• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.8
• /
• pp.3606-3629
• /
• 2018

The discovered high potential utility itemsets (HPUIs) have significant influence on a variety of areas, such as retail marketing, web click analysis, and biological gene analysis. Thus, in this paper, we propose an algorithm called HPUIM-QGA (Mining high potential utility itemsets based on a quantum genetic algorithm) to mine HPUIs over uncertain datasets based on a quantum genetic algorithm (QGA). The proposed algorithm not only can handle the problem of the non-downward closure property by developing an upper bound of the potential utility (UBPU) (which prunes the unpromising itemsets in the early stage) but can also handle the problem of combinatorial explosion by introducing a QGA, which finds optimal solutions quickly and needs to set only very few parameters. Furthermore, a pruning strategy has been designed to avoid the meaningless and redundant itemsets that are generated in the evolution process of the QGA. As proof of the HPUIM-QGA, a substantial number of experiments are performed on the runtime, memory usage, analysis of the discovered itemsets and the convergence on real-life and synthetic datasets. The results show that our proposed algorithm is reasonable and acceptable for mining meaningful HPUIs from uncertain datasets.

• Journal of the Korean Society for Library and Information Science
• /
• v.47 no.1
• /
• pp.269-286
• /
• 2013

The advent of Web 2.0 and social media is taking a leading role of emerging big data. At the same time, however, informational dysfunction such as infringement of one's rights and violation of social order has been increasing sharply. This study, therefore, aims at defining malicious usage, identifying malicious feature, and devising an automated method for classifying them. In particular, the rule-based experiment reveals statistically significant performance enhancement.

• Journal of The Korean Association of Information Education
• /
• v.5 no.3
• /
• pp.389-400
• /
• 2001

The Internet Game is widespreaded quickly on web, and there are many kinds of funny games for users to use easily, so that can be applied to ICT(Information Communication Technology)education. In this paper, we provide the analysis on the usage of Internet games for children and teachers that is conducted by the decision tree algorithm, which is one of the popular data mining techniques. The results show the pattern of children's and teachers' usages of Internet games.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.49-51
• /
• 2001

데이터 마이닝은 KDD의 분야로서, 의미 있는 정보와 관심 있는 행동 패턴을 추출해 나가는 과정이다. WWW의 발전으로, 웹 데이터가 거대해지고 있다. 이러한 데이터 마이닝 분야에서도, 웹 사용 마이닝의 목적은 의미 있는 사용자 행동 패턴을 찾아내는 것이다. 특히 현재 전자상거래가 널리 활성화되고 있는 환경에서, 사용자의 특성을 발견해내는 것은 매우 중요한 부분이다. 사용자의 특성에 따라 사용자에게 상품을 추천하거나 메일을 보내는 것이나 사용자에게 적절하게 사이트를 구축하는 것이 가능하다. 전처리 과정을 통해서 추출된 트랜잭션 데이터를 모호한 사용자의 요구를 분석할 수 있는 퍼지 집합으로 변형시켜 Fuzzy Association Rule을 통해 분석한다. 그리고 분석된 결과에 대한 규칙을 사용자의 피드백을 통해서 다시 분석하는 과정을 거치게 된다. 사용자의 요구 사항을 적절히 반영할 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.118-120
• /
• 2001

웹 이용 마이닝은 거대만 웹 데이터 저장소의 로그들을 이용하여 웹 사용자의 사용 패턴을 분석하는 데이터 마이닝 기술이다. 마이닝 기술을 적용하기 위해서는 전처리 과정 중의 사용자와 세션을 정확하게 구분해야 하는데, 표준 웹 로그 형식의 웹 로그만으로는 사용자를 완전히 구분할 수 없다. 따라서 정확한 결과를 얻기 위해 사용자와 세션을 구분할 수 있는 모듈을 웹 서버에서 제공하거나, 각각의 페이지에 적당한 실행 필드를 삽입해야 한다. 사용자와 세션을 구분하는 데는 캐시 문제, 방화벽 문제. IP(ISP)문제, 프라이버시 문제, 쿠키 문제 등 많은 문제들이 있지만, 이 문제를 해결하기 위한 명확한 방법은 아직 없다. 이 논문은 참조 로그와 에이전트 로그, 그리고 액세스 로그 등 서버측 클릭스트림 데이터만을 이용하여 사용자와 세션을 구분하는 방법을 제안한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10a
• /
• pp.133-135
• /
• 2001

