorithms outperform previous work by up to an order of magnitude for both k-distance, join and incremental distance join queries, under various operational conditions.

A Lightweight XML Repository Supporting Dynamic Partial Update of XML Data

Generic methods of storing and querying XML data are often not optimal or adequate for an embedded system with domain-specific requirement. Though embedded systems may have more strict restriction on resource and time constraints, they also pose a new chance of customization and optimization in handling XML data. This is because the characteristics and behaviours of the XML data in a specific domain are likely to be known a priori. In this paper, embedded-type XML repositories are considered where large-scale XML data are split into smaller fragments for secure and efficient delivery from the senders to the receiving devices. The proposed XML repository for the receiving devices deals with the XML data in such a way that the XML fragments can be efficiently reconstructed, searched and retrieved, while dynamic updates and versioning of the XML fragments are supported with marginal costs.

A storage and retrieval method of XML-based metadata in PVR environment

This paper presents a design of a metadata engine that supports storage and retrieval of XML-based metadata in PVR environment. As the XML-based metadata possibly consist of various parts and are described in a large hierarchical XML document, a series of preprocessing steps such as fragmentation, binary encoding, encapsulation, and indexing is performed before they are to be transferred from the sender to the receiver via broadcast networks. The proposed metadata engine is designed so that the cost of accessing the XML-based metadata can be minimized and, at the same time, dynamic updating and versioning for the metadata can be accomplished with moderate cost.

On Storing and Querying XML Data within Network-Enabled Embedded Systems

Modern digital consumer devices are now enabled to handle certain types of XML data. In many cases, such data are periodically downloaded from remote via some network. In this paper, methods of storing and querying of XML data within digital devices are considered where relatively large scale XML data are downloaded in packetized formats for secure and efficient delivery from the senders to the receiving devices. Specifically, the downloaded XML fragments are natively organized by the fragment-oriented data structures and the supporting indexes both of which can be easily implemented on top of a lightweight relational storage. The proposed design provides efficient search and retrieval of XML data as well as dynamic update and versioning within resource-limited digital devices.

Partition-Based Similarity Join in High Dimensional Data Spaces

It is not desirable in the performance perspective of search algorithms to partition a high dimensional data space by dividing all the dimensions. This is because the number of cells resulted from partitioning explodes as the number of partitioning dimensions increases, thus making any search method based on space partitioning impractical. To address this problem, we propose an algorithm to dynamically select partitioning dimensions based on a data sampling method for efficient similarity join processing. Futhermore, a disk-based plane sweeping method is proposed to minimize the cost of joins between the partitioned cells. The experimental results show that the proposed schemes substantially improve the performance of the partition-based similarity joins in high dimensional data spaces.

Tie-breaking strategies for fast distance join processing

The distance join is a spatial join that finds pairs of closest objects in the order of distance by associating two spatial data sets. The distance join stores node pairs in a priority queue, from which node pairs are retrieved while traversing R-trees in top-down manners in the order of distance. This paper first shows that a priority strategy for the tied pairs in the priority queue during distance join processing greatly affects its performance. Then it proposes a probabilistic tie-breaking priority method. The experiments show that the proposed method is always better than alternative methods in the performance perspectives.

Metadata engine for TV-Anytime compliant set-top box

This paper presents a design of metadata engine based on a relational database architecture for a TV-Anytime compliant set-top box. As TV-Anytime metadata is described in XML, it is split into a set of fragments when it is transferred. The proposed metadata engine is designed so that access to fragment metadata can be minimized when responding to metadata requests from various set-top box applications.

An XPath Reduction Algorithm for Fast Querying of Digital Libraries in XML

Digital contents in XML are rapidly increasing on the Web. In our work, we assume that the XML data is stored in a relational database using the popular node numbering scheme and propose a technique to pre-process XPath queries in order to convert them into a much more efficient form that will run several times faster. Our algorithm uses a special data structure called XIP (XML Instance Path) tree of each XML document to reduce a given XPath query (i.e., path expression) into a more simpler form. The benefits of the approach is more significant for long length queries, making it useful for digital libraries storing large and complex structured XML documents for multimedia or business applications.