臺灣博碩士論文加值系統

企業社會責任(Corporate Social Responsibility, CSR)於近二十年間受到世人重視，編制企業社會責任報告書也成為國內外企業的趨勢，在瞬息萬變的競爭環境中，企業如何扮演世界公民之角色，在獲利、環境及慈善等方面中取得平衡，將是社會大眾所矚目。然而，以往的相關研究多以量化的角度去分析企業社會責任對於財務績效等面向的表現，資料探勘技術尚未廣泛用於此領域。因此，本研究以資料探勘的視角，探討企業社會責任各面向對企業之重要性，提出一個混合的資料探勘模型—CSFSC模型，包括資料前處理、分類方法及規則擷取技術。其中，資料前處理技術包含基於相關性為基礎之特徵選取技術(Correlation-based Feature Selection, CFS)、合成少數類別技術(Synthetic Minority Over-sampling Technique , SMOTE)及Fuzzy C-Means(FCM)分群演算法，而一對一支援向量機(One-Against-One Support Vector Machine, OAOSVM)則作為多元分類方法，最後，以C5.0決策樹作為規則擷取技術。本研究使用2010年中國大陸A股上市公司企業社會責任報告書為資料來源，來測試本研究提出之模型的效能，其實驗結果顯示CSFSC模型有很好的分類準確率，並可提供清楚且有效的決策規則給予企業決策者，有助於提升企業社會責任之評級。因此，可證明在分析CSR議題方面，CSFSC模型是一個有效模型。

Over the past two decades, Corporate Social Responsibility (CSR) has received worldwide attention. Publication of CSR Reports has become the trend for domestic and foreign enterprises. In the constantly changing competition environment, it will be focus of public attention that how enterprises to play the role of corporate citizenship and to achieve a balance in profit, environmental and charitable activities. However, most of previous quantitative studies of CSR concentrate on traditional statistic approaches. The data mining technique has not been widely explored in this area. Thus, this investigation proposed a hybrid data mining CSFSC model integrating data preprocessing approaches, a classification method, and a rule generation mechanism. The data preprocessing approaches include Correlation-based Feature Selection(CFS), Synthetic Minority Over-sampling Technique (SMOTE) and Fuzzy C-Means (FCM) clustering algorithm. The One-Against-One Support Vector Machine (OAOSVM) method was employed as a classifier for performing multi-classification task. The rule-based learning algorithm C5.0 was utilized to generate rules from the results of OAOSVM model. CSR data collected from China’s Listed Firms in 2010 were employed to examine the performance of the proposed model. The empirical results showed that the designed CSFSC model can yield satisfactory classification accuracy as well as provide rules for decision makers. Therefore, the presented CSFSC model is a feasible and effective alternative in analyzing CSR.

目錄

誌謝 i
中文摘要 ii
Abstract iv
目錄 vi
圖目錄 viii
表目錄 ix
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 2
1.3 論文架構 3
第二章 文獻探討 4
2.1企業社會責任(Corporate Social Responsibility, CSR) 4
2.2.基於相關性為基礎之特徵選取技術 12
2.3合成少數類別技術 14
2.4 Fuzzy C-Means分群演算法 16
2.5 One-against-One 支援向量機 19
2.6決策樹C5.0 21
第三章 研究方法 24
3.1研究架構 24
3.2特徵選取 26
3.2.1基於相關性為基礎之特徵選取技術 29
3.3合成少數類別技術 32
3.4 Fuzzy C-Means 分群演算法 34
3.5 One-against-One支援向量機 36
3.6決策樹C5.0 40
第四章 實驗結果與分析 43
4.1 資料來源 43
4.2 實驗結果 46
4.3二因子的變異數分析 51
第五章 結論與未來研究方向 55
參考文獻 56
附錄A CSFSC模型產出之決策規則 65
附錄B CSR報告書之屬性明細 70

圖目錄

圖一 論文架構圖 3
圖二 CSFSC模型流程架構圖 25
圖三 特徵選取流程 26
圖四 CFS運作流程 30
圖五 CFS+BFS特徵選取流程 31
圖六 使用SMOTE後的資料分布概念圖 33
圖七 使用Fuzzy分群資料分布圖示 34
圖八 SVM二維空間示意圖 36
圖九 One-Against-One投票法示意圖 39
圖十 C5.0決策樹修剪(prune)示意圖 40
圖十一 C5.0決策樹處理流程(廖述賢和溫志浩,2009) 41
圖十二 CSR報告書評級標準 44
圖十三 有無資料前處理之測試資料準確率長條圖 47
圖十四 CSFSC模型及邏輯斯迴歸分析之測試準確率長條圖 50
圖十五 二因子的變異數(Two-Way ANOVA)分析流程 51

