Have library access?
IP:3.92.91.54
  • Theses

基於不同連結策略之集群式預測模式於伺服器銷售預測

Computer server sales forecasting using cluster-based forecasting model with different linkage strategies

Advisor : 呂奇傑
For better promotion, authorized us if you are the author.

Abstracts


隨著智慧型通訊裝置以及網路的蓬勃發展,讓伺服器市場成長且愈來愈重要。由於伺服器產品需求具備了前置時間長及高單價的特性,因此如何準確的預測伺服器銷售就變成重要的課題。本研究基於不同連結策略建構集群式預測模式於伺服器銷售預測。本研究所提方法先使用K-means集群技術將訓練資料予以適當分群,並針對每一個集群以支援向量迴歸(Support vector regression, SVR)及極限學習機(Extreme learning machine, ELM)建構預測模式。接著針對待預測資料,分別以中心法(Centroid linkage method)、中位數法(Median linkage method)以及遠鄰法(Furthest neighbor linkage)等三個不同的連結法(Linkage method)計算待預測資料與訓練資料子集群所對應的距離,以找到最合適於待預測資料的對應子集群,最後以對應子集群的SVR及ELM模式產生最後的預測結果。本研究以台灣某伺服器品牌商的實際伺服器銷售資料為實證資料。實證結果顯示結合遠鄰法與SVR的集群式預測模式能產生最佳的預測績效,較單純SVR、單純ELM、其他五個集群式銷售預測模式,以及實務運作上常見Seasonal Naive方法更適用於伺服器銷售預測。

Parallel abstracts


Sales forecasting is crucial for every company since it is an important task for manufacturing, inventory management and marketing. In this study, a computer server sales forecasting model using clustering method with support vector regression (SVR) and extreme learning machine (ELM) with different linkage strategies is proposed. The proposed scheme first uses k-means algorithm to partition the whole training sales data into several disjoint clusters. Then, for each group, the SVR and ELM is applied to construct forecasting model. Finally, for a given testing data, three linkage methods are used to find the cluster which the testing data belongs to and then employee the corresponding trained SVR model and ELM model to generate prediction result. A real data of computer server sales collected from a Taiwanese multinational electronics company is used as illustrative examples to evaluate the performance of the proposed model. Experimental results revealed that the proposed clustering-based sales forecasting scheme outperforms the single method and seasonal naive forecasting models and hence is an effective alternative for sales forecasting.

References


1. 吳萬益,企業研究方法,華泰文化出版社,台北,民國一百年。
2. 陳東和,「運用資料採礦技術於銀行基金客戶分群之研究」,私立中國文化大學資訊管理研究所,碩士論文,台北市,民國九十七年。
3.Badge, J., & Srivastava, N. “Selection and forecasting of stock market patterns using K-mean clustering”, International Journal of Statistics and Systems, 5, pp. 23-27,
2010.
4. Cao, L. “Support vector machines experts for time series forecasting”, Neurocomputing, 51, pp. 321-339, 2003.
5. Chang, P. C., & Lai, C. Y. “A hybrid system combining self-organizing maps with case- based reasoning in wholesaler's new-release book forecasting”, Expert Systems with Applications, 29(1), pp. 183-192 , 2005.

Read-around