Earticle

The phishing is a kind of e-commerce lure which is intended to steal the confidential information of the internet user by making identical website of legitimate one in which the contents and images most likely remains similar to the legitimate website. The other way of phishing website is to do minor changes in the URL or in the domain of the website. In this paper, an anti-phishing system is proposed which is based on the development of the Add-on tool for the web browser. The performance of the proposed system is studied with four different data mining classification algorithms which are Class Imbalance Problem (CIP), Rule based Classifier (Sequential Covering Algorithm (SCA)), Nearest Neighbour Classification (NNC), Bayesian Classifier (BC). To evaluate the performance of the proposed anti-phishing system for the detection of phishing websites, we have collected 7690 legitimate websites and 2280 phishing websites from the authorised sources like APWG database and PhishTank. After the analysis of the anti-phishing system, more than 90 percentage successful results achieved at different time periods.

Generally the analytical tools of data mining have two learning types which are supervised and unsupervised learning algorithms. Classification and prediction are main analysis tools for supervised learning. In this paper, we perform a comparison study of classification algorithms in data mining. We make comparative studies between popular classification algorithms which are LDA, QDA, kernel method, K-nearest neighbor, naive Bayesian, SVM, and CART. Also, we use almost all classification data sets of UCI machine learning repository for our experiments. According to our results, we are able to select proper algorithms for given classification data sets.

지식탐사 프로세스의 핵심적인 역할을 담당하는 데이터마이닝 단계에서는 여러 가지 목적에 따라 알고리즘을 선택하여 사용한다. 최근 통계, 비즈니스, 전자 상거래, 의학, 생물학 등의 분야에서 데이터마이닝 기술아 적극적으로 활용되고 있으며, 이를 위해 다양한 알고리즘들이 계속해서 연구.개발되고 있다. 그러나 시간이 지나면 이들 중 각 분야 별로 우수한 응용성을 보이는 알고리즘이나 방대한 양의 데이터를 다루는데 있어 좋은 성능을 보이는 몇몇 알고리즘만이 남게 될 것이며 또한 앞으로는 이러한 알고리즘들만을 선별하여 집중 연구할 필요가 있다. 따라서 본 논문에서는 데이터마이닝에 널리 사용되고 활발한 연구가 진행중인 알고리즘들 중에서 연관규칙(association rule), 클러스터링(clustering), 신경망(neural network), 결정트리(decision tree), 유전자 알고리즘(genetic algorithm), 베이지안 네트워크(bayesian network), 메모리 기반 추론(memory-based reasoning)등 7가지 카테고리에 속하는 알고리즘들을 선정하여 분류.분석하였다. 우선 각 알고리즘의 계통과 특성들을 분석하였고 이를 토대로 비교.분석을 위한 14가지의 분류 기준을 제시하였다. 이러한 분류 기준에 근거하여 세부 알고리즘들을 분석해 보고 비교 가능한 일부 알고리즘은 여러 특징과 성능을 중심으로 각각 최상의 알고리즘을 도출해 보았다. 본 연구 결과는 데이터마이닝 분야의 흔재된 알고리즘들을 분류.분석함으로써 마이닝 기술 적용시 사용자에게 알고리즘 선택의 지표를 제시할 수 있을 것이다.

Data mining plays an important role in knowledge discovery process and usually various existing algorithms are selected for the specific purpose of the mining. Currently, data mining techniques are actively to the statistics, business, electronic commerce, biology, and medical area and currently numerous algorithms are being researched and developed for these applications. However, in a long run, only a few algorithms, which are well-suited to specific applications with excellent performance in large database, will survive. So it is reasonable to focus our effort on those selected algorithms in the future. This paper classifies about 30 existing algorithms into 7 categories - association rule, clustering, neural network, decision tree, genetic algorithm, memory-based reasoning, and bayesian network. First of all, this work analyzes systematic hierarchy and characteristics of algorithms and we present 14 criteria for classifying the algorithms and the results based on this criteria. Finally, we propose the best algorithms among some comparable algorithms with different features and performances. The result of this paper can be used as a guideline for data mining researches as well as field applications of data mining.