學術產出-Theses

題名 多項分配之分類方法比較與實證研究
An empirical study of classification on multinomial data
作者 高靖翔
Kao, Ching Hsiang
貢獻者 余清祥
Yue, Jack C.
高靖翔
Kao, Ching Hsiang
關鍵詞 多項分配

相似指標
電腦模擬
支持向量機
冪次定理
Multinomial distribution
Entropy
Similarity index
Computer simulation
Support vector machine
Power Law
Zipf`s Law
日期 2008
上傳時間 18-Sep-2009 20:10:38 (UTC+8)
摘要 由於電腦科技的快速發展,網際網路(World Wide Web;簡稱WWW)使得資料共享及搜尋更為便利,其中的網路搜尋引擎(Search Engine)更是尋找資料的利器,最知名的「Google」公司就是藉由搜尋引擎而發跡。網頁搜尋多半依賴各網頁的特徵,像是熵(Entropy)即是最為常用的特徵指標,藉由使用者選取「關鍵字詞」,找出與使用者最相似的網頁,換言之,找出相似指標函數最高的網頁。藉由相似指標函數分類也常見於生物學及生態學,但多半會計算兩個社群間的相似性,再判定兩個社群是否相似,與搜尋引擎只計算單一社群的想法不同。
本文的目標在於研究若資料服從多項分配,特別是似幾何分配的多項分配(許多生態社群都滿足這個假設),單一社群的指標、兩個社群間的相似指標,何者會有較佳的分類正確性。本文考慮的指標包括單一社群的熵及Simpson指標、兩社群間的熵及相似指標(Yue and Clayton, 2005)、支持向量機(Support Vector Machine)、邏輯斯迴歸等方法,透過電腦模擬及交叉驗證(cross-validation)比較方法的優劣。本文發現單一社群熵指標之表現,在本文的模擬研究有不錯的分類結果,甚至普遍優於支持向量機,但單一社群熵指標分類法的結果並不穩定,為該分類方法之主要缺點。
Since computer science had changed rapidly, the worldwide web made it much easier to share and receive the information. Search engines would be the ones to help us find the target information conveniently. The famous Google was also founded by the search engine. The searching process is always depends on the characteristics of the web pages, for example, entropy is one of the characteristics index. The target web pages could be found by combining the index with the keywords information given by user. Or in other words, it is to find out the web pages which are the most similar to the user’s demands. In biology and ecology, similarity index function is commonly used for classification problems. But in practice, the pairwise instead of single similarity would be obtained to check if two communities are similar or not. It is dislike the thinking of search engines.
This research is to find out which has better classification result between single index and pairwise index for the data which is multinomial distributed, especially distributed like a geometry distribution. This data assumption is often satisfied in ecology area. The following classification methods would be considered into this research: single index including entropy and Simpson index, pairwise index including pairwise entropy and similarity index (Yue and Clayton, 2005), and also support vector machine and logistic regression. Computer simulations and cross validations would also be considered here. In this research, it is found that the single index, entropy, has good classification result than imagine. Sometime using entropy to classify would even better than using support vector machine with raw data. But using entropy to classify is not very robust, it is the one needed to be improved in future.
參考文獻 中文部分
1. 余清祥 (1998), “統計在紅樓夢的應用”, 政大學報, 76, 303-327.
英文部分
1. Agresti, A. (2007), An Introduction to Categorical Data Analysis, 2nd ed., John Wiley & Sons, Inc.
2. Boser, B. E., Guyon, I.M., Vapnik, V. N. (1992), “A training algorithm for optimal margin classifiers”, Proceedings of the fifth annual workshop on Computational learning theory, 144-152.
3. Cortes, C. & Vapnik, V. (1995), “Support-vector network”, Machine Learning, 20, 1-25.
4. Drucker, P. F. (1999), “Beyond the information revolution”, The Atlantic Monthly, 284, 47-59
5. Meyer, D. (2009), “Support Vector Machines: The Interface to libsvm in package e1071”, Technische Universität Wien, Austria.
6. Page, L., Brin, S., Motwani, R. and Winograd, T. (1998), “The PageRank citation ranking: Bringing order to the Web”, Standford Digital Library Technologies Project.
7. Reed, W. J. (2001), “The Pareto, Zipf and other power laws”, Economics Letters 74 (1), 15–19.
8. Shannon, C. E. (1948), “A mathematical theory of communication”, Bell System Technical Journal, 27, 379-423, 623-656.
9. Sharma, S (1996), Applied Multivariate Techniques, John Wiley & Sons, Inc.
10. Simpson, E. H. (1949), “Measurement of diversity”, Nature, 163, 688.
11. Yue, C. J. and Clayton, M. K. (2005), “A Similarity Measures based on Species Proportions”, Communications in Statistics: Theory and Methods, 34, 2123-2131.
