成功的銷售預測是公司成功的關鍵因素，它使管理人員能夠預測未來並據以訂定相應計劃。汽車工業對許多國家的經濟發展非常重要，如果汽車公司能夠開發出有效預測汽車銷售的方法，就可以安排準確的生產計劃和銷售計劃以提供良好的客戶服務並進而獲得利潤。
傳統的銷售預測研究以使用歷史銷售數據為主，但隨著資訊科技快速發展，消費者開始在網路分享他們對產品正面或負面的意見（網路情緒資料），潛在買家則習慣在購買前參考這些網路評論，因此網路情緒資料己經成為預測的重要因素。此外，深度學習演算法已廣泛且有效運用於預測，所以本研究運用網路情緒資料和CNN-LSTM模型以提升台灣汽車銷售走向的預測績效。
首先，本研究收集台灣前六大進口汽車品牌的歷史銷量和多管道的網路情緒資料，並進行預處理，以標記汽車銷售走向。然後，本研究建構傳統（MLR-Classical）、情感（MLR-Sentimental）和CNN-LSTM等三種預測模型，其中，傳統模型採用歷史銷量和季節性數據及MLR方法進行預測；情感模型採用歷史銷量、季節性數據和網路情緒資料及MLR方法進行預測；CNN-LSTM模型則採用季節性數據和網路情緒資料及CNN-LSTM方法進行預測。最後，本研究對此三種預測模型的各項績效指標進行分析比較，以驗證網路情緒資料和CNN-LSTM方法對汽車銷售走向預測績效的影響。
四項預測績效指標顯示，準確率（Accuracy）提高27.78%（從41.67% 提高到69.45%），精確率（Precision）提高0.39（從0.38提高到0.77），召回率（Recall）提高0.27（從0.42提高到0.69），F1-score提高0.33（從0.35到0.68）。因此，本研究運用的方法確實提高台灣汽車銷售走向的整體預測績效。

A successful sales forecast is a key factor in a company's success. It enables managers to anticipate the future and make plans accordingly. The automotive industry is very important to the economic development of many countries. If automobile companies can develop methods to effectively forecast car sales, they can schedule accurate production and sales plans in order to provide good customer service and thereby make a profit.
Traditionally, sales forecasting studies have been based on the application of historical sales data, but with the rapid development of information technology, modern consumers have started to share their positive or negative opinions about products online (online sentiment data), and potential buyers are accustomed to referring to these online reviews before purchasing, so online sentiment data has become an important factor in forecasting.
In addition, since deep learning algorithms for forecasting have been widely used with good results, this study uses online sentiment data and CNN-LSTM model to improve the prediction performance of car sales movement direction in Taiwan. Furthermore, due to the high market share of the six major imported car brands (BMW, LEXUS, MAZDA, Mercedes-Benz, TOYOTA, and Volkswagen) in Taiwan, this research adopts the six major imported car brands as the experimental case.
First, this study collects historical sales volume and multi-channel online sentiment data of six major imported car brands in Taiwan and pre-processes them to label the car sales movement direction. Then, three forecasting models, MLR-Classical, MLR-Sentimental, and CNN-LSTM, are constructed in this research. The MLR-Classical model uses historical sales volume, seasonal data, and MLR method for forecasting; the MLR-Sentimental model uses historical sales volume, seasonal data, online sentiment data, and MLR method for forecasting; the CNN-LSTM model uses seasonal data, online sentiment data, and CNN-LSTM method for forecasting. Finally, this research compares the performance indicators of these three forecasting models to verify the impact of the online sentiment data and CNN-LSTM model on the forecasting performance of car sales movement direction.
The results showed that four forecasting performance measures, i.e., accuracy improved 27.78% (from 41.67% to 69.45%), precision improved 0.39 (from 0.38 to 0.77), recall improved 0.27 (from 0.42 to 0.69), and F1-score improved 0.33 (from 0.35 to 0.68). This demonstrates that the forecasting performance of car sales movement direction in Taiwan can actually be improved with the online sentiment data and CNN-LSTM model.

Al-Saffar, A. A. M., Tao, H., & Talab, M. A. (2017). Review of deep convolution neural network in image classification. 2017 International Conference on Radar, Antenna, Microwave, Electronics, and Telecommunications (ICRAMET), Jakarta Selatan, Indonesia.
Alagu Dharshini, M. P., & Antelin Vijila, S. (2021). Survey of Machine Learning and Deep Learning Approaches on Sales Forecasting. 2021 4th International Conference on Computing and Communications Technologies (ICCCT), Chennal, Tamil Nadu, India.
