臺灣博碩士論文加值系統

銷售預測對製造業而言是編製預算時很重要的基石，公司未來的營運方向都有賴銷售預測的正確度。在製造業的經營管理中，管理者是根據銷售預測來訂定未來生產與採購計畫，以避免生產過剩或不足的問題，可見建立正確的銷售預測系統是很重要的。本研究收集2011年至2017年台灣兩家汽車製造商每月銷售量、經濟變數、Google Trends與意見分數(opinion score)的特徵值後，使用灰色關聯分析(Grey relational analysis, GRA) 篩選重要經濟變數，並運用決定係數(coefficient of determination，R2)作為意見分數評選的標準，設計出多種變數組合，當作最小平方支援向量迴歸(Least Square Support Vector Regression, LSSVR)的輸入變數，並以粒子群演算法(particle swarm optimization, PSO)優化，建立混合的汽車銷售量預測模型(稱為PSO-LSSVR模型)。PSO-LSSVR透過不同的變數設計，期找出預測效能較佳的變數組合，並採用MAPE(Mean Absolute Percentage Error)計算結果進行評比。實驗結果顯示，經過GRA篩選過後的經濟變數，Google Trend 及 使用R2分析後的意見分數可以增進銷售預測的效能。

Sales forecasting is a vital cornerstone of a company's budget in manufacturing industries. The future direction of the company may rest on the accuracy of sales forecasting. In manufacturing firms, product demand is forecast for the upcoming production period, and the production is planned in accordance with the forecast to avoid inventory shortage or excess. Therefore, sales forecasting is crucial for manufacturing industries. In this study—which used the monthly sales volumes of two car manufacturers in Taiwan, economic variables analyzed using grey theory (GRA), and opinion scores of sentiment analysis analyzed by coefficient of determination (R2), and the Google Trend from 2011 to 2017 as the research data—Least Square Support Vector Regression (LSSVR) optimized by Particle Swarm Optimization (PSO) was applied to forecast monthly car sales volume. The selected features together with Google Trend and analyzed opinion score are then used as the inputs to the LSSVR to build the various models. Finally, to increase model accuracy, the parameter values of the LSSVR are optimized through PSO. The mean absolute percentage Error (MAPE) is adopted to evaluate the forecasting accuracy. Our experimental results indicate that the most precise forecasts are given as a result of using the selected economic variables by GRA, analyzed opinion scores by R2, and Google Trend.

目錄

書名頁………………………………………………………………………………………………….i
論文口試委員審定書………………………………………………………………………ii
授權書………………………………………………………………………………………………….iv
中文摘要………………………………………………………………………………………………v
英文摘要………………………………………………………………………………………………vi
誌謝…………………………………………………………………………………………………….vii
目錄…………………………………………………………………………………………………….viii
表目錄………………………………………………………………………………………………………x
圖目錄.……………………………………………………………………………………………….xi
第一章 緒論…………………………………………………………………………………….1
第二章 文獻探討………………………………………………………………………………4
2.1 銷售預測…………………………………………………………………………4
2.2 預測的方法……………………………………………………………………5
第三章 研究方法………………………………………………………………………….17
3.1 研究架構………………………………………………………………………18
3.2 各階段工作說明…………………………………………………………19
3.2.1 第一階段…………………………………………………………………….19
3.2.2 第二階段…………………………………………………………………….21
3.2.3 第三階段…………………………………………………………………….23
3.2.3.1 文章資料前處理…………………………………………………23
3.2.3.2 意見傾向處理………………………………………………………24
3.2.4 第四階段…………………………………………………………………….25
3.2.4.1 最小平方支援向量迴歸……………………………………25
3.2.4.2 粒子群最佳化演算法…………………………………………27
3.2.4.3 PSO-LSSVR……………………………………………………………29
3.2.5 第五階段…………………………………………………………………….30
3.2.5.1 進行汽車銷售量模型實驗………………………………30
3.2.5.2 預測模型效能評估方式……………………………………32
第四章 實驗設計與結果…………………………………………………………………33
4.1 資料說明……………………………………………………………………………33
4.2 實驗結果……………………………………………………………………………33
4.2.1 經濟變數與灰關聯分析……………………………………….33
4.2.2 Google Trends…………………………………………………….35
4.2.3 意見分數…………………………………………………………………….36
4.2.4 預測模型…………………………………………………………………….41
4.2.4.1 Ford的銷售量預測結果……………………………………42
4.2.4.2 Toyota的銷售量預測結果………………………………43
4.2.4.3 預測模型效能評估結果……………………………………47
第五章 結論與未來研究方向…………………………………………………….48
參考文獻………………………………………………………………………………………………………49
附錄、中研院平衡語料庫詞類標記集……………………………………………56

[1] 仇士元，「加總模型的預測效果－以台灣地區汽車銷售量資料為例」，長庚大學企業管理系碩士論文，2003。
[2] 王怡心，成本與管理會計，三版，三民書局，台北，2012。
[3] 呂學慶，「台灣地區汽車市場銷售預測之探討」，國立中央大學管理學院高階主管企管碩士論文，2002。
[4] 李友錚，作業管理－創造競爭優勢，五版，前程文化事業，新北市，2014。
[5] 洪煙平，「台灣汽車銷售量與經濟因素關聯性之研究」，私立萬能科技大學經營管理研究所，碩士論文，2011。
[6] 范有寧等, “以數位口碑為基礎之流行性商品銷售預測模式,”資訊管理學報, Vol. 19, No. 1, pp. 27-50, 2012.
