• Journal of the Korean Data and Information Science Society
• /
• v.21 no.4
• /
• pp.709-717
• /
• 2010

The task of association rule mining is to find certain association relationships among a set of data items in a database. There are three primary measures for association rule, support and confidence and lift. In this paper we developed a association rule ranking function using conditional probability increment ratio. We compared our function with several association rule ranking functions by some numerical examples. As the result, we knew that our decision function was better than the existing functions. The reasons were that the proposed function of the reference value is not affected by a particular association threshold, and our function had a value between -1 and 1 regardless of the range for three association thresholds. And we knew that the ranking function using conditional probability increment ratio was very well reflected in the difference between association rule measures and the minimum association rule thresholds, respectively.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.3
• /
• pp.397-405
• /
• 2010

Data mining is the method to find useful information for large amounts of data in database, and one of the important goals is to search and decide the association for several variables. The task of association rule mining is to find certain association relationships among a set of data items in a database. There are three primary measures for association rule, support and confidence and lift. In this paper we developed a association rule ranking function by decreased lift influence to generate association rule for items satisfying at least one of three criteria. We compared our function with the functions suggested by Park (2010), and Wu et al. (2004) using some numerical examples. As the result, we knew that our decision function was better than the function of Park's and Wu's functions because our function had a value between -1 and 1regardless of the range for three association thresholds. Our function had the value of 1 if all of three association measures were greater than their thresholds and had the value of -1 if all of three measures were smaller than the thresholds.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.181-183
• /
• 2003

최근 대용량의 데이터베이스로부터 유용한 정보를 발견하고 데이터간에 존재하는 연관성을 탐색하고 분석하는 데이터 마이닝에 관한 많은 연구들이 진행되고 있다. 다양한 생체 신호를 분석하기 위하여 데이터 마이닝 기법을 이용할 수 있다. 본 논문에서는 심전도 신호의 패턴을 분류하기 위하여 신경망 기법을 적용하였다. 최근 패턴분류에 있어서 각광을 받고 있는 SVM 모델은 학습과정에서 얻어진 확률분포를 이용하여 의사결정함수를 추정한 후 이 함수에 따라 새로운 데이터를 이원분류 하는 것으로 분류 문제에 있어서 일반화 기능이 매우 높다. 기존에 많이 이용되던 BP 모델과 비교평가 하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.9
• /
• pp.2055-2062
• /
• 2013

Digital evidence which is the new form of evidence to crime makes little difference in value and function with existing evidences. As time goes on, digital evidence will be the important part of the collection and the admissibility of evidence. Usually a digital forensic investigator has to spend a lot of time in order to find clues related to the investigation among the huge amount of data extracted from one or more potential containers of evidence such as computer systems, storage media and devices. Therefore, these evidences need to be ranked and prioritized based on the importance of potential relevant evidence to decrease the investigate time. In this paper we propose a methodology which prioritizes order in which evidences are to be examined in order to help in selecting the right evidence for investigation. The proposed scheme is based on Fuzzy Multi-Criteria Decision Making, in which uncertain parameters such as evidence investigation duration, value of evidence and relation between evidence, and relation between the case and time are used in the decision process using the aggregation function in fuzzy set theory.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.353-362
• /
• 2016

By Wikipedia, data mining is the process of discovering patterns in a big data set involving methods at the intersection of association rule, decision tree, clustering, artificial intelligence, machine learning. and database systems. Association rule is a method for discovering interesting relations between items in large transactions by interestingness measures. Association rule interestingness measures play a major role within a knowledge discovery process in databases, and have been developed by many researchers. Among them, the Hellinger measure is a good association threshold considering the information content and the generality of a rule. But it has the drawback that it can not determine the direction of the association. In this paper we proposed a signed Hellinger measure to be able to interpret operationally, and we checked three conditions of association threshold. Furthermore, we investigated some aspects through a few examples. The results showed that the signed Hellinger measure was better than the Hellinger measure because the signed one was able to estimate the right direction of association.