Alkahtani, H., Aldhyani, T. H. H., & Meraoumia, A. (2021). Botnet Attack Detection by Using CNN-LSTM Model for Internet of Things Applications. Security and Communication Networks, 2021, 1-23. https://doi.org/10.1155/2021/3806459
Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., . . . Farhan, L. (2021). Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. J Big Data, 8(1), 53. https://doi.org/10.1186/s40537-021-00444-8
Ankita, Rani, S., Bashir, A. K., Alhudhaif, A., Koundal, D., & Gunduz, E. S. (2022). An efficient CNN-LSTM model for sentiment detection in #BlackLivesMatter. Expert Systems with Applications, 193. https://doi.org/10.1016/j.eswa.2021.116256
Armstrong, J. S., & Brodie, R. J. (2001). Forecasting for Marketing. Mark Health Serv, 21(3), 16-22. https://www.ncbi.nlm.nih.gov/pubmed/11525136
Aroef, C., Yuda, R., Rustam, Z., & Pandelaki, J. (2019). Multinomial Logistic Regression and Support Vector Machine for Osteoarthritis Classification. Journal of Physics: Conference Series.
Bampoula, X., Siaterlis, G., Nikolakis, N., & Alexopoulos, K. (2021). A Deep Learning Model for Predictive Maintenance in Cyber-Physical Production Systems Using LSTM Autoencoders. Sensors (Basel), 21(3). https://doi.org/10.3390/s21030972
Bataineh, A. Q. (2015). The Impact of Perceived e-WOM on Purchase Intention: The Mediating Role of Corporate Image. International Journal of marketing studies, 7(1). https://doi.org/10.5539/ijms.v7n1p126
Bayaga, A. (2010). Multinomial Logistic Regression: Usage and Application in Risk Analysis. Journal of applied quantitative methods, 5(2).
Biau, G., & Scornet, E. (2016). A random forest guided tour. Test, 25(2), 197-227. https://doi.org/10.1007/s11749-016-0481-7
Brühl, B., Hülsmann, M., Borscheid, D., Friedrich, C. M., & Reith, D. (2009). A sales forecast model for the German automobile market based on time series analysis and data mining methods. Industrial Conference on Data Mining, Leipzig, Germany.
Cai, Z., & Vasconcelos, N. (2018). Cascade r-cnn: Delving into high quality object detection. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA.
Cao, J., & Wang, J. (2019). Stock price forecasting model based on modified convolution neural network and financial time series analysis. International Journal of Communication Systems, 32(12). https://doi.org/10.1002/dac.3987
Chaudharya, M., Gastlia, M. S., Nassara, L., & Karraya, F. (2021). Deep Learning Approaches for Forecasting Strawberry Yields and Prices Using Satellite Images and Station-Based Soil Parameters. AAAI 2021 Spring Symposium on Combining Machine Learning and Knowledge Engineering (AAAI-MAKE 2021), Palo Alto, California, USA,
Chong, A. Y. L., Li, B. Y., Ngai, E. W. T., Ch'ng, E., & Lee, F. (2016). Predicting online product sales via online reviews, sentiments, and promotion strategies A big data architecture and neural network approach. International Journal of Operations & Production Management, 36(4), 358-385. https://doi.org/10.1108/Ijopm-03-2015-0151
El-Habil, A. M. (2012). An application on multinomial logistic regression model. Pakistan journal of statistics and operation research, 271-291.