[7] 徐嘉鴻、鄭富烝、徐苑玲、邱南星、邱昭彰, “利用網路言論與關鍵字熱門度推測汽車銷售量,” 2013 International Conference on Advanced Information Technologies, 2013.
[8] 張甘青，「從使用者網路點閱習慣探討搜尋引擎最佳化」，東海大學圖書館館訊，第149期，39~54頁，2014。
[9] 黃俊英，企業研究方法，四版，東華書局，台北，2010。
[10] 鄧聚龍，灰色系統基本方法，華中理工大學出版社，武昌喻家山，1987。
[11] 鄧聚龍，郭洪，溫坤禮，張廷政，張偉哲，灰色預測模型方法與應用，高立圖書有限公司，台北，1999。
[12] 鄭光清，「汽車電子資訊產品對車輛銷售之研究--傳統迴歸與模糊迴歸之預測比較」，大葉大學工業工程與科技管理學系碩士論文，2005。
[13] 謝朋昇，「應用類神經網路於國產車銷售預測之研究」，大葉大學工業工程與科技管理學系碩士論文，2008。
[14] 魏夢麗，呂秀英，「決定係數(R2)在迴歸分析中的解釋及正確使用」，科學農業，第47卷，341~345頁，1999。
[15] 龔曙明，市場調查與預測，一版，清華大學出版社，北京，2005。
[16] Ahn, H. and Spangler, W. S., "Sales Prediction with Social Media Analysis, " 2014 SRII Global Conference, pp. 213-222, 2014.
[17] Arndt, J., "Role of product-related conversations in the diffusion of a new product, " Journal of Marketing Research, Vol. 4, No. 3, pp. 291-295, 1967.
[18] Bangwayo-Skeete P.F. and Skeete, R.W., "Can Google data improve the forecasting performance of tourist arrivals? Mixed-data sampling approach, " Tourism Management, Vol. 46, pp. 454-464, 2015.
[19] Bennett, K. and Campbell, C, "Support vector machines: hype or hallelujah?," ACM SIGKDD Explorations Newsletter, Vol. 2, pp. 1-13, 2000.
[20] Barreira, N., Godinho, P., and Melo, P., "Nowcasting unemployment rate and new car sales in south-western Europe with Google Trends, " NETNOM-ICS: Economic Research and Electronic Networking, Vol. 14, No. 3, pp. 129-165, 2013.
[21] Berger, J., Sorensen, A.T., Rasmussen, S.J., "Positive effects of negative publicity: when negative reviews increase sales, " Marketing Science Vol. 29, No. 5, pp. 815–827, 2010.
[22] Brabanter, K.D., Brabanter, J.D., Suykens, A.K., "Approximate Confidence and Prediction Intervals for Least Squares Support Vector Regression," IEEE Transactions on Neural Networks, Vol. 22, No. 1, pp.110–120, 2011.
[23] Chiu, C., Chiu, N.H., Sung, R.J., Hsieh, P.Y., "Opinion mining of hotel customer-generated contents in Chinese weblogs," Current Issues in Tourism, Vol. 18, No.5, pp. 477-495, 2013.
