簡易檢索 / 詳目顯示

回結果列表
研究生： 吳麗莎
Monalisa - Gosumolo
論文名稱： 應用多目標自適應粒子群最佳化演算法於數值關聯法則之研究
Multi-Objective Particle Swarm Optimization Algorithm using Adaptive Archive Grid for Numerical Association Rules Mining
指導教授： 郭人介
Ren-Jieh Kuo
口試委員: 王孔政
Kung-Jeng Wang
駱至中
Zhi-Zhong Luo
學位類別： 碩士
Master
系所名稱： 管理學院 - 工業管理系
Department of Industrial Management
論文出版年： 2015
畢業學年度： 103
語文別： 英文
論文頁數： 66
中文關鍵詞： MOPSO 、multi-objective optimization 、association rules 、particle swarm optimization
外文關鍵詞： MOPSO, multi-objective optimization, association rules, particle swarm optimization
相關次數： 點閱：377  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • There are two challenges in association rules mining: dealing with numerical attribute containing with categorical and quantitative data; and accommodating several criteria for discovering the optimal rules result without any pre-processing activities or pre-defined parameter values. To deal with these problems, this study solves the numerical association rules mining using a multi-objective particle swarm optimization using adaptive archive grid based on Pareto optimal strategy. This method aims to optimize confidence, comprehensibility and interestingness factors for discovering rules. By implementing this method, the numerical association rule does not require any major pre-processing activities such as discretization. Moreover, minimum support and confidence are not a pre-requisite. The proposed method is evaluated by using three benchmark data sets contains with numerical attributes. Furthermore, it is applied to the real case data set taken from a weight loss application for discovering association rules in terms of the behavior of customer page usage.

    There are two challenges in association rules mining: dealing with numerical attribute containing with categorical and quantitative data; and accommodating several criteria for discovering the optimal rules result without any pre-processing activities or pre-defined parameter values. To deal with these problems, this study solves the numerical association rules mining using a multi-objective particle swarm optimization using adaptive archive grid based on Pareto optimal strategy. This method aims to optimize confidence, comprehensibility and interestingness factors for discovering rules. By implementing this method, the numerical association rule does not require any major pre-processing activities such as discretization. Moreover, minimum support and confidence are not a pre-requisite. The proposed method is evaluated by using three benchmark data sets contains with numerical attributes. Furthermore, it is applied to the real case data set taken from a weight loss application for discovering association rules in terms of the behavior of customer page usage.

    ABSTRACT i CONTENTS ii LIST OF TABLES v LIST OF FIGURES vi CHAPTER 1 INTRODUCTION 1 1.1 Research Background 1 1.2 Research Objectives 2 1.3 Research Scope and Constraints 3 1.4 Thesis Organization 3 CHAPTER 2 LITERATURE REVIEW 5 2.1 Association Rules 5 2.1.1 Basic Concept 5 2.1.2 Numerical Association Rules 6 2.2 Multi-objective Optimization 6 2.3 Meta-Heuristic Method 7 2.3.1 Basic Concept 7 2.3.2 Meta-Heuristic in Association Rules Mining 8 2.3.3 Particle Swarm Optimization (PSO) Algorithm 9 2.3.4 Multi-objective Particle Swarm Optimization using Adaptive Archive Grid (MOPSO) 10 2.4 Measurement 11 2.4.1 Support 11 2.4.2 Confidence 11 2.4.3 Comprehensibility 11 2.4.4 Interestingness 12 CHAPTER 3 RESEARCH METHODOLOGY 13 3.1 Pre-processing 13 3.2 Multi-Objective Particle Swarm Optimization Algorithm using Adaptive Archived Grid (MOPSO) 14 3.2.1 Objectives design 14 3.2.2 Measurement factors 15 3.2.3 MOPSO Algorithm 16 3.3 Post-processing 24 CHAPTER 4 EXPERIMENTAL RESULT 25 4.1 Datasets 25 4.2 Parameter Settings 25 4.3 Experimental Results and Analysis 28 4.3.1 Algorithm Convergences 28 4.3.2 Computational Result 31 CHAPTER 5 CASE STUDY 36 5.1 Source and dataset 36 5.2 Pre-processing 39 5.3 MOPSO implementation 41 5.4 Post-processing 42 5.5 Association Rules result 42 CHAPTER 6 CONCLUSIONS AND FUTURE STUDY 47 6.1 Conclusions 47 6.2 Contributions 47 6.3 Future Research 48 REFERENCES 49 Appendix I GENERAL FACTORIAL DESIGN OF DETERMINING TUNING PARAMETERS FOR MOPSO 51 Appendix II LIST OF ASSOCIATION RULES 56