表目錄

表一 企業社會責任量化相關文獻整理 7
表二 CFS特徵選取相關文獻整理 13
表三 SMOTE資料前處理相關文獻整理 15
表四 Fuzzy C-Means分群相關文獻整理 17
表五 一對一支援向量機分類相關文獻整理 20
表六 C5.0決策樹分類預測相關文獻整理 22
表七 資料類別分布表 43
表八 潤靈環球責任評級之一級指標 45
表九 有無資料前處理之測試資料準確率結果 46
表十 CFS特徵選取後之屬性 47
表十一 CFS與皮爾森相關係數選取之條件屬性對照表 48
表十二 CSFSC模型及邏輯斯迴歸分析之測試準確率比較表 49
表十三 二因子變異數(Two-Way ANOVA)分析表 52
表十四 簡單主要效果分析 53
表十五 “完整性”兩兩配對比較表 53
表十六 “完整性”兩兩配對比較表產生之決策規則 54
表十七 CSFSC模型產出之包含“利益相關方溝通訊息”及“完整性”的規則 54

[1]世界企業永續發展協會(World Business Council for Sustainability and Development,WBCSD)。檢索日期：2013.01.23。取自http://www.bcsd.org.tw/domain_node/domainnode/23
[2]商業社會責任(Business Social Responsibility,BSR)。檢索日期：2012.09.03。取自 http://www.bsr.org/
[3]遠見編輯部，(2008,4月)。第四屆遠見企業社會責任大調查 大企業領軍協助供應商做CSR。遠見，262，126~132。檢索日期：2013.01.16。取自http://www.gvm.com.tw/Boardcontent_14052.html
[4]廖述賢、溫志浩， (2009)。資料採礦與商業智慧，雙葉書廊有限公司，台北。
[5]潤靈環球責任評級（Rankins CSR Ratings, RKS）。檢索日期：2012.04.27。取自http://www.rksratings.com/
[6]Baden, D.A., Harwood, I.A., Woodward, D.G.(2009). The effect of buyer pressure on suppliers in SMEs to demonstrate CSR practices: An added incentive or counter productive?. European Management Journal 27, pp.429– 441.
[7]Bae, Jae Kwon, Kim. Jinhwa (2011). Product development with data mining techniques: A case on design of digital camera. Expert Systems with Applications Volume 38 Issue 8. pp. 9274–9280.
[8]Bezdek, J.C. (1981). Pattern Recognition with Fuzzy Objective Function Algorithms. Plenum Press, New York.
[9]Boli, J. & Hartsuijker, D. (2001). World Culture and Transnational Corporations: Sketch of a Project. International Conference on Effects of and Responses to Globalization.
[10]Bovolo, Francesca(2010). A Novel Technique for Subpixel Image Classification Based on Support Vector Machine. Image Processing, Vol. 19, No. 11, pp. 2983-2999.
[11]Bowen, H. R. (1953).Social responsibilities of the businessman. New York：Harper & Row.
[12]Breiman, L., Friedman, J.H., Olshen, R.A.(1984). Classification and Regression Trees. Wadsworth International Group, Belmont, California.
[13]Budayan, C., Dikmen, I., & Birgonul, M. T. (2009). Comparing the performance of traditional cluster analysis, self-organizing maps and fuzzy C-means method for strategic grouping. Expert Systems with Applications, 36,pp. 11772-11781.
[14]Bujlow, T., Riaz, T., Pedersen, J. M. (2012). A method for classification of network traffic based on C5.0 Machine Learning Algorithm. Computing, Networking and Communications (ICNC), 2012 International Conference on. pp.237-241.
[15]Carroll, Archie B. (1979, October). A three-dimensional conceptual model of corporate performance. Academy of Management Review 4(4), pp. 497-505.
[16]Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic Minority Over-sampling Technique. Artificial Intelligence Research 16, pp.321-357.
[17]Chen, Hsiang-Yang, Hou, Ting-Wei, Chuang, Chao-Hua , and TBPS Research Group. (2010). Applying data mining to explore the risk factors of parenting stress. Expert Systems with Applications Volume 37, Issue 1. pp. 598-601.
[18]Chen, T. (2007). Incorporating fuzzy c-means and a back-propagation network ensemble to job completion time prediction in a semiconductor fabrication factory. Fuzzy Sets and Systems 158, pp. 2153-2168.
[19]Chiu, Shih-Chi & Sharfman, Mark (2011, November). Legitimacy, Visibility, and the Antecedents of Corporate Social Performance: An Investigation of the Instrumental Perspective. Journal of Management Vol. 37 No. 6, pp.1558-1585.
[20]Choi, Jong-Seo, Kwak, Young-Min, Choe, Chongwoo (2010). Corporate social responsibility and corporate financial performance: Evidence from Korea. Australian Journal of Management 35(3), pp. 291–311.