[24] Choi, H. and Varian, H., "Predicting the present with Google Trends, " Economic Record, Vol. 88, pp. 2-9, 2012.
[25] DeLurgio, S. A., Forecasting principles and applications, McGraw-Hill, Boston, 1998.
[26] Deng, J. L., "Control problems of grey systems, " Systems and Control Letters, Vol. 1, pp. 288-294, 1982.
[27] Fantazzini, D. and Toktamysova, Z., "Forecasting German Car Sales Using Google Data and Multivariate Models, " International Journal of Production Economics, Vol. 170, pp. 97-135, 2015.
[28] Gao, J., Wang, J., Wang, B. and Song, X., "Network flow prediction based on wavelet kernel-based least squares SVR algorithm," Journal of Computational Information Systems, Vol. 8, pp. 9011-9016, 2012.
[29] Ginsberg, J. , Mohebbi, M. H. , Patel, R. S. , Brammer, L. , Smolinski, M. S. & Brilliant, L., "Detecting Influenza Epidemics using Search Engine Query Data, " Nature, Vol. 457, No. 7232, pp. 1012-1014, 2009.
[30] Goldberg, D. E., "Genetic algorithms in search, Optimization and Machine Learning, " Addison-Wesley Publishing Company, USA, 1989.
[31] Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D., "Electronic word-of-mouth via consumer-opinion platforms: What motivates consumers to articulate themselves on the internet?, " Journal of Interactive Marketing, Vol. 18, No. 1, pp. 38-52, 2004.
[32] Hilton, R. W., Maher, M. W., and Selto, F. H., "Cost Management: Strategies for Business Decisions, " Fourth edition, McGraw-Hill, New York, 2008.
[33] Holland, J. H., "Adaptation in Natural and Artificial Systems," University of Michigan Press, Ann Arbor, Michigan, USA, 1975.
[34] Jiang, B.T., & Zhao, F.Y., "Particle swarm optimization-based least squares support vector regression for critical heat flux prediction, " Annals of Nuclear Energy Vol. 53, pp. 69-81, 2013.
[35] Jones, K.O., "Comparison of genetic algorithm and particle swarm optimization, " in: Proceedings of the 2005 International Conference on Computer System and Technologies, Varna, Bulgaria, pp. IIIA. 1-6, 2005.
[36] Keerthi, S.S. and Lin, C. J., "Asymptotic behaviors of support vector machines with Gaussian kernel," Neural Computation, Vol. 15, No. 7, pp. 1667-1689, 2003.
[37] Kennedy, J. and Eberhart, R.C., "Particle Swarm Optimization, " Proceedings of the IEEE International Conference on Neural Networks, Perth,Australia, pp.1942-1948, 1995.
[38] Kuo, R. J. and Xue, K. C., "A Decision Support System for Sales Forecasting through Fuzzy Neural Networks with Asymmetric Fuzzy Weights, " Decisions Support Systems, Vol. 24, No. 2, pp. 105-126, 1998.
[39] Kuo, R.J. and Xue, K.C., "Fuzzy neural networks with application to sales forecasting, " Fuzzy Sets and Systems, Vol. 108, No. 2, pp. 123-143, 1999.
[40] Liao, R.j., Zheng, H.B., Grzybowski, S., Yang, L.J., "Particle swarm optimization -least squares support vector regression based forecasting model on dissolved gases in oil-filled power transformers, " Electric Power Systems Research Vol. 81, No. 12, pp. 2074–2080, 2011.
[41] Lin, C.T., Chang, C.W., Chen, C.B., "The worst ill-conditioned silicon wafer machine detected by using grey relational analysis, " International Journal of Advanced Manufacturing Technology, Vol. 31, pp. 388-395, 2006.
[42] Lin, K., Lin, Q., Zhou, C., Yao, J., "Time Series Prediction based on Linear Regression and SVR, " Third International Conference on Natural Computation (ICNC 2007), 2007.
[43] Liu, R., Ren, H., "Housing Price Forecasting Based on Least Squares Support Vector Regression with Particle Swarm Optimization," Journal of Convergence Information Technology, Vol. 6, No. 11, pp. 325-333, 2011.
