(3.227.235.183) 您好!臺灣時間:2021/04/13 21:01
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:郭家榮
研究生(外文):Chia-Jung Kuo
論文名稱:資料壓縮技術應用於類神經網路之研究
論文名稱(外文):Improving Training Time of Neural Network - Applying Dataset Condense Technology
指導教授:黃謙順黃謙順引用關係
指導教授(外文):Chein-Shung Hwang
學位類別:碩士
校院名稱:中國文化大學
系所名稱:資訊管理研究所
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2003
畢業學年度:92
語文別:中文
論文頁數:90
中文關鍵詞:資料挖掘資料分類資料壓縮k-means叢集方法倒傳遞網路類神經網路
外文關鍵詞:Data MiningData ClassificationDataset Condensek-means algorithmBackpropagationNeural Network
相關次數:
  • 被引用被引用:2
  • 點閱點閱:286
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:40
  • 收藏至我的研究室書目清單書目收藏:0
由於類神經網路具有相當優異的學習能力,促進許多關於型態識別、資料分類、資料預測等領域研究紛紛利用類神經網路來解決其問題。然而類神經網路卻有著令人詬病的問題─建立網路分類模型耗費相當多的時間,尤其是在具有大量筆數之資料集時更甚。
許多相關研究使用不同的技巧來縮短網路訓練時間。本研究參考相關文獻,藉由減少資料集資料量進入網路訓練來縮短網路訓練時間,提出一個以k-means叢集技術為基礎的叢集資料壓縮演算法,其主要特色為具有可調節壓縮率參數,在應用上可提供更大的彈性。
研究中發現,藉由叢集壓縮方法來減少訓練資料集資料量確實能夠有效縮短網路訓練時間。本研究探討k-means叢集資料壓縮方法在各種不同特性之資料集中其表現,與CNN相較之下,雖然k-means叢集壓縮演算法可縮短的網路訓練時間幅度不如CNN,但是其網路準確度誤差比CNN來的小,且k-means叢集資料壓縮演算法在各種特性之資料集中皆具有相當穩定之表現。
總結本研究提出的k-means叢集壓縮演算法,為具有可調節壓縮率參數、與執行速度快、網路準確誤差極小、以及能夠有效的縮短網路訓練整體時間等優點,確實能夠改善類神經網路中訓練耗時問題。
Many researches have interested in using artificial neural network (ANN) to solve problems in pattern recognition, pattern classification and data prediction due to its ex-cellent learning capability. In spite of that, ANN has the drawback of requiring a large amount of training time, especially with dataset which have huge records.
Many researches attempt to use various techniques to improve the problem of training time while modeling the neural networks. In this paper, we propose a dataset reduction algorithm called k-means cluster-based data condensing algorithm that allows us to adapt the compressed ratio parameter. The results showed that the condensed dataset made from our proposed algorithm indeed could reduce the training time, and resulted in just a little accuracy down or even no accuracy down than the original huge dataset. In comparison with CNN condensing algorithm, although our method required a little more training time than CNN, it showed much better performance in terms of model accuracy.