12. Wikipedia, Web search engine, http://en.wikipedia.org/wiki/Web_search_engine (as of June 15, 2009).
13. Zipf, G. K. (1935), The Psychobiology of Language: An Introduction to Dynamic Phinology, Houghton-Mifflin.
14. Zipf, G. K. (1949), Human behavior and the principle of least effort: An introduction to human ecology, Addison-Wesley, Cambridge, MA.
描述 碩士
國立政治大學
統計研究所
96354005
97
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0096354005
資料類型 thesis
dc.contributor.advisor 余清祥zh_TW
dc.contributor.advisor Yue, Jack C.en_US
dc.contributor.author (Authors) 高靖翔zh_TW
dc.contributor.author (Authors) Kao, Ching Hsiangen_US
dc.creator (作者) 高靖翔zh_TW
dc.creator (作者) Kao, Ching Hsiangen_US
dc.date (日期) 2008en_US
dc.date.accessioned 18-Sep-2009 20:10:38 (UTC+8)-
dc.date.available 18-Sep-2009 20:10:38 (UTC+8)-
dc.date.issued (上傳時間) 18-Sep-2009 20:10:38 (UTC+8)-
dc.identifier (Other Identifiers) G0096354005en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/36927-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 統計研究所zh_TW
dc.description (描述) 96354005zh_TW
dc.description (描述) 97zh_TW
dc.description.abstract (摘要) 由於電腦科技的快速發展,網際網路(World Wide Web;簡稱WWW)使得資料共享及搜尋更為便利,其中的網路搜尋引擎(Search Engine)更是尋找資料的利器,最知名的「Google」公司就是藉由搜尋引擎而發跡。網頁搜尋多半依賴各網頁的特徵,像是熵(Entropy)即是最為常用的特徵指標,藉由使用者選取「關鍵字詞」,找出與使用者最相似的網頁,換言之,找出相似指標函數最高的網頁。藉由相似指標函數分類也常見於生物學及生態學,但多半會計算兩個社群間的相似性,再判定兩個社群是否相似,與搜尋引擎只計算單一社群的想法不同。
本文的目標在於研究若資料服從多項分配,特別是似幾何分配的多項分配(許多生態社群都滿足這個假設),單一社群的指標、兩個社群間的相似指標,何者會有較佳的分類正確性。本文考慮的指標包括單一社群的熵及Simpson指標、兩社群間的熵及相似指標(Yue and Clayton, 2005)、支持向量機(Support Vector Machine)、邏輯斯迴歸等方法,透過電腦模擬及交叉驗證(cross-validation)比較方法的優劣。本文發現單一社群熵指標之表現,在本文的模擬研究有不錯的分類結果,甚至普遍優於支持向量機,但單一社群熵指標分類法的結果並不穩定,為該分類方法之主要缺點。
zh_TW
dc.description.abstract (摘要) Since computer science had changed rapidly, the worldwide web made it much easier to share and receive the information. Search engines would be the ones to help us find the target information conveniently. The famous Google was also founded by the search engine. The searching process is always depends on the characteristics of the web pages, for example, entropy is one of the characteristics index. The target web pages could be found by combining the index with the keywords information given by user. Or in other words, it is to find out the web pages which are the most similar to the user’s demands. In biology and ecology, similarity index function is commonly used for classification problems. But in practice, the pairwise instead of single similarity would be obtained to check if two communities are similar or not. It is dislike the thinking of search engines.
This research is to find out which has better classification result between single index and pairwise index for the data which is multinomial distributed, especially distributed like a geometry distribution. This data assumption is often satisfied in ecology area. The following classification methods would be considered into this research: single index including entropy and Simpson index, pairwise index including pairwise entropy and similarity index (Yue and Clayton, 2005), and also support vector machine and logistic regression. Computer simulations and cross validations would also be considered here. In this research, it is found that the single index, entropy, has good classification result than imagine. Sometime using entropy to classify would even better than using support vector machine with raw data. But using entropy to classify is not very robust, it is the one needed to be improved in future.