[44] Liu, S., Xu, L., Li, D., Li, Q., Jiang, Y., Tai, H., Zeng, L., "Prediction of dissolved oxygen content in river crab culture based on least squares support vector regression optimized by improved particle swarm optimization, " Computers and Electronics in Agriculture, Vol. 95, pp. 82-91, 2013.
[45] Lütkepohl, H. , Xu, F., "The role of the log transformation in forecasting economic variables," Empirical Economics, Vol. 42, No. 3, pp. 619–638, 2012.
[46] Luxhoj, J. T., Riis. Jens. O., and Stensballe, Brian. "A Hybrid Econometric-Neural Network Modeling Approach for Sales Forecasting, " The International Journal of Production Economics, 43, pp.175-192, 1996.
[47] Pai, P.F., Hung, K.C., and Lin, K.P., "Tourism demand forecasting using novel hybrid system," Expert Systems with Applications, Vol. 41, pp. 3691-3702, 2014.
[48] Salton, G. and McGill, M.J., "Introduction to modern information retrieval," McGraw-Hill, 1983.
[49] Sa-ngasoongsong, A., Bukkapatnam, S. T. S., Kim, J., Iyer, P.S. and Suresh, R. P., "Multi-step sales forecasting in automotive industry based on structural relationship identification, " International Journal of Production Economics, Vol. 140, pp. 875-887, 2012.
[50] Santos, J.D.A., and Barreto G.A., "Novel sparse LSSVR models in primal weight space for robust system identification with outliers," Journal of Process Control, Vol. 67, pp. 129-140, 2018.
[51] Shi, Y. and Eberhart, R.C., "A Modified Particle Swarm Optimizer, " Proceedings of the IEEE International Conference on Evolutionary Computation, Anchorage, Alaska, pp. 69-73, 1998.
[52] Smola, A.J. and Scholkopf, B., "A Tutorial on Support Vector Regression," Statistics and Computing, Vol. 14, pp. 199–222, 2004.
[53] Suykens, J.A.K. and Vandewalle J., "Least squares support vector machine classifiers," Neural Process. Letters, Vol. 9, No. 3, pp.293-300, 1999.
[54] Suykens, J.A.K., T. Van Gestel, J. De Brabanter, B. De Moor, and J. Vandewalle, "Least Squares Support Vector Machines," World Scientific, Singapore, 2002.
[55] Wang, F., K., Chang, K. K., Tzeng, C. W., "Using adaptive network-based fuzzy inference system to forecast automobile sales, " Expert Systems with Applications, Vol. 38, No. 8, pp. 10587-10593, 2011.
[56] Wu, L. and Brynjolfsson, E., "The Future of Prediction: How Google Searches Foreshadow Housing Prices and Sales, " Technical report, MIT., 2010.
[57] Yan, S., "A novel prediction method for stock index applying grey theory and neural networks," The 7th International Symposium on Operations Research and Its Applications (ISORA’08), pp. 104-111, 2008.
[58] Yang, Z., Gu, X., Liang, X., & Ling, L., "Genetic algorithm-least squares support vector regression based predicting and optimizing model on carbon fiber composite integrated conductivity, " Materials & Design, Vol. 31, pp. 1042-1049, 2010.
[59] Yuan, F.C., "Parameters optimization using genetic algorithms in support vector regression for sales volume forecasting," Applied Mathematics, Vol. 3, No.10, pp. 1480–1486, 2012.
[60] Yu, L., Xu, H. and Tang, L., "LSSVR ensemble learning with uncertain parameters for crude oil price forecasting," Applied Soft Computing, Vol. 56, pp. 692-701, 2017.
[61] http://www.businessdictionary.com/definition/sales-forecast.html
[62] http://ckipsvr.iis.sinica.edu.tw/ (中文斷詞系統CKIP)
[63] http://www.keenage.com/html/c_index.html (Hownet)
[64] http://www.itu.int/net/pressoffice/press_releases/2015/17.aspx#.V1QgIdJ961s (國際電信聯盟)
[65] https://www.statista.com/statistics/267161/market-share-of-search-engines-in-the-united-states/ (圖3資料來源)
[66] http://www.swarmintelligence.org/tutorials.php
[67] http://www.ttvma.org.tw/cht/industrial-survey.php (台灣區車輛工業同業公會)