中文摘要 ..................... iii
英文摘要 ..................... iv
誌謝辭  ..................... v
內容目錄 ..................... vi
表目錄  ..................... viii
圖目錄  ..................... ix
公式目錄 ..................... xi
第一章  緒論................... 1
  第一節  研究背景與動機 ........... 1
  第二節  研究目的 .............. 4
  第三節  研究預期貢獻 ............ 6
  第四節  研究範圍與架構 ........... 7
第二章  文獻探討 ................ 8
  第一節  類神經網路 ............. 8
  第二節  倒傳遞網路 ............. 13
  第三節  監督式學習 ............. 15
  第四節  資料縮減 .............. 16
  第五節  叢集技術 .............. 25
  第六節  CNN資料壓縮方法 ......... 30
第三章  叢集資料壓縮演算法 ........... 33
  第一節  k-means叢集資料壓縮演算法 ..... 33
  第二節  叢集資料壓縮範例 .......... 36
第四章  實驗結果 ................ 41
  第一節  實驗設計 .............. 41
  第二節  資料來源 .............. 42
  第三節  實驗架構 .............. 46
  第四節  實驗一-探討k-means叢集資料壓縮中較佳       
          叢集邊緣/內部壓縮組合... 47
  第五節  實驗二-探討分群數目對叢集壓縮演算法的
           影響 ............ 54
  第六節  實驗三-探討資料筆數對叢集壓縮演算法的
           影響 ............ 57
  第七節  實驗四-探討資料集包含雜訊程度對資料壓
           縮演算法之影響 ....... 60
  第八節  實驗五-探討資料集遺失欄位程度對資料壓
           縮演算法之影響 ....... 62
  第九節  實驗六-探討資料集類別個數對資料壓算法
           之影響 ........... 64
  第十節  實驗七-探討資料維度對資料壓縮演算法之
           影響 ............ 67
  第十一節 實驗八-以真實資料驗證資料壓縮演算法. 69
  第十二節 實驗九-比較CNN資料壓縮與k-means叢集
資料壓縮之優劣 ....... 72
第五章  結論與未來方向 ............. 79
參考文獻 ..................... 82

黃智育(2002),資料探勘於即時線上推薦系統之應用研究,私立朝陽科技大學資訊管理研究所未出版之碩士論文。
葉怡成(2001),類神經網路模式應用與實作,台北:儒林圖書公司。
羅華強(2001),類神經網路-MATLAB的應用,新竹:清蔚科技股份有限公司。
Ankerst, M., Breunig, M. M., Kriegel, H. P., & Sander, J. (1999). OP-TICS: Ordering points to identify the clustering structure. Pro-ceedings of the 1999 ACM SIGMOD international conference on Management of data (pp. 49-60), Philadelphia, Pennsylvania: ACM Press.
Agrawal, R., Gehrke, J., Gunopulos, D., & Raghavan, P. (1998). Automatic subspace clustering of high dimensional data for data mining applications. Proceeding of the ACM SIGMOD Interna-tional Conference on Management of Data, Seattle, Washington: ACM Press.
Balakrishnan, K., & Honavar, V. (1992). Faster learning in multi-layer networks by handling output layer flat-spots. Proceedings of the International Joint Conference on Neural Networks (IJCNN’92). Beijing, China.
Ben-Dor, A., Shamir, R., & Yahkini, Z. (1999). Clustering gene ex-pression patterns. Journal of Computational Biology, 6(3/4), 281-289.
Breiman, L. (1999). Pasting small votes for classification in large da-tabase and on-line. Machine Learning, 36, 85-103.
Carpnter, G. A., & Grossberg, S. (1998). The art of adaptive pattern recognition by self-organizing neural network, Computer, 21(3), 77-88.
Catlett, J. (1991). Megainduction: A test flight. In L. A. Birnbaum & G. C. Collins (Eds.), Proceedings of the 8th International Work-shop on Machine Learning (ML-91) (pp. 596-599), Evanston, Illinois: Morgan Kaufmann.
Chang, C. L. (1974). Finding prototypes for nearest neighbor classifi-ers. IEEE Transactions on Computers, 23(11), 1179-1184.
Choi, S. H., & Rockett, P. (2002). The training of neural classifier with condensed datasets. IEEE Transaction on System. Man. And Cybernetics-Part B: Cybernetics, 32(2), 202-206.
Craven, M. P., Curtis, K. M., Hayes-Gill, B. R., & Thursfield, C. D. (1997). A hybrid neural network/rule based technique for on-line gesture and hand-written character recognition. Pro-ceeding of the Fourth IEEE International Conference on Elec-tronics, Circuits and Systems, Cairo, Egypt, 2, 850-853.
Dayhoff, J. (1990). Neural-network architectures: An introduction. New York: Van Nostrand Reinhold.
Domingos, P. (1996). Unifying instance-based and rule-based induc-tion. Machine Learning, 24(2), 141-168.
Domingos, P. (1995). Rule induction and instance-based learning: A unified approach. Proceedings of the Fourteenth of Interna-tional Joint Conference on Artificial Intelligence (IJCAI-95) (pp. 1266-1232), Montreal, Canada: Morgan Kaufmann.
Ester, M., Kriegel, H. P., Sander, J., & Xu, X. (1996). A density-based algorithm for discovering clusters in large spatial databases with noise. In E. Simoudis, J. Han, & U. Fayyad (Eds.), Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining (pp. 226-231), Portland, Oregon: AAAI Press.
