Journal of Intelligence and Information Systems,
Vol. 10, No. 2, November 2004
Multilingual Product Retrieval Agent through Semantic Web and Semantic Networks
Yoo-Jin Moon
Vol. 10, No. 2, Page: 1 ~ 13
Keywords : Multilingual Product Retrieval, XML, Semantic Web, E-commerce, UNSPSC
This paper presents a method for the multilingual product retrieval agent through XML and the semantic networks in e-commerce. Retrieval for products is an important process, since it represents interfaces of the customer contact to the e-commerce. Keyword-based retrieval is efficient as long as the product information is structured and organized. But when the product information is expressed across many online shopping malls, especially when it is expressed in different languages with cultural backgrounds, buyers' product retrieval needs language translation with ambiguities resolved in a specific context. This paper presents a RDF modeling case that resolves semantic problems in the representation of product information and across the boundaries of language domains. With adoption of UNSPSC code system, this paper designs and implements an architecture for the multilingual product retrieval agents. The architecture is based on the central repository model of product catalog management with distributed updating processes. It also includes the perspectives of buyers and suppliers. And the consistency and version management of product information are controlled by UNSPSC code system. The multilingual product names are resolved by semantic networks, thesaurus and ontology dictionary for product names.
Performance Comparison of Clustering Techniques for Spatio-Temporal Data
Nayoung Kang, Juyoung Kang, and Hwan-Seung Yong
Vol. 10, No. 2, Page: 15 ~ 37
Keywords : Data Mining, Spatio-Temporal Data Mining, Clustering, Performance Evaluation
With the growth in the size of datasets, data mining has recently become an important research topic. Especially, interests about spatio-temporal data mining has been increased which is a method for analyzing massive spatio-temporal data collected from a wide variety of applications like GPS data, trajectory data of surveillance system and earth geographic data. In the former approaches, conventional clustering algorithms are applied as spatio-temporal data mining techniques without any modification. In this paper, we focused to SOM that is the most common clustering algorithm applied to clustering analysis in data mining wet and develop the spatio-temporal data mining module based on it. In addition, we analyzed the clustering results of developed SOM module and compare them with those of K-means and Agglomerative Hierarchical algorithm in the aspects of homogeneity, separation, separation, silhouette width and accuracy. We also developed specialized visualization module fur more accurate interpretation of mining result.
A Hybrid Product Design System for Financial Product Factory
Seong-ha Lee, Jung-eun Ju, Seong-cheol Choi, and Sang-hoe Koo
Vol. 10, No. 2, Page: 39 ~ 51
Product factory is a real-time financial product design system for the Internet customers. The hybrid product is a product taking combined characteristics of two different products. Hybrid product factory is a product factory that designs hybrid products from two different products based on both business rules and customer requirements. Though the importance of product factory is emphasized in the industry, there has not been much research peformed regarding product factory. In this research, we developed a product factory system that designs hybrid products. To design a hybrid product, it is necessary to have a method to combine attributes and values of two different products, and a method to control the combining operations to properly reflect business requirements. In this research, we developed low different combining operators and business rule representations. rn addition, to prove the effectiveness of this methods, we implemented a prototypical system and demonstrated on cases regarding financial loan products.
Generalization of Recurrent Cascade Correlation Algorithm and Morse Signal Experiments using new Activation Functions
Sang-Wha Lee, and Hae-Sang Song
Vol. 10, No. 2, Page: 53 ~ 63
Keywords : RCC, Second order RCC, sigmoid, new activation function
Recurrent-Cascade-Correlation(RCC) is a supervised teaming algorithm that automatically determines the size and topology of the network. RCC adds new hidden neurons one by one and creates a multi-layer structure in which each hidden layer has only one neuron. By second order RCC, new hidden neurons are added to only one hidden layer. These created neurons are not connected to each other. We present a generalization of the RCC Architecture by combining the standard RCC Architecture and the second order RCC Architecture. Whenever a hidden neuron has to be added, the new RCC teaming algorithm automatically determines whether the network topology grows vertically or horizontally. This new algorithm using sigmoid, tanh and new activation functions was tested with the morse-benchmark-problem. Therefore we recognized that the number of hidden neurons was decreased by the experiments of the RCC network generalization which used the activation functions.
Front Classification using Back Propagation Algorithm
Minchul Jung
Vol. 10, No. 2, Page: 65 ~ 77
Keywords : font classification, optical character recognition(OCR), Artificial neural network, Chain code
This paper presents a priori and the local font classification method. The font classification uses ascenders, descenders, and serifs extracted from a word image. The gradient features of those sub-images are extracted, and used as an input to a neural network classifier to produce font classification results. The font classification determines 2 font styles (upright or slant), 3 font groups (serif sans-serif or typewriter), and 7-font names (Postscript fonts such as Avant Garde, Helvetica, Bookman, New Century Schoolbook, Palatine, Times, and Courier). The proposed a priori and local font classification method allows an OCR system consisting of various font-specific character segmentation tools and various mono-font character recognizers. Experiments have shown font classification accuracies reach high performance levels of about 95.4 percent even with severely touching characters. The technique developed for tile selected 7 fonts in this paper can be applied to any other fonts.