[21]Consolandi, Costanza, Innocenti, Alessandro, Vercelli, Alessandro (2009). CSR, rationality and the ethical preferences of investors in a laboratory experiment. Research in Economics 63, pp.242–252.
[22]Dash, M. and Liu, H.. (1997). Feature Selection for Classification. Intelligent Data Analysis 1, pp.131–156.
[23]Davis, Keith and Blomstrom, Robert L. (1975). Business and Society: Environment and Responsibility (3rd ed.), McGraw-Hill Book Company, New York, pp.39.
[24]Duan, K., Keerthi, S., Poo, A.(2003, April).Evaluation of simple performance measures for tuning SVM hyperparameters. Technical Report, Department of Mechanical Engineering, pp. 41-59.
[25]Feng, Jie, Jiao, L.C., Zhang, Xiangrong, Gong, Maoguo, Sun, Tao (2013). Robust non-local fuzzy c-means algorithm with edge preservation for SAR image segmentation. Signal Processing 93, pp.487–499.
[26]Foroughi, Homa, Rezvanian, Alireza, Paziraee, Amirhossien (2008). Robust Fall Detection using Human Shape and Multi-class Support Vector Machine. Computer Vision, Graphics & Image Processing, 2008. ICVGIP '08. Sixth Indian Conference on, pp. 413-420.
[27]Freeman, R. E. Strategic Management: A Stakeholder Approach, Boston: Pitman. 1984.
[28]Friedman, M. (1970, September).The Social Responsibility of Business is to increase its Profits. The New York Times Magazine.
[29]Friedman, J. (1996). Another Approach to Polychotomous Classification. Dept. Statist., Stanford Univ., Stanford, CA. [Online].Available: http://www-stat.stanford.edu/reports/friedman/poly.ps.Z
[30]Galbreath, Jeremy & Shum, Paul (2012). Do customer satisfaction and reputation mediate the CSR−FP link? Evidence from Australia. Australian Journal of Management 37(2), pp. 211–229.
[31]Gao, Ming, Hong, Xia, Chen, Sheng, Harris, Chris J. (2011). A combined SMOTE and PSO based RBF classifier for two-class imbalanced problems. Neurocomputing 74, pp.3456–3466.
[32]Garay, Luis & Font, Xavier (2012). Doing good to do well? Corporate social responsibility reasons, practices and impacts in small and medium accommodation enterprises. International Journal of Hospitality Management 31.pp. 329–337.
[33]Ghaderi, Hamid & Kabiri, Peyman (2012). Fourier Transform and Correlation-based Feature Selection for Fault Detection of Automobile Engines. The 16th CSI International Symposium on Artificial Intelligence and Signal Processing (AISP 2012), pp. 514-519.
[34]Ghoul, S. E., Guedhami, O., Kwok, Chuck C.Y., Mishra, Dev R. (2011). Does corporate social responsibility affect the cost of capital?. Journal of Banking & Finance 35, pp. 2388–2406.
[35]Hall, Mark A., Smith, Lloyd A. (1999). Feature Selection for Machine Learning: Comparing a Correlation-based Filter Approach to the Wrapper. Proceedings of the Twelfth International FLAIRS Conference
[36]Hall, Mark A. (1999). Correlation-based feature subset selection for machine learning. PhD thesis. Department of Computer Science. University of Waikato. Hamilton. New Zealand.
[37]Hall, Mark, Frank, Eibe, Holmes, Geoffrey, Pfahringer, Bernhard, Reutemann. Peter, Witten, Ian H. (2009). The WEKA Data Mining Software: An Update; SIGKDD Explorations, Volume 11, Issue 1
[38]Hasan, Abid & Adnan, Md. A. (2012). High Dimensional Microarray Data Classification Using Correlation Based Feature Selection. 2012 International Conference on Biomedical Engineering (ICoBE), pp. 319-321.
[39]Hung, W. L., Yang, M. S., & Chen, D. H. (2008). Bootstrapping approach to feature-weight selection in fuzzy c-means algorithms with an application in color image segmentation. Pattern Recognition Letters 29, pp.1317-1325.
[40]Huseynov, Fariz & Klamm, Bonnie K. (2012). Tax avoidance, tax management and corporate social responsibility. Journal of Corporate Finance 18, pp.804–827