Estivill-Castro, V., & Lee, I. (2000a). AMOEBA: Hierarchicl cluster-ing based on spatial proximity using delaunay diagram. In P. Forer, A. Yeh, J. He (Eds.), Proceedings of the 9th International Symposium on Spatial Data Handling (SDH2000) (pp. 7a.26-7a.41), HongKong: Study Group on Geographical Infor-mation Science of the International Geographical Union.
Estivill-Castro, V. & Lee, I. (2000b). AUTOCLUST: Automatic clus-tering via boundary extraction for massive point data sets. In R. J. Abrahart & B. H. Carlisle (Eds.), Proceedings of the 5th In-ternational Conference on GeoComputation. (pp. 23-25), Uni-versity of Greenwich, Kent, UK: GeoComputatuin.
Fahlman, S. E. (1988). Faster-learning variations on back-propagation: An empirical study. In D. Touretzky, G. E. Hinton & Sejnowski (Eds.), Proceedings of the 1988 Connectionist Models Summer School, San Mateo, California: Morgan Kaufmann.
Fraley, C. & Raftery A. E. (1998). How many clusters? Which clus-tering method? Answers via model-based cluster analysis. The Computer Journal, 41(8), 578-588.
Gate, G. W. (1972). The reduced nearest neighbor rule. IEEE Transac-tion on Information Theory, 18(5), 431-433.
Guha, S., Rastogi, R., & Shim, K. (1998). CURE: An efficient clus-tering algorithm for large databases. Proceedings of the 1998 ACM SIGMOD international conference on Management of data (pp. 73-84), Seattle, Washington: ACM Press.
Guha, S., Rastogi, R., & Shim, K. (1999). ROCK: A robust clustering algorithm for categorical attributes. Information Systems, 25(5), 512-521.
Han, J. & Kamber, M. (2000). Data Mining: Concepts and techniques. San Francisco, California: Morgan Kaufmann.
Hart, P. E. (1968). The condensed nearest neighbor rule. IEEE trans-action on Information Theory, 14(3), 515-516.
Hinneburg, A., & Keim, D. A. (1998). An efficient approach to clus-tering in multimedia databases with noise. In R. Agrawal & P. Stolorz (Eds.), Proceeding of the 4th International Conference on Knowledge Discovery and Data Mining (pp. 58-65), New York: AAAI Press.
Jain, A. K., Murty, M. N., & Flynn, P. J. (1999). Data clustering: A re-view. ACM Computing Surveys, 31(3), 264-323.
Karypis, G., Han, E. H., & Kumar, V. (1999). CHAMELEON: Hier-archical clustering using dynamic modeling. IEEE Computer, 32(8), 68-75.
Kaufman, L., & Rousseeuw, P. J. (1990). Finding groups in data: An introduction to cluster analysis. New York: John Wiley & Sons.
Kohavi, R. (1996). Scaling up the accuracy of Naive-Bayes classifiers: A decision-tree hybrid. In E. Simoudis, J. Han, & U. Fayyad (Eds.), Proceeding of the Second International Conference on Knowledge Discovery and Data Mining (pp. 202-207), Menlo Park, California: AAAI Press.
MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. In L. L. Cam & J. Neyman (Eds.), Proceedings on 5th Berkeley Symposium on Mathematics, Sta-tistics and Probability (pp. 281-297), Berkeley: University of California.
Mehrotra, K., Mohan, C. K., & Ranka, S. (1997). Element of artificial neural networks. Cambridge, Massachusetts: MIT Press.
Melli, G. (2003). Datgen: A program that creates structured data. [Online], Available: http://www.datasetgenerator.com [No date].
More, J. J. (1977). The Levenberg-Marquardt algorithm: Implementa-tion and theory. In G. A. Watson (Ed.), Numerical Analysis: Lecture notes in mathematics (pp. 105-116), New York: Springer Verlag.
Ng, R. T., & Han, J. (1994). Efficient and effective clustering method for spatial data mining. Proceeding of 1994 International Conference Very Large Databases (VLDB’94) (pp. 144-155), Santiago, Chile: Morgan Kaufmann.