en_US
dc.description.tableofcontents 第一章 前言 1
第二章 文獻探討 3
第一節 搜尋引擎原理 3
第二節 索引指標 5
一、Simpson指標 6
二、熵 6
三、成對樣本相似指標 7
四、成對樣本熵 8
第三節 機器學習 8
一、邏輯斯迴歸分析 9
二、支持向量機 9
第三章 使用資料與模擬設定 11
第一節 似幾何分配資料與分類模擬設定 11
第二節 Zipf’s Law資料與分類模擬設定 13
第三節 紅樓夢實證資料之介紹 15
第四章 模擬結果與實證研究 17
第一節 似幾何分配下之分類模擬結果 17
一、分類結果與設定討論 17
二、變數選取與子集討論 21
三、多物種資料延伸討論 24
第二節 Zipf’s Law下之分類模擬結果 28
一、分類結果與設定討論 28
二、變數選取與子集討論 31
三、多物種資料延伸討論 35
第三節 紅樓夢實證之分類結果 39
一、分類結果 39
二、變數選取與子集討論 40
第五章 結論與建議 42
第一節 結論 42
第二節 建議 42
參考文獻 44
附錄 46
附錄一 似幾何分配其他相關附錄資料 46
附錄二 Zipf’s Law相關附錄資料 53
附錄三 紅樓夢相關附錄資料 60
zh_TW
dc.format.extent 95625 bytes-
dc.format.extent 120594 bytes-
dc.format.extent 165681 bytes-
dc.format.extent 307664 bytes-
dc.format.extent 139546 bytes-
dc.format.extent 266137 bytes-
dc.format.extent 308082 bytes-
dc.format.extent 2006549 bytes-
dc.format.extent 1975195 bytes-
dc.format.extent 263717 bytes-
dc.format.extent 177603 bytes-
dc.format.extent 169392 bytes-
dc.format.extent 3959073 bytes-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.format.mimetype application/pdf-
dc.language.iso en_US-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0096354005en_US
dc.subject (關鍵詞) 多項分配zh_TW
dc.subject (關鍵詞) zh_TW
dc.subject (關鍵詞) 相似指標zh_TW
dc.subject (關鍵詞) 電腦模擬zh_TW
dc.subject (關鍵詞) 支持向量機zh_TW
dc.subject (關鍵詞) 冪次定理zh_TW
dc.subject (關鍵詞) Multinomial distributionen_US
dc.subject (關鍵詞) Entropyen_US
dc.subject (關鍵詞) Similarity indexen_US
dc.subject (關鍵詞) Computer simulationen_US
dc.subject (關鍵詞) Support vector machineen_US
dc.subject (關鍵詞) Power Lawen_US
dc.subject (關鍵詞) Zipf`s Lawen_US
dc.title (題名) 多項分配之分類方法比較與實證研究zh_TW
dc.title (題名) An empirical study of classification on multinomial dataen_US
dc.type (資料類型) thesisen
dc.relation.reference (參考文獻) 中文部分zh_TW
dc.relation.reference (參考文獻) 1. 余清祥 (1998), “統計在紅樓夢的應用”, 政大學報, 76, 303-327.zh_TW
dc.relation.reference (參考文獻) 英文部分zh_TW
dc.relation.reference (參考文獻) 1. Agresti, A. (2007), An Introduction to Categorical Data Analysis, 2nd ed., John Wiley & Sons, Inc.zh_TW
dc.relation.reference (參考文獻) 2. Boser, B. E., Guyon, I.M., Vapnik, V. N. (1992), “A training algorithm for optimal margin classifiers”, Proceedings of the fifth annual workshop on Computational learning theory, 144-152.zh_TW
dc.relation.reference (參考文獻) 3. Cortes, C. & Vapnik, V. (1995), “Support-vector network”, Machine Learning, 20, 1-25.zh_TW
dc.relation.reference (參考文獻) 4. Drucker, P. F. (1999), “Beyond the information revolution”, The Atlantic Monthly, 284, 47-59zh_TW
dc.relation.reference (參考文獻) 5. Meyer, D. (2009), “Support Vector Machines: The Interface to libsvm in package e1071”, Technische Universität Wien, Austria.zh_TW
dc.relation.reference (參考文獻) 6. Page, L., Brin, S., Motwani, R. and Winograd, T. (1998), “The PageRank citation ranking: Bringing order to the Web”, Standford Digital Library Technologies Project.zh_TW
dc.relation.reference (參考文獻) 7. Reed, W. J. (2001), “The Pareto, Zipf and other power laws”, Economics Letters 74 (1), 15–19.zh_TW
dc.relation.reference (參考文獻) 8. Shannon, C. E. (1948), “A mathematical theory of communication”, Bell System Technical Journal, 27, 379-423, 623-656.zh_TW
dc.relation.reference (參考文獻) 9. Sharma, S (1996), Applied Multivariate Techniques, John Wiley & Sons, Inc.zh_TW
dc.relation.reference (參考文獻) 10. Simpson, E. H. (1949), “Measurement of diversity”, Nature, 163, 688.zh_TW
dc.relation.reference (參考文獻) 11. Yue, C. J. and Clayton, M. K. (2005), “A Similarity Measures based on Species Proportions”, Communications in Statistics: Theory and Methods, 34, 2123-2131.zh_TW
dc.relation.reference (參考文獻) 12. Wikipedia, Web search engine, http://en.wikipedia.org/wiki/Web_search_engine (as of June 15, 2009).zh_TW
dc.relation.reference (參考文獻) 13. Zipf, G. K. (1935), The Psychobiology of Language: An Introduction to Dynamic Phinology, Houghton-Mifflin.zh_TW
dc.relation.reference (參考文獻) 14. Zipf, G. K. (1949), Human behavior and the principle of least effort: An introduction to human ecology, Addison-Wesley, Cambridge, MA.zh_TW