Jaeyoung Yang

MORPHEUS: a more scalable comparison-shopping agent

This paper proposes a more scalable comparison-shopping agent named MORPHEUS. MORPHEUS presents simple but robust inductive learning algorithm that automatically constructs wappers.

Topic-Specific Web Content Adaptation to Mobile Devices

Mobile content adaptation is a technology of effectively representing the contents originally built for the desktop PC on wireless mobile devices. Previous approaches for Web content adaptation are mostly device-dependent. Also, the content transformation to suit to a smaller device is done manually. As a result, the user has difficulty in selecting relevant information from a heavy volume of contents since the context information related to the content is not provided. To resolve these problems, this paper proposes an enhanced method of Web content adaptation for mobile devices. In our system, the process of Web content adaptation consists of 4 stages including block filtering, block title extraction, block content summarization, and personalization through learning. As a result of learning, personalization is realized by showing the information for the relevant block at the top of the content list

A Shopping Agent That Automatically Constructs Wrappers for Semi-Structured Online Vendors

This paper proposes a shopping agent with a robust inductive learning method that automatically constructs wrappers for semi-structured online stores. Strong biases assumed in many existing systems are weakened so that the real stores with reasonably complex document structures can be handled. Our method treats a logical line as a basic unit, and recognizes the position and the structure of product descriptions by finding the most frequent pattern from the sequence of logical line information in output HTML pages. This method is capable of analyzing product descriptions that comprise multiple logical lines, and even those with extra or missing attributes. Experimental tests on over 60 sites show that it successfully constructs correct wrappers for most real stores.

ScalableWeb News Adaptation To Mobile Devices Using Visual Block Segmentation for Ubiquitous Media Services

This paper describes an enhanced method of Web content adaptation to mobile devices for online News article provision in ubiquitous environments. Our system exploits a scheme of visual block segmentation for Web pages that filters out unnecessary blocks and extracts useful article information from content blocks. This method resolves the problems of previous approaches to Web content adaptation in which the content transformation to suit to a smaller device is device-dependent and manually-driven. Our method also employs a learning module that is initiated when the user selects to view the full content in the content summary page. As a result of learning, personalization is realized by showing the information for the relevant block at the top of the content list. A series of experiments are performed to evaluate our mobile content adaptation method for a number of well-known Web News sites, and the result of evaluation is satisfactory both in block filtering accuracy and in user satisfaction by personalization.

An interface agent for wrapper-based information extraction

This paper proposes a new method of building information extraction rules for Web documents by exploiting a user interface agent that combines the manual and automatic approaches of rule generation. We adopt the scheme of supervised learning in which the interface agent is designed to get information from the user regarding what to extract from a document and XML-based wrappers are generated according to these inputs. The interface agent is used not only to generate new extraction rules but also to modify and extend existing ones to enhance the precision and the recall measures of Web information extraction systems. We have done a series of experiments to test the system, and the results are very promising.

Knowledge-Based Wrapper Induction for Intelligent Web Information Extraction

This chapter discusses some of the issues in Web information extraction, focusing on automatic extraction methods that exploit wrapper induction. In particular, we point out the limitations of traditional heuristic-based wrapper-generation systems, and as a solution, we emphasize the importance of the domain knowledge in the process of wrapper generation. We demonstrate the effectiveness of domain knowledge by presenting our scheme of knowledge-based wrapper generation for semistructured and labeled documents. Our information-extraction system, XTROS, represents both the domain knowledge and the wrappers using XML documents to increase modularity, flexibility, and interoperability. XTROS shows good performance on several Web sites in the domain of real estate, and it is expected to be easily adaptable to different domains by plugging in appropriate XML-based domain knowledge.

A Knowledge-Based Information Extraction System for Semi-structured Labeled Documents

This paper presents a scheme of knowledge-based wrapper generation for semi-structured and labeled documents. The implementation of an agent-oriented information extraction system, XTROS, is described. In contrast with previous wrapper learning agents, XTROS represents both the domain knowledge and the wrappers by XML documents to increase modularity, flexibility, and interoperability. XTROS shows good performance on several Web sites in the domain of real estate, and it is expected to be easily adaptable to different domains by plugging in appropriate XML-based domain knowledge.

A focused crawler with document segmentation

The focused crawler is a topic-driven document-collecting crawler that was suggested as a promising alternative of maintaining up-to-date Web document indices in search engines. A major problem inherent in previous focused crawlers is the liability of missing highly relevant documents that are linked from off-topic documents. This problem mainly originated from the lack of consideration of structural information in a document. Traditional weighting method such as TFIDF employed in document classification can lead to this problem. In order to improve the performance of focused crawlers, this paper proposes a scheme of locality-based document segmentation to determine the relevance of a document to a specific topic. We segment a document into a set of sub-documents using contextual features around the hyperlinks. This information is used to determine whether the crawler would fetch the documents that are linked from hyperlinks in an off-topic document.

Detecting Collaborative Fields Using Social Networks

It is generally difficult for researchers to obtain information related to their own fields and novel technologies from huge data residing in the World Wide Web. Furthermore, they often try to apply them to other particular fields which are different from theirs. The main motivation of this phenomenon is to solve existing problems or improve the performance of their systems. Hence, it is important to detect collaborative fields in which technologies of particular fields are applied to another area to find various trends. In this paper, we propose a method to detect collaborative fields by using social networks representing the relations among authors of papers, and describe some experimental results to show the effectiveness of the proposed method when collaborative fields are detected by using social networks.

Agents for Intelligent Information Extraction by Using Domain Knowledge and Token-Based Morphological Patterns

Knowledge-based information extraction is known to have flexibility in recognizing various kinds of target information by exploiting the domain knowledge to automatically generate information-extraction rules. However, most of previous knowledge-based information-extraction systems are only applicable to labeled documents, and as a result, ontology terms must appear in the document in order to guide the system to determine the existence of the target information.