Olave, M., Rajkovic, V., & Bohanec, M. (1989). An application for admission in public school systems. In I. Th. M. Snellen, W. B. H. J. van de Donk, & J. P. Baquiast (Eds.), Expert Systems in Pubic Administration (pp. 140-160), North Holland: Elsevier Science Publishers.
Parker, D. B. (1985). Learning-logic: Casting the cortex of the human brain in silicon (Technical report TR-47, center for computa-tional research in economics and management science). Cam-bridge, Massachusetts: Massachusetts Institute of the Technol-ogy.
Quinlan, J. R. (1993). C4.5: Programs for Machine Learning. San Mateo, California: Morgan Kaufmann.
Quinlan, J. R. (1986). Induction od decision trees. Machine Learning, 1(2), 81-106.
Riedmiller, M., & Braun, H. (1993). A direct adaptive method for faster backpropagation learning: The RPROP algorithm. In H. Ruspini (Ed.), Proceeding of the IEEE International Conference on Neural Networks (pp. 586-591), San Francisco, California.
Ritter, G. L., Woodruff, H. B., Lowry, S. R., & Isenhour, T. L. (1975). An algorithm for a selective nearest neighbor decision rule. IEEE Transaction on Information Theory, 21(6), 665-669.
Romero, R., Touretzky D., & Thibadeau, R. (1997). Optical chinese character recognition using probabilistic neural network. Pat-tern Recognition 8(30), 1279-1292.
Rumelhart, D.E., McClelland, J.L., & the PDP Research Group (1986). Parallel Distributed Processing, 1(1), Cambridge, Massa-chusetts: MIT Press.
Salomon, R., & van Hemmen, J. L. (1996). Accelerating backpropa-gation through dynamic self-adaptation. Neural Networks, 9(4), 589-601.
Salzberg, S. (1991). A nearest hyperrectangle learning method. Ma-chine Learning, 6, 277-309.
Schmidhuber, J. (1989). Accelerated learning in back-propagation nets. In R. Pfeifer, Z. Schreter, F. Fogelman-Soulie, & L. Steels (Eds.), Connectionism in perspective (pp. 439-445), Amsterdam, North Holland: Elsevier Science Publishers.
Sheikholeslami, G., Chatterjee, S., & Zhang, A. (1998). WaveCluster: A multi-resolution clustering approach for very large spatial da-tabases. In A. Gupta, O. Shmueli, J. Widom (Eds.), Proceedings of the 24th Very Large Databases Conference (VLDB’98) (pp. 428-439), New York City, New York: Morgan Kaufmann.
Sutton, R. S., Barto, A. G. (1981). Toward a modern theory of adaptive networks: Expectation and prediction. Psychological Review, 88, 135-170.
Wang, W., Yang, J. & Muntz, R. (1997). STING: A statistical informa-tion grid approach to spatial data mining. In M. Jarke, M. J. Carey, K. R. Dittrich, F. H. Lochovsky, P. Loucopoulos, & M. A. Jeusfeld (Eds.), Proceeding of the 23rd International Confer-ence on Very Large Databases (VLDB’97) (pp. 186-195), Ath-ens, Greece: Morgan Kaufmann.
Wettschereck & Dietrich (1994). A hybrid nearest-neighbor and near-est-hyperrectangle algorithm. In F. Bergadamo, & L. de Raedth (Eds.), Proceedings of 7th European Conference on Machine Learning (ECML’94) (pp. 323-335), Catania, Italy: Springer Verlag.
Wilson, D. R. & Martinez, T. R. (2000). Reduction techniques for in-stance-based learning algorithms. Machine Learning, 38(3), 257-286.
Zhang, T., Ramakrishnan, R., & Livny M. (1996). BIRCH: An effi-cient data clustering method of very large databases. Pro-ceeding 1996 ACM-SIGMOD International Conference on Management of Data (SIGMOD’96) (pp. 103-114), Montreal, Canada: ACM Press.
Zupan B., Bohanec M., Bratko I., & Demsar J. (1997). Machine learning by function decomposition. Proceedings of the 14th International Conference on Machine Learning (ICML’97) (pp. 421-429), San Mateo, California: Morgan Kaufmann.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