Fan, Z.-P., Che, Y.-J., & Chen, Z.-Y. (2017). Product sales forecasting using online reviews and historical sales data: A method combining the Bass model and sentiment analysis. Journal of Business Research, 74, 90-100. https://doi.org/10.1016/j.jbusres.2017.01.010
Fantazzini, D., & Toktamysova, Z. (2015). Forecasting German car sales using Google data and multivariate models. International Journal of Production Economics, 170, 97-135. https://doi.org/10.1016/j.ijpe.2015.09.010
Feldman, R. (2013). Techniques and applications for sentiment analysis. Communications of the Acm, 56(4), 82-89. https://doi.org/10.1145/2436256.2436274
Filieri, R., Lin, Z., Pino, G., Alguezaui, S., & Inversini, A. (2021). The role of visual cues in eWOM on consumers’ behavioral intention and decisions. Journal of Business Research, 135, 663-675. https://doi.org/10.1016/j.jbusres.2021.06.055
Frid-Adar, M., Diamant, I., Klang, E., Amitai, M., Goldberger, J., & Greenspan, H. (2018). GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification. Neurocomputing, 321, 321-331. https://doi.org/10.1016/j.neucom.2018.09.013
Fu, L., Liang, P., Li, X., & Yang, C. (2021). A Machine Learning Based Ensemble Method for Automatic Multiclass Classification of Decisions. 2021 Evaluation and Assessment in Software Engineering (EASE), Trondheim, Norway. https://doi.org/10.1145/3463274.3463325
G, S., KP, S., & R, V. (2018). Automated detection of diabetes using CNN and CNN-LSTM network and heart rate signals. Procedia Computer Science, 132, 1253-1262. https://doi.org/10.1016/j.procs.2018.05.041
Hülsmann, M., Borscheid, D., & Friedrich, C. M. (2012). General sales forecast models for automobile markets and their analysis. Transactions on Machine Learning and Data Mining, 5 (2012)(2). https://opus.bsz-bw.de/fhdo/frontdoor/index/index/docId/414
Hai, P. N., Tien, N. M., Hieu, H. T., Chung, P. Q., Son, N. T., Ha, P. N., & Son, N. T. (2020). An Empirical Research on the Effectiveness of Different LSTM Architectures on Vietnamese Stock Market. 2020 International Conference on Control, Robotics and Intelligent System, Xiamen, China. https://doi.org/10.1145/3437802.3437827
Hand, D., & Christen, P. (2018). A note on using the F-measure for evaluating record linkage algorithms. Statistics and Computing, 28(3), 539-547. https://doi.org/10.1007/s11222-017-9746-6
Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronic word-of-mouth via consumer-opinion platforms: What motivates consumers to articulate themselves on the Internet? Journal of Interactive Marketing, 18(1), 38-52. https://doi.org/10.1002/dir.10073
Hershey, S., Chaudhuri, S., Ellis, D. P., Gemmeke, J. F., Jansen, A., Moore, R. C., . . . Seybold, B. (2017). CNN architectures for large-scale audio classification. 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, USA.
Hosmer Jr, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression (Vol. 398). John Wiley & Sons.
Hu, N., Koh, N. S., & Reddy, S. K. (2014). Ratings lead you to the product, reviews help you clinch it? The mediating role of online review sentiments on product sales. Decision Support Systems, 57, 42-53. https://doi.org/10.1016/j.dss.2013.07.009
Huang, T., Zhang, Q., Tang, X., Zhao, S., & Lu, X. (2021). A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems. Artificial Intelligence Review, 55(2), 1289-1315. https://doi.org/10.1007/s10462-021-09993-z
Huang, W., Nakamori, Y., & Wang, S.-Y. (2005). Forecasting stock market movement direction with support vector machine. Computers & Operations Research, 32(10), 2513-2522. https://doi.org/10.1016/j.cor.2004.03.016
Karevan, Z., & Suykens, J. A. K. (2020). Transductive LSTM for time-series prediction: An application to weather forecasting. Neural Netw, 125, 1-9. https://doi.org/10.1016/j.neunet.2019.12.030
Kaur, K., & Singh, T. (2021). Impact of Online Consumer Reviews on Amazon Books Sales: Empirical Evidence from India. Journal of theoretical and applied electronic commerce research, 16(7), 2793-2807. https://doi.org/10.3390/jtaer16070153
Kiranyaz, S., Avci, O., Abdeljaber, O., Ince, T., Gabbouj, M., & Inman, D. J. (2021). 1D Convolutional Neural Networks and Applications: A Survey. Mechanical Systems and Signal Processing, 151. https://doi.org/10.1016/j.ymssp.2020.107398
Kiranyaz, S., Ince, T., Hamila, R., & Gabbouj, M. (2015). Convolutional Neural Networks for patient-specific ECG classification. 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Milan, Italy.
Kudeshia, C., & Kumar, A. (2017). Social eWOM: does it affect the brand attitude and purchase intention of brands? Management Research Review, 40(3), 310-330. https://doi.org/10.1108/MRR-07-2015-0161
López, M., & Sicilia, M. (2014). Determinants of E-WOM Influence: The Role of Consumers' Internet Experience. Journal of theoretical and applied electronic commerce research, 9(1), 7-8. https://doi.org/10.4067/s0718-18762014000100004
Lee, I. (2017). A study of the effect of social shopping deals on online reviews. Industrial Management & Data Systems, 117(10), 2227-2240. https://doi.org/10.1108/imds-09-2016-0378
Liu, G., & Long, B. (2009). The research on Combination forecasting model of the automobile sales forecasting system. 2009 International Forum on Computer Science-Technology and Applications (IFCSTA), ChongQing, China.
