Atsuyuki Morishima

Efficient evaluation of XML middle-ware queries

We address the problem of efficiently constructing materialized XML views of relational databases. In our setting, the XML view is specified by a query in the declarative query language of a middle-ware system, called SilkRoute. The middle-ware system evaluates a query by sending one or more SQL queries to the target relational database, integrating the resulting tuple streams, and adding the XML tags. We focus on how to best choose the SQL queries, without having control over the target RDBMS.

CyLog/Crowd4U: a declarative platform for complex data-centric crowdsourcing

This demo presents a principled approach to the problems of data-centric human/machine computations with Crowd4U, a crowdsourcing platform equipped with a suite of tools for rapid development of crowdsourcing applications. Using the demo, we show that declarative database abstraction can be used as a powerful tool to design, implement, and analyze data-centric crowdsourcing applications. The power of Crowd4U comes from CyLog, a database abstraction that handles complex data-centric human/machine computations. CyLog is a Datalog-like language that incorporates a principled feedback system for humans at the language level so that the semantics of the computation not closed in machines can be defined based on the game theory. We believe that the demo clearly shows that database abstraction can be a promising basis for designing complex data-centric applications requiring human/machine computations.

A machine learning approach to rapid development of XML mapping queries

We present XLearner, a novel tool that helps the rapid development of XML mapping queries written in XQuery. XLearner is novel in that it learns XQuery queries consistent with given examples (fragments) of intended query results. XLearner combines known learning techniques, incorporates mechanisms to cope with issues specific to the XQuery learning context, and provides a systematic way for the semiautomatic development of queries. We describe the XLearner system. It presents algorithms for learning various classes of XQuery, shows that a minor extension gives the system a practical expressive power, and reports experimental results to demonstrate how XLearner outputs reasonably complicated queries with only a small number of interactions with the user.

SMART: a tool for semantic-driven creation of complex XML mappings

We focus on the problem of data transformations, i.e., how to transform data to another structure to adapt it to new application requirements or given environments. Here, we define data transformation as the process of taking as input two schemas A and B and an instance of A, and producing an instance of B. Today, data transformations are required in many situations: to integrate multiple information sources, to construct and receive data for Web services, and to migrate data from legacy systems to new systems, from local databases to data warehouses. This demonstration focuses on XML transformations, since XML is the de facto standard for data exchange.

Wraplet: Wrapping Your Web Contents with a Lightweight Language

Wrapping of Web sources is known to be one of the key tasks in information integration problems. This paper proposes Wraplet, a wrapping language for extracting structured data from Web contents written in HTML. Unlike existing solutions, Wraplet is designed as a lightweight language in which users can write scripts for wrapping easily with text editors. Its simple syntax and the library of useful patterns help the user write wrapping descriptions by hand. We explain the motivation of its development and the language design and then shows the result of a preliminary experiment about applicability of the language to real Web sources. We conducted a statistical analysis and obtained the result that the applicability of Wraplet is more than 90% at the 95% confidence level in the experimental setting.

A Data Modelling and Query Processing Scheme for Integration of Structured Document Repositories and Relational Databases

Integration of h.eterogen.eous information resources h,as been. one of the most im,portant issues in recent advanced application environments. In addition to conventional databases, structured documents have been recogn.ized as important information resources recently. In this paper, we first present a data m.odel named the NR/SD model as a basic data modeling framework for integration of strzlctured documents and relationsal databases. Then, we discuss ‘the query processing and optimization scheme for ewvironments including structured document repositories and relational databases. The NR/SD model combines an. abstract data type named the structured document type and the nested relational structures, and features operators named converters to dynam.ically convert structured documents into nested relational structures and vice versa. Th.erefore, we can manipulate information in eithei forms of structured documents and relations. This feature poses the following issues in query processing: (1) Utilizution of the local query processing capability of the doc,ument repository and the relation,al database, and (2) Eficient manipulation of structured document data whose volume is potentially quite large. We discuss the query processing and optim.ization scheme mainly focusing on these issues.

A data modeling approach to the seamless information exchange among structured documents and databases

Integration of heterogeneous information resources has been one of the most important issues in recent advanced application environments. In addition to conventional databases, structured documents have been recognized as important information resources recently. In this paper, we present a data model named the NR/SD model, which is used as a basic data modeling framework for the seamless integration of structured documents and relational databases. The NR/SD model combines an abstract data type named the structured document type and the nested relational structures, and features operators named converters to dynamically convert structured documents into nested relational structures and vice versa. Therefore, we can manipulate information in either forms of structured documents and relations. The operators can also be used to develop user views on the stored structured documents and relations. We show data structures and operators in the NR/SD model and its applicability to system federation environments. Some basic properties of the converters are also studied.

Dynamic creation of multimedia Web views on heterogeneous information sources

Integration of the World Wide Web and other information sources is strongly required in the recent advanced application environments. Although information sources contain various types of data objects, the volume of multimedia objects has been increasing drastically. The paper proposes a multimedia integration scheme using SMIL in a mediator based integration system. The scheme achieves dynamic creation of various Web views combining multimedia objects stored in heterogeneous information sources such as the Web, structured documents, and databases. A key point is to utilize SMIL as a framework for dynamic multimedia integration. Since there already exist some players for SMIL documents, dynamic integration results including multimedia objects can be rendered in such players, without special-purpose ones, and are ready to be broadly distributed. The main contributions of the paper are as follows. (1) We propose a dynamic, SMIL-based multimedia integration scheme. This scheme utilizes an extended relational data model named WebNR/SD. (2) We report on implementation of the proposed scheme. We have already implemented the first prototype system which realizes integration of heterogeneous information sources including multimedia objects. Actually, it can construct SMIL-based multimedia views dynamically. This shows the practical significance of our scheme.

A Rule-Oriented Architecture to Incorporate Dissemination-Based Information Delivery into Information Integration Environments

Integration of heterogeneous information sources has been one of important research issues in recent advanced application environments. Today, various types of information sources are available. Dissemination-based information delivery services that autonomously deliver information from the server sites to users are among the useful and promising information sources. In this paper, we present incorporation of dissemination-based information delivery into information integration environments. The integration here has two goals: (1) Users can utilize dissemination-based information services as other information sources such as databases and the Web. Namely, they can be sources of information integration. (2) Users can obtain integrated information through dissemination-based delivery. We explain this requirement can be met by a combination of an information integration engine and event-driven rule processing scheme. We also explain prototype system development.

A Support Tool for XML Schema Matching and Its Implementation

We are developing a software tool to support schema matching for data transformations. Schema matching is the process of finding relationships between components of two given database schemas. The tool is unique in that it first extracts conceptual schemas from the two database schemas and allows the user to use the extracted conceptual schemas as clues to perform schema matching. This paper overviews the tool and explains its implementation.