[41]Husted, B. W. & Allen, D. B. (2007). Strategic Corporate Social Responsibility and Value Creation among Large Firms:Lessons from the Spanish Experience. Long Range Planning 40, pp.594-610.
[42]Iliadis, L. S., Vangeloudh, M., & Spartalis, S. (2010). An intelligent system employing an enhanced fuzzy c-means clustering model: Application in the case of forest fires. Computers and Electronics in Agriculture, 70, pp.276-284.
[43]Inoue, Yuhei & Lee, Seoki (2011). Effects of different dimensions of corporate social responsibility on corporate financial performance in tourism-related industries. Tourism Management 32, pp.790-804.
[44]Kaur, Prabhjot, Soni, A.K., Gosain, Anjana (2012). A Robust Kernelized Intuitionistic Fuzzy C-Means Clustering Algorithm in Segmentation of Noisy Medical Images. Pattern Recognition Letters Vol. 34 Issue 2, pp.163–175.
[45]Karegowda, Asha Gowda & Jayaram, M.A. (2009). Cascading GA & CFS for Feature Subset selection in Medical Data Mining. 2009 IEEE International Advance Computing Conference (IACC 2009) Patiala, India, pp.6-7.
[46]Knerr, S., Personnaz, L. and Dreyfus, G. (1990). Single-layer learning revisited: A stepwise procedure for building and training a neural network.in Neurocomputing: Algorithms, Architectures and Applications, J. Fogelman, Ed. New York: Springer-Verlag.
[47]Kong, Dongmin (2012). Does corporate social responsibility matter in the food industry? Evidence from a nature experiment in China. Food Policy 37, pp.323–334.
[48]Li, Z. ( 2012). Lack of corporate social responsibility behind recent China accidents. Retrieved January 16, 2013, from http://www.worldwatch.org/node/3859
[49]Liu, Bo, Hao, Zhifeng and Tsang, Eric C. C. (2008). Nesting One-Against-One Algorithm Based on SVMs for Pattern Classification. Neural Networks, Vol. 19, No. 12, pp.2044-2052.
[50]Maciejewski, Tomasz & Stefanowski, Jerzy (2011). Local Neighbourhood Extension of SMOTE for Mining Imbalanced Data. Computational Intelligence and Data Mining (CIDM), pp.104-111.
[51]Mahjoobi, J. and Etemad-Shahidi, A., (2008). An alternative approach for the prediction of significant wave heights based on classification and regression trees. Applied Ocean Research Volume 30, Issue 3, pp.172-177.
[52]Maignan, Isabelle & Ferrell, O.C. (2003). Nature of corporate responsibilities Perspectives from American, French, and German consumers. Journal of Business Research 56, pp.55– 67.
[53]McWilliams, Abagail & Siegel, Donald (2000). Research Notes And Communications: Corporate Social Responsibility And Financial Performance: Correlation Or Misspecification? Strategic Management Journal Strat. Mgmt. J. 21, pp. 603-609.
[54]Melgani, F. (2004, August). Classification of hyperspectral remote sensing images with support vector machines. Geoscience and Remote Sensing, Vol. 42, No. 8, pp.1778-1790.
[55]Michalak , K and Kwasnicka, H. (2006). Correlation–Based Feature Selection Strategy in Classification Problems. Int. J. Appl. Math. Comput. Sci., Vol. 16, No. 4, pp. 503–511.
[56]Muata, K. & Bryson, O. (2007). Post-pruning in decision tree induction using multiple performance measures. Computers & Operations Research 34, pp. 3331-3345.
[57]Niblett, T. (1987). Constructing decision trees in noisy domains. in: Proceedings of the Second European Working Session on Learning. Sigma Press, Bled. pp.67-78.
[58]Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill.
[59]Özçift, Akin (2011). Random forests ensemble classifier trained with data resampling strategy to improve cardiac arrhythmia diagnosis. Computers in Biology and Medicine 41, pp. 265–271.
[60]Park, D. C. (2009). Classification of audio signals using Fuzzy c-Means with divergence-based Kernel. Pattern Recognition Letters 30, pp.794-798.