Liu, N., Ren, S., Choi, T.-M., Hui, C.-L., & Ng, S.-F. (2013). Sales Forecasting for Fashion Retailing Service Industry: A Review. Mathematical Problems in Engineering, 2013, 1-9. https://doi.org/10.1155/2013/738675
Livieris, I. E. (2020). An advanced active set L-BFGS algorithm for training weight-constrained neural networks. Neural Computing and Applications, 32(11), 6669-6684. https://doi.org/10.1007/s00521-019-04689-6
Livieris, I. E., Pintelas, E., & Pintelas, P. (2020). A CNN–LSTM model for gold price time-series forecasting. Neural Computing and Applications, 32(23), 17351-17360. https://doi.org/10.1007/s00521-020-04867-x
Loureiro, A. L. D., Miguéis, V. L., & Da Silva, L. F. M. (2018). Exploring the use of deep neural networks for sales forecasting in fashion retail. Decision Support Systems, 114, 81-93. https://doi.org/10.1016/j.dss.2018.08.010
Lu, H.-K., Yang, L.-W., Lin, P.-C., Yang, T.-H., & Chen, A. N. (2017). A study on adoption of bitcoin in Taiwan –Using Big Data Analysis of Social Media. 2017 the 3rd International Conference on Communication and Information Processing (ICCIP), Tokyo, Japan.
Lu, W., Li, J., Li, Y., Sun, A., Wang, J., & Hassanien, A. E. I. B. (2020). A CNN-LSTM-Based Model to Forecast Stock Prices. Complexity, 2020, 1-10. https://doi.org/10.1155/2020/6622927
Lu, X., Ba, S., Huang, L., & Feng, Y. (2013). Promotional Marketing or Word-of-Mouth? Evidence from Online Restaurant Reviews. Information Systems Research, 24(3), 596-612. https://doi.org/10.1287/isre.1120.0454
Ma, Y., Han, R., & Wang, W. (2020). Prediction-Based Portfolio Optimization Models Using Deep Neural Networks. IEEE Access, 8, 115393-115405. https://doi.org/10.1109/access.2020.3003819
Manowska, A. (2020). Using the LSTM Network to Forecast the Demand for Electricity in Poland. Applied Sciences, 10(23), 8455. https://doi.org/10.3390/app10238455
Moon, S.-H., & Kim, Y.-H. (2020). An improved forecast of precipitation type using correlation-based feature selection and multinomial logistic regression. Atmospheric Research, 240, 104928. https://doi.org/10.1016/j.atmosres.2020.104928
MOTC. (2020). Commonly Used Transportation Statistics. Ministry of Transportation and Communications R.O.C. Retrieved 2020/9/28 from https://stat.motc.gov.tw/mocdb/stmain.jsp?sys=100&funid=defjsp#
Mudinas, A., Zhang, D., & Levene, M. (2019). Market Trend Prediction using Sentiment Analysis: Lessons Learned and Paths Forward. arXiv preprint arXiv:1903.05440. https://doi.org/10.1145/nnnnnnn.nnnnnnn
Nguyen, H. D., Tran, K. P., Thomassey, S., & Hamad, M. (2021). Forecasting and Anomaly Detection approaches using LSTM and LSTM Autoencoder techniques with the applications in supply chain management. International Journal of Information Management, 57. https://doi.org/10.1016/j.ijinfomgt.2020.102282
OpView Insight. (2020). OpView Insight: Social Media Monitoring Tool. Retrieved 2020/8/15 from https://www.opview.com.tw/
Ouyang, Q., Lv, Y., Ma, J., & Li, J. (2020). An LSTM-Based Method Considering History and Real-Time Data for Passenger Flow Prediction. Applied Sciences, 10(11). https://doi.org/10.3390/app10113788
Pierdzioch, C., Rülke, J.-C., & Stadtmann, G. (2011). Forecasting U.S. car sales and car registrations in Japan: Rationality, accuracy and herding. Japan and the World Economy, 23(4), 253-258. https://doi.org/10.1016/j.japwor.2011.09.002
Poh, H. L., Yao, J., & Jašic, T. (1998). Neural networks for the analysis and forecasting of advertising and promotion impact. Intelligent Systems in Accounting, Finance & Management, 7(4), 253-268.