    ADAMO, J.-M. 2001. Data Mining for Association rules and Sequential Patterns, Springer.
    AGRAWAL, R., IMIELINSKI, T. & SWAMI, A. 1993. Mining Association Rules between Sets of Items in Large Databases.
    ALATAS, B. & AKIN, E. 2008. Rough particle swarm optimization and its applications in data mining. Soft Computing, 12, 1205-1218.
    ALATAS, B. & AKIN, E. 2009. Multi-objective rule mining using a chaotic particle swarm optimization algorithm. Knowledge-Based Systems, 22, 455-460.
    ALATAŞ, B. & AKIN, E. 2005. An efficient genetic algorithm for automated mining of both positive and negative quantitative association rules. Soft Computing, 10, 230-237.
    ALATAS, B., AKIN, E. & KARCI, A. 2008. MODENAR: Multi-objective differential evolution algorithm for mining numeric association rules. Applied Soft Computing, 8, 646-656.
    BEIRANVAND, V., MOBASHER-KASHANI, M. & ABU BAKAR, A. 2014. Multi-objective PSO algorithm for mining numerical association rules without a priori discretization. Expert Systems with Applications, 41, 4259-4273.
    COELLO, C. A. C., PULIDO, G. T. & LECHUGA, M. S. 2004. Handling multiple objectives with particle swarm optimization. Evolutionary Computation, IEEE Transactions on, 8, 256-279.
    EBERHART, R. & KENNEDY, J. A new optimizer using particle swarm theory. Micro Machine and Human Science, 1995. MHS '95., Proceedings of the Sixth International Symposium on, 4-6 Oct 1995 1995. 39-43.
    EBERHART, R. C. & YUHUI, S. Particle swarm optimization: developments, applications and resources. Evolutionary Computation, 2001. Proceedings of the 2001 Congress on, 2001 2001. 81-86 vol. 1.
    ENGELBRECHT, A. P. 2005. Fundamentals of Computation Swarm Intelligence, England, Wiley.
    FIDELIS, M. V., LOPES, H. S. & FREITAS, A. A. 2002. Discovering Comprehensible Classification Rules with a Genetic Algorithm.
    FREITAS, A. A. 1998. Data Mining and Knowledge Discovery with Evoluntionary Algorithm, Springer.
    GHOSH, A. & NATH, B. 2004. Multi-objective rule mining using genetic algorithms. Information Sciences, 163, 123-133.
    KNOWLES, J. & CORNE, D. 2000. Approximating the Nondominated Front Using the Pareto Archived Evolution Strategy. Evolutionary Computation, 8, 149-172.
    LAROSE, D. T. & LAROSE, C. D. 2014. Discovering Knowledge in Data, Wiley.
    MATA, J., ALVAREZ, J.-L. & RIQUELME, J.-C. 2002. Discovering Numerical Association Rules via Evoluntionary Algorithm.
    MINAEI-BIDGOLI, B., BARMAKI, R. & NASIRI, M. 2013. Mining numerical association rules via multi-objective genetic algorithms. Information Sciences, 233, 15-24.
    QODMANAN, H. R., NASIRI, M. & MINAEI-BIDGOLI, B. 2011. Multi objective association rule mining with genetic algorithm without specifying minimum support and minimum confidence. Expert Systems with Applications, 38, 288-298.
    TALBI, E.-G. 2009. Metaheuristics From Design to Implementation, New Jersey, Wiley.

    無法下載圖示 全文公開日期 2020/06/09 (校內網路)
    全文公開日期 本全文未授權公開 (校外網路)
    全文公開日期 本全文未授權公開 (國家圖書館：臺灣博碩士論文系統)
    QR CODE