• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.1
• /
• pp.60-64
• /
• 2010

To discover association rules from nontransactional data, there have been many studies on discretization of attribute values. These studies do not reflect the change of discovered rules' confidence according to the change of the ranges of the discretized attributes, and perform the discretization stage and the rule discovery stage independently. This causes the ranges of attributes not properly discretized, thereby making the rules having high confidence excluded in the result set. To solve this problem, we propose a novel method that performs the discretization and rule discovery stages simultaneously in order to discretize ranges of attributes in such a way that the rules having high confidence are discovered well. To the end, we perform hierarchical clustering on the attributes in the right hand side of rules, then do characterization on every cluster thus obtained. The experimental result demonstrates that our method discovers the rules having high confidence better than existing methods.

• Information Systems Review
• /
• v.8 no.2
• /
• pp.119-138
• /
• 2006

Association Rule Mining has been successfully used for mining knowledge in static environment but it provides limited features to discovery time-dependent knowledge from multi-point data set. The aim of this paper is to develop a methodology which detects changes of customer behavior automatically from customer profiles and sales data at different multi-point snapshots. This paper proposes a procedure named 'Recursive Change Mining' for detecting continuous change of customer purchase behavior. The Recursive Change Mining Procedure is basically extended association rule mining and it assures to discover continuous and repetitive changes from data sets which collected at multi-periods. A case study on L department store is also provided.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2000.11a
• /
• pp.317-324
• /
• 2000

주기적인 연관규칙은 타겟데이터베이스를 일정 단위시간으로 나누었을 때 연관규칙이 만족하는 구간이 일정한 주기마다 발생하는 패턴을 탐색하는 방법이다. 하지만, 이 방법은 엄격한 주기를 가지도록 하여 실제 데이터에 그대로 적용하기가 어려웠다. 예를 들이 편의점 데이터에서 매일 오전 7시-8시 사이에 주기적으로 발생하는 연관규칙을 발견할 때, 이러한 연관규칙을 주기적인 연관규칙이라고 한다. 하지만, 실제 데이터에서는 날씨와 같이 사람의 행동에 영향을 미치는 다른 요인 때문에 항상 일정한 주기를 가지는 연관규칙을 찾기는 어렵다. 본 논문에서는 주기가 일정하지 않은 연관규칙을 찾기 위해서 연관규칙의 주기성을 허용 오차를 포함하며 재정의하고, 오차를 허용하기 위한 탐색 알고리즘을 보완하였다. 반면에, 오차를 허용함으로써 오차를 허용하지 않는 경우보다 더 많은 주기성을 찾을 수 있을 뿐만 아니라, 동일한 주기를 가지지만 오프셋이 다른 여러 개의 비슷한 주기가지 찾게 되어 사용자가 의미 있는 연관규칙을 찾는데 방해가 된다. 본 논문에서는 이를 해결하기 위해서 오차를 허용하는 주기적 연관규칙의 오차의 정도를 측정하기 위한 단위로 집중도(intensity)와 경향성(tendency)을 제안한다. 주기적 연관규칙이 매 주기마다 정확한 세그먼트에 나타나는 정도를 나타내는 집중도와, 최소 평균오차를 의미하는 경향성을 이용하여 유사한 주기들 중에서 대표주기만을 찾을 수 있도록 한다. 또한, 오차를 허용하는 주기적 연관규칙에서 오차가 주로 발생하는 패턴을 분석함으로써 고객들의 수요 경향성을 더 잘 파악할 수 있다. 예를 들어, 평소에는 매일 오진 7시∼8시에 나타나던 연관성이 지각하는 사람들이 같은 월요일에는 1시간 늦은 8시∼9시에 나타난다는 오타 정보까지 파악할 수 있다. 이러한 월요일마다 1시간 늦게 나타나는 오차의 경향성을 나타내는 오차 주기(error cyc1e)를 이용함으로써 고객들의 수요의 경향성을 좀 더 세밀한 부분까지 파악할 수 있게 해 준다.

• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.759-770
• /
• 2009

A problem of separating signals from noises is considered, when they are randomly mixed in the observation. It is assumed that the noise follows a Gaussian distribution and the signal follows a Gamma distribution, thus the underlying distribution of an observation will be a mixture of Gaussian and Gamma distributions. The parameters of the mixture model will be estimated from the EM algorithm. Then the signals and noises will be classified by a fixed threshold approach based on multiple testing using positive false discovery rate and Bayes error. The proposed method is applied to a real optical emission spectroscopy data for the quantitative analysis of inclusions. A simulation is carried out to compare the performance with the existing method using 3 sigma rule.

