Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/92959
標題: Feature Selection based on an Improved Cat Swarm Optimization
特徵選取基於改良式貓群演算法
作者: 張愷元
Kai-Yuan Zhang
關鍵字: feature selection
cat swarm optimization
support vector machines
貓演算法
特徵選取
支援向量機
仿生演算法
引用: [1]蔡佾翰,「使用 TF-IDF 和 SVM 評量中文文章適讀性」,國立嘉義大學資訊工程學系研究所碩士論文,2011 [2]國家發展委員會, 政府資料開放平臺」 來源網址:http://data.gov.tw/,查詢時間:2014 [3]Zheng Zhao, Fred Morstatter, Shashvata Sharma, Salem Alelyani, Aneeth Anand, Huan Liu, 'Advancing Feature Selection Research-ASU feature selection repository,' School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, 2010. [4]Richard Ernest Bellman, 'Rand Corporation. Dynamic programming,' Princeton University Press, ISBN 978-0-691-07951-6, 1957. [5]Holland, J.H., 'Adaptation in Natural and Artificial Systems,' The University Michigan Press, Ann Arbor, 1975. [6]Kennedy, J.; Eberhart, 'Particle Swarm Optimization,' Proceedings of IEEE International Conference on Neural Networks IV. pp. 1942–1948, 1995. [7]Rania Hassan, Babak Cohanim, 'A comparison of particle swarm optimization and the genetic algorithm American Institute of Aeronautics and Astronautics,' Olivier de Weck, Gerhard Venter , 2005. [8]Poli, R. 'An analysis of publications on particle swarm optimisation applications,' Technical Report CSM-469 (Department of Computer Science, University of Essex, UK), 2007. [9]Chung-Jui Tu, Li-Yeh Chuang, Jun-Yang Chang, and Cheng-Hong Yang, Member, IAENG, 'Feature Selection using PSO-SVM,' IAENG International Journal of Computer Science. 33:1, IJCS_33_1_18. [10]D. Karaboga, 'An Idea Based On Honey Bee Swarm for Numerical Optimization,' Technical Report-TR06,Erciyes University, Engineering Faculty, Computer Engineering Department, 2005. [11]Alberto Colorni, Marco Dorigo, Vittorio Maniezzo, 'Distributed Optimization by Ant Colonies', Proceedings of the 1st European Conference on Artificial Life, pp. 134-142, Paris, 1991. [12]Chu, S.C. and P.W. Tsai, 'Computational Intelligence Based On The Behavior Of Cats,' International Journal of Innovative Computing Information and Control, 3(1), pp. 163-173, 2007. [13]Deivaseelan.A, PG Student, P.Babu, Associate Professor, 'Modified Cat Swarm Optimization For Iir System Identification', ISSN 1995-0772, pp.731-740, 2012. [14]Meysam Orouskhani , Yasin Orouskhani , Mohammad Mansouri , Mohammad Teshnehlab, 'A Novel Cat Swarm Optimization Algorithm for Unconstrained Optimization Problems,' Information Technology and Computer Science, Volume. 5, No. 11, pp. 32-41, 2013. [15]Corinna Cortes, Vladimir Vapnik, 'Support-Vector Networks,' Machine Learning , Volume 20, Issue 3, pp. 273-297, 1995. [16]Gerard Salton Cornell Univ., Ithaca, NY Edward A. Fox 'Extended Boolean Information Retrieval,' Communications of the ACM Volume 26 Issue 11, pp. 1022-1036, 1983. [17]Hettich, S., C.L. Blake, and C.J. Merz. 'UCI Repository of Machine Learning Databases,' from website : http//www.ics.uci.edu/~mlearn/MLRepository.html, 1998. [18]Stephen D. Bay, Dennis Kibler, Michael J. Pazzani, and Padhraic Smyth, 'The UCI KDD Archive of Large Data Sets for Data Mining Research and Experimentation,' ACM SIGKDD, Volume 2, Issue 2, pp. 14-18, 2000. [19]Lin, K.C. and H.Y. Chien, 'CSO-based feature selection and parameter optimization for support vector machine,' in Joint Conference on Pervasive Computing, pp. 783-788, 2009. [20]S. L. Salzberg, 'On Comparing Classifiers: Pitfalls to Avoid and a Recommended Approach,' Data Mining and Knowledge Discovery, Volume 1, pp. 317–328, 1997. [21]Tsai, C.-H.'MMSEG: A Word Identification System for Mandarin Chinese Text'Based on Two Variants of the Maximum Matching Algorithm ,' from website : http://technology.chtsai.org/mmseg/, 2000.
摘要: Feature selection is a research which has been discussed widely. We often use meta-heuristic algorithm for feature selection. Cat Swam Optimization is a meta-heuristic algorithm. The performance of Cat Swam Optimization algorithm is better than Particle Swarm Optimization algorithm, but Cat Swam Optimization has disadvantage that taking much time for best solution. The research improve algorithm Cat Swam Optimization. Proposing a new algorithm named Improved Cat Swam Optimization. We improve a formula of Seeking Mode and increase a Crossover Technique to Seeking Mode. The aim is improving disadvantage of Cat Swam Optimization algorithm. By this improvement, we improve the convergence speed let Cat Swam Optimization can find the best solution quickly. In the research, we use the algorithms with support vector machine for feature selection. In the section of experimental results, Improved Cat Swam Optimization has better classification accuracy in different UCI datasets. We also use Improved Cat Swam Optimization for classification of articles.
特徵選取是近來被廣泛討論的研究議題。特徵選取是組合最佳化議題,目的是從特徵組合中找出最佳的組合並降低特徵維度。但要從特徵維度中找出最佳的子集合需要大量的時間,因為資料的分析時間會隨著維度上升而成指數成長,因此我們常以仿生演算法來找出較佳的特徵子集合。貓演算法是在 2007 年被提出的仿生演算法,它模仿貓群的行為,以搜尋模式與追蹤模式兩個模式來找尋最佳解。現今,貓演算法已被應用在各種組合最佳化問題上,且目前有研究指出貓演算法的搜尋效果優於粒子群演算法,但貓演算法有著收斂時間過長而難以找出最佳解的缺點。本研究針對貓演算法提出改良,提出了改良式的貓演算法。研究中針對貓演算法在搜尋模式中的移動方式進行改良,並在搜尋模式中加入一個變化版的交叉操作,目的是強化貓演算法的收斂速度並改善貓演算法難以往最佳解收斂的缺點。我們將改良式的貓演算法結合支援向量機進行特徵選取,並以 UCI 資料集進行特徵選取實驗,實驗結果顯示改良式貓演算法的分類正確率優於原始的貓演算法,並以改良式貓演算法應用於文章分類。
URI: http://hdl.handle.net/11455/92959
其他識別: U0005-0107201513041500
文章公開時間: 10000-01-01
Appears in Collections:資訊管理學系

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.