LED的製造生產流程需經過多道的製程加工，在每道製程結束後產線人員會針對其加工站點進行單項製程的簡易量測，以確認產品是否能繼續生產。當LED產品完成所有製程加工後，會進行晶圓允收測試，在這項量測中會得知晶圓的點測良率，若當點測良率開始下滑，則代表前段製程機台開始出現異常，進而影響到產品產出。
過往，工程師都是自行透過MES系統先撈取晶圓點測良率，當點測良率開始下滑時才進一步去分析產品生產資料，並逐一去排除可能發生異常的站點及機台，這樣的過程不但花費了過多時間，也造成經驗無法適當傳承。
本研究將採用資料探勘(Data Mining)中的關聯法則分析(Association Rule)及決策樹(Decision Trees)來解決上述問題，並以台灣某一LED廠為例。

LED manufacturing has several production processes. At the end of each process, the operator of that process will perform simple test to verify whether the product is qualified to proceed with next step. After completion of the whole processes, the wafer acceptance test will be carried out for probing yield. If the probing yield is decreased, it indicates an error occurred at the front-end tools during operation, thereby affecting product output.
In the past, engineers retrieve the data of wafer probing yield through MES system on their own. Only when the data shows declining yield, the engineers start to analyze the production information to rule out possible abnormal processes or tools. By doing so, it spends too much time and can’t accumulate experiences.
The study will use Association Rule and Decision Tree in data mining techniques to solve the above problems based on a case study of a Taiwan LED manufacturer.

中文文獻
[1] 中華民國光電學會，LED工程師基礎概念與應用，中華民國一百零一年4月
[2] 林原慶，中國大陸LED產業特輯，中華民國一百年10月
[3] 莊達人，VLSI 製造技術，中華民國八十七年3月
[4] 簡禎富、許嘉裕，資料挖礦與大數據分析，前程文化，中華民國一百零五年2月
學位論文
[5] 吳鴻志，「運用關連式法則分析異常WAT良率之機台組合」，國立中正大學，碩士論文，中華民國九十年
[6] 林政傑，「運用序列探勘技術分析LED產業之最佳機台組合」，國立高雄應用科技大學，碩士論文，中華民國九十七年
[7] 張家銘，「以CART決策樹方法建構晶圓良率值與晶圓允收參數間之關聯性模式」國立清華大學工業工程與工程管理學系，碩士論文，中華民國九十五年
[8] 郭信宏，「應用資料探勘技術於面板檢測實證研究」，國立中央大學，碩士論文，中華民國九十六年
[9] 楊景晴，「整合決策樹與關聯規則之資料挖礦架構及其實證研究」，國立清華大學工業工程與工程管理學系，碩士論文，中華民國九十二年
[10] 簡協修，「運用資料探勘技術於製程品質分析改善-以LCD驅動IC封裝產業為例」，國立交通大學，碩士論文，中華民國九十三年
[11] 謝宗霖，「資料探勘於半導體測試工程之應用」，國立高雄第一科技大學，碩士論文，中華民國九十五年
英文文獻
[12] Aggarwal, C. C. (Ed.),”Data classification: algorithms and applications”, CRC Press, 2014.
[13] Agrawat, R., Imietinski, T., & Swami, A., "Mining Association Rules between Sets of Items in Very Large Database", Proc eedinng of the ACMSIGMOD Conference on Management of data, 1993, pp.207-216.
[14] Bettini, C., Wang, X. S., Jajodia, S., & Lin, J. L., ”Discovering frequent event patterns with multiple granularities in time sequences”, IEEE Transactions on Knowledge and Data Engineering, Vol.10, Issue2, 1998, pp.222-237.
[15] Breiman, L., Friedman, J., Stone, C. J., & Olshen, R. A., “Classification and regression trees”, CRC press, 1984.
[16] Cao, L., Zhang, H., Zhao, Y., Luo, D., & Zhang, C., “ Combined mining: discovering informative knowledge in complex data”, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), Vol.41, Issue3, 2011, pp.699-712.
[17] Fayyad, U., Piatetsky-Shapiro, G., & Smyth, P., ” From data mining to knowledge discovery in databases”, AI magazine,Vol.17, No.3, 1996, pp.37-54.
[18] Fayyad, U. M., ”Data mining and knowledge discovery: Making sense out of data”,
IEEE Expert: Intelligent Systems and Their Applications, Vol.11, Issue5, 1996, pp.20-25.
[19] Frawley, W. J., Piatetsky-Shapiro, G., & Matheus, C. J., ”Knowledge discovery in databases: An overview”, AI magazine,Vol.3, No.3, 1992, pp.57-70.
[20] Grupe, F. H., & Mehdi Owrang, M. , ”Data base mining discovering new knowledge and competitive advantage”, Information System Management, Vol.12, Issue4, 1995, pp.26-31.
[21] Hessinger, U., Chan, W. K., & Schafman, B. T., ”Data mining for significance in yield-defect correlation analysis”, IEEE Transactions on Semiconductor Manufacturing, Vol.27, Issue3, 2014, pp.347-356.
[22] Kittler, J., Hatef, M., Duin, R. P., & Matas, J., ”On combining classifiers”, IEEE transactions on pattern analysis and machine intelligence, Vol.20, Issue3, 1998, pp.226-239.
[23] Kureshi, N., Abidi, S. S. R., & Blouin, C., “ A predictive model for personalized therapeutic interventions in non-small cell lung cancer”, IEEE journal of biomedical and health informatics, Vol.20, Issue1, 2016, pp.424-431.
[24] Lee, C., Song, B., & Park, Y.,” Design of convergent product concepts based on functionality: An association rule mining and decision tree approach”, Expert Systems with Applications, Vol.39, Issue10, 2012, pp.9534-9542.
[25] Lenard, M. J., Madey, G. R., & Alam, P. , ” The design and validation of a hybrid information system for the auditor’s going concern decision”, Journal of Management Information Systems, Vol.14, Issue4, 1998, pp.219-237.
[26] Li, Z., Li, L., Yan, K., & Zhang, C. Automatic image annotation using fuzzy association
rules and decision tree. Multimedia Systems, 2016, pp.1-12.
[27] Ma, H., ”A Study on Customer Segmentation for E-Commerce Using the Generalized Association Rules and Decision Tree”, American Journal of Industrial and Business Management, Vol.5, No.12, 2015, pp.813.
[28] Michael, J. A., & Gordon, S. L., ” Data mining technique: For marketing, sales and customer support”, New York: John Wiley&Sons INC, 1997, pp.445.
[29] Simoudis, E., ”Reality check for data mining”, IEEE Expert: Intelligent systems and their applications,Vol.11, Issue5, 1996, pp.26-33.
[30] Wang, C. H., & Chuang, J. J. , ”Integrating decision tree with back propagation network to conduct business diagnosis and performance simulation for solar companies”, Decision Support Systems, Vol.81, 2016, pp.12-19.
[31] Yadav, N., Rajeswari, K., & Vaithiyanathan, V. V., ”Usage of Classification based Association for Removal of Noisy Attributes”, International Journal of Computer Applications, Vol.108, No.13, 2014.
[32] Yen, S. J., & Lee, Y. S., ”A neural network approach to discover attribute dependency for improving the performance of classification”, Expert Systems with Applications, Vol.38, Issue10, 2011, pp.12328-12338.
[33] Zadeh, L. A., ”Fuzzy sets”, Information and control, Vol.8, Issue3, 1965, pp338-353.
網路資料
[34] 科技網(2017)， http://www.digitimes.com.tw/tech/rpt/rpt_show.asp?cnlid=3&v=20170103-1&n=1