• Journal of KIISE:Software and Applications
• /
• v.31 no.8
• /
• pp.999-1009
• /
• 2004

As a new approach to the diagnosis of cancers, bioinformatics attracts great interest these days. Machine teaming techniques have produced valuable results, but the field of medicine requires not only highly accurate classifiers but also the effective analysis and interpretation of them. Since gene expression data in bioinformatics consist of tens of thousands of features, it is nearly impossible to represent their relations directly. In this paper, we propose a method composed of a feature selection method and genetic programming. Rank-based feature selection is adopted to select useful features and genetic programming based arithmetic operators is used to generate classification rules with features selected. Experimental results on Lymphoma cancer dataset, in which the proposed method obtained 96.6% test accuracy as well as useful classification rules, have shown the validity of the proposed method.

• Journal of information and communication convergence engineering
• /
• v.6 no.3
• /
• pp.255-260
• /
• 2008

In the recently years, there has been a big interest in ad hoc wireless network as they have tremendous military and commercial potential. An Ad hoc wireless network is composed of mobile computing devices that use having no fixed infrastructure of a multi-hop wireless network formed. So, the fact that limited resource could support the network of robust, simple framework and energy conserving etc. In this paper, we propose a new ad hoc multicast routing protocol for based on the ontology scheme called inference network. Ontology knowledge-based is one of the structure of context-aware. And the ontology clustering adopts a tree structure to enhance resilient against mobility and routing complexity. This proposed multicast routing protocol utilizes node locality to be improve the flexible connectivity and stable mobility on local discovery routing and flooding discovery routing. Also attempts to improve route recovery efficiency and reduce data transmissions of context-awareness. We also provide simulation results to validate the model complexity. We have developed that proposed an algorithm have design multi-hierarchy layered networks to simulate a desired system.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.805-812
• /
• 2005

Smart home is one of the ubiquitous environment platforms with various complex sensor-and-control network. In this paper, a now learning methodology for learning user's behavior preference pattern is proposed in the sense of reductive user's cognitive load to access complex interfaces and providing personalized services. We propose a fuzzy inductive learning methodology based on life-long learning paradigm for knowledge discovery, which tries to construct efficient fuzzy partition for each input space and to extract fuzzy association rules from the numerical data pattern.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.6
• /
• pp.7-17
• /
• 2009

The precise analysis of exercise data for designing an effective rehabilitation system is very important as a feedback for planing the next exercising step. Many subjective and reliable research outcomes that were obtained by analysis and evaluation for the human motor ability by various methods of biomechanical experiments have been introduced. Most of them include quantitative analysis based on basic statistical methods, which are not practical enough for application to real clinical problems. In this situation, data mining technology can be a promising approach for clinical decision support system by discovering meaningful hidden rules and patterns from large volume of data obtained from the problem domain. In this research, in order to find relational rules between posture training type and muscle activation pattern, we investigated an application of the WAR(Weishted Association Rule) to the biomechanical data obtained mainly for evaluation of postural control ability. The discovered rules can be used as a quantitative prior knowledge for expert's decision making for rehabilitation plan. The discovered rules can be used as a more qualitative and useful priori knowledge for the rehabilitation and clinical expert's decision-making, and as a index for planning an optimal rehabilitation exercise model for a patient.

• International Journal of Computer Science & Network Security
• /
• v.22 no.6
• /
• pp.339-345
• /
• 2022

Outcome-based education (OBE) is a tried-and-true teaching technique based on a set of predetermined goals. Program Educational Objectives (PEOs), Program Outcomes (POs), and Course Outcomes (COs) are the components of OBE. At the end of each year, the Program Outcomes are evaluated, and faculty members can submit many recommended measures which dependent on the relationship between the program outcomes and its courses outcomes to improve the quality of program and hence the overall educational program. When a vast number of courses are considered, bad actions may be proposed, resulting in unwanted and incorrect decisions. In this paper, a recommender system, using collaborative filtering and association rules algorithms, is proposed for predicting the best relationship between the program outcomes and its courses in order to improve the attributes of the graduates. First, a parallel algorithm is used for Collaborative Filtering on Data Model, which is designed to increase the efficiency of processing big data. Then, a parallel similar learning outcomes discovery method based on matrix correlation is proposed by mining association rules. As a case study, the proposed recommender system is applied to the Computer Information Systems program, College of Computer Sciences and Information Technology, Al-Baha University, Saudi Arabia for helping Program Quality Administration improving the quality of program outcomes. The obtained results revealed that the suggested recommender system provides more actions for boosting Graduate Attributes quality.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.7
• /
• pp.1588-1593
• /
• 2005

In coagulant control of water treatment plants, rule extraction, one of datamining categories, was performed for coagulant control of a water treatment plant. Clustering methods were applied to extract control rules from data. These control rules can be used for fully automation of water treatment plants instead of operator's knowledge for plant control. To perform fuzzy clustering, there are some coefficients to be determined and these kinds of studies have been performed over decades such as clustering indices. In this study, statistical indices were taken to calculate the number of clusters. Simultaneously, seed points were found out based on hierarchical clustering. These statistical approaches give information about features of clusters, so it can reduce computing cost and increase accuracy of clustering. The proposed algorithm can play an important role in datamining and knowledge discovery.