فهرست:
فصل اول 1
مقدمه 1
1-1- مقدمه. 2
1-2- روشهای دستهبندی.. 3
1-3- ارزیابی دستهبند. 4
1-4- تصدیق متقابل.. 6
1-5- الگوریتم نزدیکترین همسایه. 7
1-7- سر فصلها 9
فصل دوم 10
الگوریتم نزدیکترن همسایه و روشهای موجود برای بهبود آن.. 10
2-1-الگوریتم نزدیکترین همسایه. 11
2-2- محدودیتهای روش نزدیکترین همسایه. 14
2-3- مروری بر راهکارهای ارائه شده در گذشته برای بهبود الگوریتم نزدیکترین همسایه. 15
فصل سوم 18
روشهای تصمیمگیری دستهجمعی.. 18
3-1- مقدمه. 19
3-2- روشهای متفاوت برای ایجاد یک تصمیمگیر دستهجمعی.. 21
3-3- ساختارهای مختلف در روش تصمیمگیری دستهجمعی.. 22
3-4- رایگیری بین دستهبندها 23
3-5- معرفی چند روش تصمیمگیری دستهجمعی پرکاربرد. 24
فصل چهارم 28
روش پیشنهادی برای دستهجمعی کردن الگوریتم نزدیکترین همسایه. 28
4-1- مقدمه. 29
4-2- ایدهی اصلی.. 30
4-3- دستهجمعی کردن مجموعه دستهبندهای وزندار نزدیکترین همسایه. 31
فصل پنجم 39
نتایج آزمایشات پیاده سازی و نتیجهگیری.. 39
5-1- نتایج.. 40
فصل ششم 45
نتیجهگیری 45
فهرست منابع.. 48
Abstract 1
منبع:
Nearest Neighbor Pattern Classification, T. Cover, and P. Hart, , IEEE Transactions on Information Theory,1967, 13(1): 21-27.
Full Bayesian network Classifiers. Zhang, Jiang Su and Harry. ACM, 2006, international Conference on Machine Learning, Vol. 148, pp. 897-904.
A Simple Decomposition Method for Support Vector Machines. Chih-Wei Hsu and Chih-Jen Lin. ACM, 2002, Machine Learning , Vol. 46.
Neural networks for classification: a survey. Zhang, Guoqiang Peter. 2000, IEEE Transactions on Systems, Man, and Cybernetics, Part C , pp. 451-462.
Kohavi, Ron.A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection. 1995. pp. 1137-1145.
Locally adaptive metric nearest neighbor classification, C. Domeniconi, J. Peng, D. Gunopulos, IEEE Transaction on Pattern Analysis and Machine Intelligence 24 (2002) 1281–1285.
Improving nearest neighbor rule with a simple adaptive distance measure,J. Wang, P. Neskovic, L.N. Cooper, Pattern Recognition Letters 28 (2007) 207–213.
Flexible Metric Nearest-neighbor Classification, J. Friedman, Technical Report 113, Department of Statistices, Stanford University, 1994.
A method of learning weighted similarity function to improve the performance of nearest neighbor, M. Zolghadri Jahromi, E. Parvinnia, R. John, Information Sciences 179 (2009)
Neighborhood rough set based heterogeneous feature subset selection, Q. Hu, D. Yu, J. Liu, C. Wu, Information Sciences 178 (2008) 3577–3594.
A new approach to attribute reduction of consistent and inconsistent covering decision systems with covering rough sets, G. Chen, C.Z. Wang, Q.H. Hu Information Sciences 177 (2007) 3500–3518.
A novel feature selection approach: combining feature wrappers and filters, O. Uncu, I.B. Turks, Information Sciences 177 (2007) 449–466.
A cost sensitive learning algorithm for intrusion detection, S. Ghodratnama, M. R. Moosavi, M. Taheri, and M. Zolghadri Jahromi, Proceedings of ICEE 2010, art. no. 5507006: 559-565.
Bagging predictors, L.Brreima, Machine Learning 24:123-140, 1996.
Experiments with a new boosting algorithm. Y. Freund, and R. Schapire. Thirteenth International Conference on Machine Learning, 1996.
Active Learning for kNN based on Bagging Features, Shuo, S., Yuhai, L., Yuehua, H., Shihua, Z.,and Yong, L., Fourth International Conference on Natural Computation ICNC 2008 7, art. no. 4667945: 61-64.
Boosting k-nearest neighbor classifier by means of input space projection, García-Pedrajas, and Ortiz-Boyer, Expert Systems with Applications, 2009 36(7): 10570-10582.
Nearest Neighbor Ensemble, C. Domeniconi, and B. Yan, in Proceedings of the 17th International Conference on Pattern Recognition, Cambridge, UK, August 23-26, 2004.
A Proposed Method for Learning Rule Weights in Fuzzy Rule Based Classification Systems, M. Zolghadri Jahromi, and M. Taheri, Fuzzy Sets and Systems, 2008, 159 (4): 449–459.
A proposed method for learning rule weights in fuzzy rule based classification systems , M. Zolghadri Jahromi and M. Taheri, Fuzzy Sets and Systems 159, 449–459, 2008.
A method of learning weighted similarity function to improve the performance of nearest neighbor , M. Zolghadri Jahromi and E. Parvinnia and R John, Information Sciences 179, 2964–2973, 2009.
A cost sensitive learning algorithm for intrusion detection , S. Ghodratnama and M. R. Moosavi and M. Taheri and M. Zolghadri Jahromi, Proceedings of ICEE 2010, May 11-13, 2010.
A Novel Piecewise Linear Clustering Technique Based on Hyper Plane Adjustment , M. Taheri and E. Chitsaz and S. D. Katebi and M. Zolghadri Jahromi, Communications in Computer and Information Science, 2009, Volume 6, Part 1, 1-8.
UCI Machine Learning Repository , http://www.ics.uci.edu/~mlearn/databases/
A Novel Prototype Reduction Method for the K-Nearest Neighbor Algorithm with K ≥ 1,T. Yang, L. Cao, and C. Zhang, Lecture Notes in Computer Science, 2010, vol. 6119: 89-100.
http://www.cs.waikato.ac.nz/ml/weka