[61]PANG, Su-lin, GONG, Ji-zhang (2009). C5.0 Classification Algorithm and Application on Individual Credit Evaluation of Banks,” Systems Engineering - Theory & Practice Volume 29, Issue 12.pp. 94-104.
[62]Porter, Michael E., Kramer, Mark R. (2006, December). Strategy & Society - The Link Between Competitive Advantage and Corporate Social Responsibility and Environmental Management. Harvard Business Review (84:12), pp.78-92.
[63]Quinlan, J. R. (2003). C5.0 Online Tutorial. Retrieved January 26, 2013, from http://www.rulequest.com.
[64]Rahim, R. A., Jalaludin, F. W., Tajuddin, K. (2011, January). The Importance Of Corporate Social Responsibility On Consumer Behaviour In Malaysia. Asian Academy of Management Journal, Vol. 16, No. 1, pp. 119–139.
[65]Shannon, Claude E. (1948, July and October). A mathematical theory of communication. Bell System Technical Journal, vol. 27, pp. 379–423 and 623–656.
[66]Siegel, Donald S., Vitaliano, Donald F. (2007). An Empirical Analysis of the Strategic Use of Corporate Social Responsibility. Journal of Economics & Management Strategy, Vol.16, Number 3, pp.773–792.
[67]Smith, Victoria & Langford, Peter (2011). Responsible or redundant? Engaging the workforce through corporate social responsibility. Australian Journal of Management 36(3), pp. 425–447.
[68]Soman, K. P., Diwakar, Shyam, Ajay, V. (2000). Insight into Data Mining : Theory and Practice. Prentice Hall of India.
[69]Taft, L. M., Evans, R. S., Shyu, C. R., Egger, M. J., Chawla, N., Mitchell, J. A., et al. (2009). Countering imbalanced datasets to improve adverse drug event predictive models in labor and delivery. Journal of Biomedical Informatics 42, pp.356-364.
[70]Tan, K. S., & Isa, N. A. M. (2011). Color image segmentation using histogram thresholding – Fuzzy C-means hybrid approach. Pattern Recognition Letters 44, pp.1-15.
[71]Tang, Lu & Li, Hongmei (2009). Corporate social responsibility communication of Chinese and global corporations in China. Public Relations Review (35:3), pp.199–212.
[72]Usunier, Jean-Claude, Furrer, O., Furrer-Perrinjaquet, A. (2011). The perceived trade-off between corporate social and economic responsibility: A cross-national study. International Journal of Cross Cultural Management 11(3), pp. 279–302.
[73]Wang, He-Yong (2008). Combination Approach of SMOTE and Biased-SVM for Imbalanced Datasets. 2008 International Joint Conference on Neural Networks (IJCNN 2008), pp.228-231.
[74]Wang, Tai-Yue, Chiang, Huei-Min (2009). One-against-one fuzzy support vector machine classifier: An approach to text categorization. Expert Systems with Applications 36, pp.10030-10034.
[75]Wang, Yajuan, Simon, Marc, Bonde, Pramod, Harris, Bronwyn U., Teuteberg, Jeffrey J., Kormos, Robert L., Antaki, James F. (2012). Prognosis of Right Ventricular Failure in Patients With Left Ventricular Assist Device Based on Decision Tree With SMOTE. Information Technology In Biomedicine, Vol. 16, No. 3, pp.383-390.
[76]Weber, Manuela (2008).The business case for corporate social responsibility: A company-level measurement approach for CSR. European Management Journal (26:4). pp. 247-261.
[77]Wei, Chih-Chiang and Hsu, Nien-Sheng (2009). Optimal tree-based release rules for real-time flood control operations on a multipurpose multireservoir system. Journal of Hydrology Volume 365, Issues 3–4. pp.213-224.
[78]Wong, Jehn-Yih and Chung, Pi-Heng (2007). Managing valuable Taiwanese airline passengers using knowledge discovery in database techniques. Journal of Air Transport Management Volume 13, Issue 6, pp.362-370.
[79]Zairi, Mohamed, & Peters, John (2002). The Impact of Social Responsibility on Business Performance. Managerial Auditing Journal (17:4), pp. 174-178.