웹 사용 마이닝의 단계중 패턴 발전을 위해 초기 데이타를 정제하는 전처리 과정은 매우 중요한 작업이다. 전처리 과정의 결과가 높은 정확도를 가지고 있다면 마이닝의 결과 역시 보다 정확한 결과를 생성한다는 것은 여러 연구를 통해 널리 알려진 사실이다. 본 논문에서는 전처리 과정중 내용 페이지를 구분하기 위해 자주 이용되는 기법중 하나인 최대 전진 참조(M.F.R : Maximal Forward Reference) 기법을 개선한 인기도 기반 전진 참조(P.F.R : Popularity-weighted Forward Reference) 기법을 제안하고 예제를 통해 두 기법의 결과를 비교하였다. 그 결과 최대 전진 참조 기법에서 발생할 수 있는 오류를 극복한 인기도 기반 기법이 좀더 정확한 내용 페이지 구분이 가능하여 웹 사용 마이닝 단계에서 유용하게 활용 할 수 있음을 보였다.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.9
• /
• pp.35-46
• /
• 2010

Sequential pattern mining is one of the essential data mining tasks, and it is widely used to analyze data generated in various application fields such as web-based applications, E-commerce, bioinformatics, and USN environments. Recently data generated in the application fields has been taking the form of continuous data streams rather than finite stored data sets. Considering the changes in the form of data, many researches have been actively performed to efficiently find sequential patterns over data streams. However, conventional researches focus on reducing processing time and memory usage in mining sequential patterns over a target data stream, so that a research on mining more interesting and useful sequential patterns that efficiently reflect the characteristics of the data stream has been attracting no attention. This paper proposes a mining method of sequential patterns over data streams with a gap constraint, which can help to find more interesting sequential patterns over the data streams. First, meanings of the gap for a sequential pattern and gap-constrained sequential patterns are defined, and subsequently a mining method for finding gap-constrained sequential patterns over a data stream is proposed.

• Journal of the Korean Society for information Management
• /
• v.27 no.1
• /
• pp.25-40
• /
• 2010

The purpose of this study is to analyze the usage pattern between each academic database through social network analysis, and to support the academic database for users's needs. For this purpose, we have extracted log data to construct the academic database networks in the proxy server of K university library and have analyzed the usage pattern among each research area and among each social position. Our results indicate that the specialized academic database for the research area has more cohesion than the generalized academic database in the full-time professors' network and the doctoral students' network, and the density, degree centrality and degree centralization of the full-time professors' network and the doctoral students' network are higher than those of the other social position networks.

• Journal of Internet Computing and Services
• /
• v.10 no.5
• /
• pp.65-78
• /
• 2009

The most commonly adopted approach to find valuable information from tree data is to extract frequently occurring subtree patterns from them. Because mining frequent tree patterns has a wide range of applications such as xml mining, web usage mining, bioinformatics, and network multicast routing, many algorithms have been recently proposed to find the patterns. However, existing tree mining algorithms suffer from several serious pitfalls in finding frequent tree patterns from massive tree datasets. Some of the major problems are due to (1) modeling data as hierarchical tree structure, (2) the computationally high cost of the candidate maintenance, (3) the repetitious input dataset scans, and (4) the high memory dependency. These problems stem from that most of these algorithms are based on the well-known apriori algorithm and have used anti-monotone property for candidate generation and frequency counting in their algorithms. To solve the problems, we base a pattern-growth approach rather than the apriori approach, and choose to extract maximal frequent subtree patterns instead of frequent subtree patterns. The proposed method not only gets rid of the process for infrequent subtrees pruning, but also totally eliminates the problem of generating candidate subtrees. Hence, it significantly improves the whole mining process.

• Journal of the Korean Society for information Management
• /
• v.33 no.3
• /
• pp.321-343
• /
• 2016

This study aimed to design a semi-automatic web-based pilot system 1) to build a Korean novel geo-name, 2) to update the database using automatic geo-name extraction for a scalable database, and 3) to retrieve/visualize the usage of an old geo-name on the map. In particular, the problem of extracting novel geo-names, which are currently obsolete, is difficult to solve because obtaining a corpus used for training dataset is burden. To build a corpus for training data, an admin tool, HTML crawler and parser in Python, crawled geo-names and usages from a vocabulary dictionary for Korean New Novel enough to train a named entity tagger for extracting even novel geo-names not shown up in a training corpus. By means of a training corpus and an automatic extraction tool, the geo-name database was made scalable. In addition, the system can visualize the geo-name on the map. The work of study also designed, implemented the prototype and empirically verified the validity of the pilot system. Lastly, items to be improved have also been addressed.