Pranckevičius, T., & Marcinkevičius, V. (2017). Comparison of Naive Bayes, Random Forest, Decision Tree, Support Vector Machines, and Logistic Regression Classifiers for Text Reviews Classification. Baltic Journal of Modern Computing, 5(2). https://doi.org/10.22364/bjmc.2017.5.2.05
Qiu, J., Wang, B., & Zhou, C. (2020). Forecasting stock prices with long-short term memory neural network based on attention mechanism. PLoS ONE, 15(1), e0227222. https://doi.org/10.1371/journal.pone.0227222
Rambocas, M., & Pacheco, B. G. (2018). Online sentiment analysis in marketing research: a review. Journal of Research in Interactive Marketing, 12(2), 146-163. https://doi.org/10.1108/jrim-05-2017-0030
Ren, R., Wu, D. D., & Liu, T. (2019). Forecasting Stock Market Movement Direction Using Sentiment Analysis and Support Vector Machine. IEEE Systems Journal, 13(1), 760-770. https://doi.org/10.1109/jsyst.2018.2794462
Sa’ait, N., Kanyan, A., & Nazrin, M. F. (2016). The effect of e-WOM on customer purchase intention. International Academic Research Journal of Social Science, 2(1), 73-80.
Sezer, O. B., & Ozbayoglu, A. M. (2018). Algorithmic financial trading with deep convolutional neural networks: Time series to image conversion approach. Applied Soft Computing, 70, 525-538. https://doi.org/10.1016/j.asoc.2018.04.024
Shin, H. C., Roth, H. R., Gao, M., Lu, L., Xu, Z., Nogues, I., . . . Summers, R. M. (2016). Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning. IEEE Trans Med Imaging, 35(5), 1285-1298. https://doi.org/10.1109/TMI.2016.2528162
Singh, A., Jenamani, M., Thakkar, J. J., & Rana, N. P. (2022). Quantifying the effect of eWOM embedded consumer perceptions on sales: An integrated aspect-level sentiment analysis and panel data modeling approach. Journal of Business Research, 138, 52-64. https://doi.org/10.1016/j.jbusres.2021.08.060
Smagulova, K., & James, A. P. (2019). A survey on LSTM memristive neural network architectures and applications. The European Physical Journal Special Topics, 228(10), 2313-2324. https://doi.org/10.1140/epjst/e2019-900046-x
Stevenson, W. J. (2015). Operations Management, 12th ed. Irwin/McGraw-Hill.
Tesla. (2022). Tesla orders had surged significantly following Super Bowl 2022 — the result of other automakers advertising their EVs. Retrieved 2022/4/24 from https://www.teslaoracle.com/2022/04/23/tesla-orders-surged-significantly-after-super-bowl-2022-the-result-of-other-automakers-advertising-their-evs/
Thiesing, F. M., & Vornberger, O. (1997). Sales forecasting using neural networks. Proceedings of International Conference on Neural Networks (ICNN'97), Houston, Texas.
Upadhyay, A., Bandyopadhyay, G., & Dutta, A. (2012). Forecasting stock performance in indian market using multinomial logistic regression. Journal of Business Studies Quarterly, 3(3), 16.
Wang, F.-K., Chang, K.-K., & Tzeng, C.-W. (2011). Using adaptive network-based fuzzy inference system to forecast automobile sales. Expert Systems with Applications, 38(8), 10587-10593. https://doi.org/10.1016/j.eswa.2011.02.100
Wang, P.-S., Liu, Y., Guo, Y.-X., Sun, C.-Y., & Tong, X. (2017). O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis. ACM Transactions on Graphics, 36(4), 1-11. https://doi.org/10.1145/3072959.3073608
Wang, Y. (2005). A multinomial logistic regression modeling approach for anomaly intrusion detection. Computers & Security, 24(8), 662-674. https://doi.org/10.1016/j.cose.2005.05.003
Wu, S.-J., Chiang, R.-D., & Chang, H.-C. (2018). Applying sentiment analysis in social web for smart decision support marketing. Journal of Ambient Intelligence and Humanized Computing, 1-10. https://doi.org/10.1007/s12652-018-0683-9
Xu, X., & Yoneda, M. (2021). Multitask Air-Quality Prediction Based on LSTM-Autoencoder Model. IEEE Trans Cybern, 51(5), 2577-2586. https://doi.org/10.1109/TCYB.2019.2945999
Yenter, A., & Verma, A. (2017). Deep CNN-LSTM with combined kernels from multiple branches for IMDb review sentiment analysis. 2017 IEEE 8th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference (UEMCON), New York City, NY, USA.
Yu, Q., Wang, K., Strandhagen, J. O., & Wang, Y. (2018). Application of Long Short-Term Memory Neural Network to Sales Forecasting in Retail—A Case Study. In Lecture Notes in Electrical Engineering (pp. 11-17). Springer Singapore. https://doi.org/10.1007/978-981-10-5768-7_2
Zhou, J., YAN, J.-f., YANG, L., WANG, M., & XIA, P. (2019). Customer Churn Prediction Model Based on LSTM and CNN in Music Streaming. DEStech Transactions on Engineering and Technology Research(aemce), 254-261.