Classification of e-mail Using Dynamic Category Hierarchy and Automatic Category Generation
Sun Park, Chan Min Ahn, Sang Ho Park, Ju-Hong Lee, and Bum-Ghi Choi
Vol. 10, No. 2, Page: 79 ~ 89
Keywords : E-mail classification, dynamic category hierarchy, automatic classification
Since the amount of E-mail messages has increased , we need a new technique for efficient e-mail classification. E-mail classifications are grouped into two classes: binary classification, multi-classification. The current binary classification methods are mostly spm mail classification methods which are based on rule driven, bayesian, SVM, etc. The current multi- classification methods are based on clustering which groups e-mails by similarity. In this paper, we propose a novel method for e-mail classification. It combines the automatic category generation method based on the vector model and the dynamic category hierarchy construction method. This method can multi-classify e-mail automatically and manage a large amount of e-mail efficiently. In addition, this method increases the search accuracy by dynamic reclassification of e-mails.
A Hybrid Approach Using Case-Based Reasoning and Fuzzy Logic for Corporate Bond Rating
Hyun-jung Kim, and Kyung-shik Shin
Vol. 10, No. 2, Page: 91 ~ 109
Keywords : Fuzzy sets, Case-based reasoning, Corporate bond rating
This study investigates the effectiveness of a hybrid approach using fuzzy sets that describe approximate phenomena of the real world. Compared to the other existing techniques, the approach handles inexact knowledge in common linguistic terms as human reasoning does it. Integration of fuzzy sets with case-based reasoning (CBR) is important in that it helps to develop a successful system far dealing with vague and incomplete knowledge which statistically uses membership value of fuzzy sets in CBR. The preliminary results show that the accuracy of the integrated fuzzy-CBR approach proposed for this study is higher that of conventional techniques. Our proposed approach is applied to corporate bond rating of Korean companies.
OntCIA: Software Change Impact Analysis System Based on the Semantic Web
Hee Seok Song
Vol. 10, No. 2, Page: 111 ~ 131
Keywords : Semantic Web, Ontology, Change Impact Analysis
Software change is an essential operation for software evolution. To maintain the system competently, managers as well as developers must be able to understand the structure of the system but the structure of software is hidden to the developers and managers who need to change it. In this paper, we present a system (OntCIA) for supporting change impact analysis for rating and billing domain based on the semantic web technology. The basic idea of OntCIA is to build a domain knowledge base using an OWL ontology and RDF to implement change impact analysis system that would support the managers and software developers in finding out information about structure of large software system. OntCIA allows users to incrementally build an ontology in rating and billing domain and provides useful information in response to user queries concerning the code, such as, for example 'Find the modules which have a role for confirming new subscription'. The strengths of OntCIA are its architecture for easy maintenance as well as semantic indexing by automatic reasoning.
Hybrid Multiple Classifier Systems
In-cheol Kim
Vol. 10, No. 2, Page: 133 ~ 145
Keywords : Multiple Classifier System, Bagging, Boosting, Meta Learner, Meta Learner, Bias, Bootstrap, Sampling
Combining multiple classifiers to obtain improved performance over the individual classifier has been a widely used technique. The task of constructing a multiple classifier system(MCS) contains two different issues : how to generate a diverse set of base-level classifiers and how to combine their predictions. In this paper, we review the characteristics of the existing multiple classifier systems: bagging, boosting, and stacking. And then we propose new MCSs: stacked bagging, stacked boosting, bagged stacking, and boasted stacking. These MCSs are a sort of hybrid MCSs that combine advantageous characteristics of the existing ones. In order to evaluate the performance of the proposed schemes, we conducted experiments with nine different real-world datasets from UCI KDD archive. The result of experiments showed the superiority of our hybrid MCSs, especially bagged stacking and boosted stacking, over the existing ones.
Designing Intelligent Agent System for Purchase Decision Making in Retail Electronic Commerce
Seok Chin Chu, and June Suk Hong
Vol. 10, No. 2, Page: 147 ~ 163
Keywords : Intelligent Agent, Electronic Commerce, Multi-Agent Negotiation
For the purchase of a cheaper product on the Internet, many customers have been trying to search online shopping mall sites and visit comparison-pricing shops that compare prices and other criteria of the product. Others have been participating into online auction markets or group-buying markets. However, a lot of online shopping malls, auction markets, and group-buying markets provide the same product with different prices. Since these marketplaces have different price settlement mechanism, it is very difficult for the customers to determine marketplace to purchase, considering different kinds of marketplaces at the same time. To overcome such limitations, decision rules and solution procedures for purchase decision making are necessary, which can cover multiple marketplaces simultaneously. For this purpose, purchase decision making in each market must be conducted to maximize customer's utility, and conflicts with other marketplaces must be resolved. Therefore, we have developed the rules and methods that can negotiate cooperatively the purchase decision making in several marketplaces, and designed an architecture of Intelligent Buyer Agent and a message structure to support the idea.

Advanced Search
Date Range