臺灣博碩士論文加值系統

農作物交易價、量會因季節、氣候的變遷、消費行為、政策因素等多種因素而波動。本研究以分析全台農產品批發市場之交易價量資料作為研究標的。發展一套農作物的價量(二維)關係為基礎之分群方式。此方法透過資料正規化找出農產品之特徵值來代表每種作物，再以階層式分群法探討農作物彼此之間的相似性。結果發現農作物之價量關係具有數個特定的價量變動特徵，依此了解及分析農作物價量變動趨勢。本研究將558種作物依兩階段分群方法進行分群，第一階段為依據價量的外型特徵進行分群，以價量關係外型的相似程度作為該階段的分群結果；第二階段再依第一階段的分群結果，依交易價格和交易量進行分群，實驗結果發現兩階段分群方法與一階段分群方法相比，兩階段分群方法的SSE較小。希望藉由此模型，將農產品的價量資訊揭露，提供消費者購買時和經營者進貨時的參考依據。

The sale prices and transaction volumes of agricultural crops are fluctuated depending on multiple factors such as seasonal effect, climate change, consumer purchasing behavior, and so on. Analyzing the dynamic between sale price and transaction volume of agricultural goods can enhance the retailers’ capability of understanding the market trend. This research developed a two-step clustering method to conduct the data clustering on agriculture crop price-volume (P-V) data. The first step, for each crop, we proposed the p-v shape characteristics including the average, largest, and smallest price and volume values, the slope of p-v distribution, and R square of p-v linear regression to represent the shape of p-v distribution. The hierarchical clustering method was applied on the developed characteristic dataset to cluster 558 crops to several groups. The second step continued to perform the hierarchical clustering method on the price and volume dataset of each group to study the p-v seasonal pattern. The experimental results show the two-stage clustering method is able to have smaller sum of square error (SEE) while comparing with one-stage method. Except the better clustering performance, the proposed two-stage method also has the managerial advantage on providing easy-to-read clustering structure. The agriculture crop p-v clustering result can be provided to either customers or retailers for better information sharing and proactive marketing strategy.

致 謝 I
中文摘要 II
ABSTRACT III
目 錄 IV
表目錄 VI
圖目錄 VIII
第1章 緒論 1
1.1 研究動機 1
1.2 研究目的 5
1.3 本文架構 6
第2章 文獻探討 7
2.1 農產品價量關係 7
2.2 群集分析方法 8
2.3 相關分群研究 14
第3章 研究設計與方法 17
3.1 資料介紹 17
3.2 資料前處理 18
3.3 特徵值選取 21
3.3.1 第一階段 22
3.3.2 第二階段 28
第4章 實驗結果 31
4.1 第一階段分群結果 31
4.2 第二階段分群結果 41
4.2.1 群集1 41
4.2.2 群集2 46
4.2.3 群集3 51
4.2.4 群集4 55
4.3 結果探討 62
第5章 結論與未來發展 65
5.1 結論 65
5.2 未來研究發展與建議 66
參考文獻 67
附錄 69

Aşcıoğlu, A., Comerton-Forde, Carole, M., &; H., T. (2007). Price Clustering on the Tokyo Stock Exchange. Financial Review, 42(2), 289-301. doi: 10.1111/j.1540-6288.2007.00172.x
Batagelj, V., Bock, H.-H., Ferligoj, A., &; Žiberna, A. (2006). Data Science and Classification.
Berkhin, P. (2006). A Survey of Clustering Data Mining Techniques: Springer Berlin Heidelberg.
Chia-Hsuan, C. (2013). Genetic Algorithm-Based Multiobjective Fuzzy Clustering Method on Categorical Data (Master), National Taiwan University of Science and Technology
Dent, I., Aickelin, U., &; Rodden, T. (2013). Application of a Clustering Framework to UK Domestic Electricity. 1.
Gan, G., Ma, C., &; Wu, J. (2007). Data Clustering: Theory, Algorithms, and Applications: SIAM, Society for Industrial and Applied Mathematics.
Google, t. w. (2013). How Mobile is Transforming the Shopping Experience in Stores. from http://www.thinkwithgoogle.com/research-studies/mobile-in-store.html
Hatamlou, A., &; Abdullah, S. (2011). A Two-Stage Algorithm for Data Clustering.
Hu, Y.-C. (2005). Discovering of Quarterly Equity Styles with Clustering. (Master), National Central University.
Hubert, L., &; Arabie, P. (1985). Comparing Partitions. Journal of Classification, 2(1), 193-218. doi: 10.1007/BF01908075
Ka Yee Yeung, &; Ruzzo, W. L. (2001). Details of the Adjusted Rand Index and Clustering Algorithms Supplement to the Paper “An Empirical Study on Principal Component Analysis for Clustering Gene Expression Data” (to appear in Bioinformatics).
lai, Y.-s., &; Chang, D. H.-K. (2012). Cluster Analysis on the Paths and Scales of Typhoons Driving to Taiwan. (Master), National Chiao Tung University.
Li, X.-B., &; Jacob, V. S. (2008). Adaptive Data Reduction for Large-Scale Transaction Data. European Journal of Operational Research, 188(3), 910-924. doi: http://dx.doi.org/10.1016/j.ejor.2007.08.008
Liu, Y., &; Zhang, H. (2009). Analyzing Agricultural Economy Data with Computational Intelligence Method Paper presented at the International Conference on Computational Intelligence and Software Engineering (CiSE 2009), Wuhan.
Malliaris, A. G., &; L.Urrutia, J. (1998). Volume and Price Relationships: Hypotheses and Testing for Agricultural Futures. The Journal of Futures Markets, 18(1), 53-72.
Marcotte, &; Claude. (2012). Measuring Entrepreneurship at the Country Level: A Review and Research Agenda. Entrepreneurship &; Regional Development, 25(3-4), 174-194. doi: 10.1080/08985626.2012.710264
Prayag, G., Disegna, M., Cohen, S. A., &; Yan, H. (2013). Segmenting Markets by Bagged Clustering: Young Chinese Travelers to Western Europe. Journal of Travel Research, 1-17. doi: 10.1177/0047287513514299
Punj, G., &; Stewart, D. W. (1983). Cluster Analysis in Marketing Research: Review and Suggestions for Application. Journal of Marketing Research, 20(2), 134-148. doi: 10.2307/3151680
Shen, J., &; Li, W. (2013). Cluster Analysis of Larger-Scale Discrete Data With Application to Estimating Dwell Time of Bus Route. Pakistan Journal of Statistics, 29(5), 873-886.
Tan, P.-N., Steinbach, M., &; Kumar, V. (2005). Introduction to Data Mining (1 edition ed.): Addison-Wesley.
Visalakshi, N. K., &; Thangavel, K. (2009). Impact of Nomalization in Distributed K-Means Clusteing. International Journal of Soft Computing, 4, 168-172.
行政院農業委員會. (2011). 糧食供需重要統計指標. from http://www.coa.gov.tw/view.php?catid=8
呂秀英. (2006). 多變數分析在農業科技之應用. Paper presented at the Crop, Environment &; Bioinformatics.
林啟賢. (2013). 大學優秀籃球運動員的目標取向及意象使用. 大學體育學刊, 第15卷 第2期, 195-207. doi: 10.5279/ser.1502.007
陳順宇. (2004). 多變量分析. 台灣: 華泰文化, 6-1-53
楊新輝. (1998). 台北市第一果菜批發市場價量關係之研究. (碩士), 國立台灣大學.
網摘, I. (2013). 四大情境告訴你蘋果 iBeacon 技術的無限可能. from http://share.inside.com.tw/posts/3442