由於網路通訊技術的快速發展，人們在生活中愈來愈倚賴透過網路平台獲取所需的資訊。網路評論成為影響人們決策時的關鍵因素之一，但網路中的評論並非全部都值得人們信任。事實上，網路中存在約15%到30%的假評論，而這些假評論可能會誤導閱聽者的決策。在過往有很多關於假評論的研究，但鮮少將字詞之間的單調性關聯考慮在分類預測模型中。單調性的概念通常須由該領域的專家定義，但在文字評論資料中，很難找到專家能賦予龐雜的評論資料單調性的定義。
本研究使用統計檢定方法，找出文章中文字的單調性關聯。並透過修改神經網路中的損失函數，提出一個具有單調性先驗知識之深度模型架構。在模型預測結果分析的部分，運用檢測出的單調性關鍵字，對模型的分類結果進行解釋，提升模型的可解釋性。透過實驗的驗證，本研究提出的單調性深度模型在評論資料數量大的情況下比起一般的深度模型能有更佳的分類成效。

With the rapid development of machine learning and deep learning, which are widely used in our life. Though the models usually can achieve high accuracy, it is difficult to understand the prediction causality for the users. To improve the users’ confidence in the models, this research tries to fuse the prior knowledge into the deep model to make the model more interpretable. We use the explainable deep models for the fake reviews problems. There are about 15% to 30% fake reviews on the internet, which may mislead readers to make inappropriate decision. There are lots of previous research in the fake reviews detection region, but the monotonicity between words is rarely considered in the classification prediction model.
This study uses statistical methods to find out the monotonic association of the texts in the article. Furthermore, we modify the loss function in the neural network to make the deep model architecture with monotonic prior knowledge. This research takes some real-world datasets to verify the efficiency of our method. According to the experiments, our method performs better than original method when the number of reviews are larger. We also use monotonic words distribution in the confusion matrix to explain the reason of the model prediction results. By confusing the monotonic prior knowledge into the deep model, we improve the model performance and explain the model classification results.

Belharbi, S., Ben Ayed, I., McCaffrey, L., & Granger, E. (2019). Non-parametric Uni-modality Constraints for Deep Ordinal Classification. arXiv, arXiv: 1911.10720.

Bhargava, R., Baoni, A., & Sharma, Y. (2019). Composite sequential modeling for identifying fake reviews. Journal of Intelligent Systems, 28(3), 409-422.

Blitzer, J., Dredze, M., & Pereira, F. (2007). Biographies, bollywood, boom-boxes and blenders: Domain adaptation for sentiment classification. Paper presented at the Proceedings of the 45th annual meeting of the association of computational liguistics, Prague, Czech Republic

Bonchi, F., Giannotti, F., Mazzanti, A., & Pedreschi, D. (2003). ExAMiner: Optimized level-wise frequent pattern mining with monotone constraints. Paper presented at the Third IEEE International Conference on Data Mining, Melbourne, FL, USA.

Chawla, N. V., Japkowicz, N., & Kotcz, A. (2004). Special issue on learning from imbalanced data sets. ACM SIGKDD explorations newsletter, 6(1), 1-6.

Chen, C.-C., & Li, S.-T. (2014). Credit rating with a monotonicity-constrained support vector machine model. Expert Systems with Applications, 41(16), 7235-7247.

Cheng, X., Ren, X., Ma, S., & Li, S. (2015). A modeling method based on artificial neural network with monotonicity knowledge as constraints. Chemometrics and Intelligent Laboratory Systems, 145, 93-102.

Diligenti, M., Roychowdhury, S., & Gori, M. (2017). Integrating prior knowledge into deep learning. Paper presented at the 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA), Cancun, Mexico.

Ding, J. H., Li, X. C., Kang, X. J., & Gudivada, V. N. (2019). A Case Study of the Augmentation and Evaluation of Training Data for Deep Learning. Acm Journal of Data and Information Quality, 11(4), 22. doi:10.1145/3317573

Drummond, C., & Holte, R. C. (2003). C4. 5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling. Paper presented at the Workshop on learning from imbalanced datasets II, Washington, DC.

Filieri, R., Alguezaui, S., & McLeay, F. (2015). Why do travelers trust TripAdvisor? Antecedents of trust towards consumer-generated media and its influence on recommendation adoption and word of mouth. Tourism management, 51, 174-185.

Hoang, N.-D., Nguyen, Q.-L., & Tran, V.-D. (2018). Automatic recognition of asphalt pavement cracks using metaheuristic optimized edge detection algorithms and convolution neural network. Automation in Construction, 94, 203-213.

Jianqiang, Z., Xiaolin, G., & Xuejun, Z. (2018). Deep convolution neural networks for twitter sentiment analysis. IEEE Access, 6, 23253-23260.

Kotlowski, W., & Slowinski, R. (2012). On nonparametric ordinal classification with monotonicity constraints. IEEE Transactions on Knowledge and Data Engineering, 25(11), 2576-2589.

Kumar, N., Venugopal, D., Qiu, L., & Kumar, S. (2018). Detecting review manipulation on online platforms with hierarchical supervised learning. Journal of Management Information Systems, 35(1), 350-380.

Kursuncu, U., Gaur, M., & Sheth, A. (2019). Knowledge infused learning (K-IL): Towards deep incorporation of knowledge in deep learning. arXiv preprint arXiv:1912.00512.

Lappas, T., Sabnis, G., & Valkanas, G. (2016). The impact of fake reviews on online visibility: A vulnerability assessment of the hotel industry. Information Systems Research, 27(4), 940-961.

LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521, 436-444. doi:10.1038/nature14539

Luca, M., & Zervas, G. (2016). Fake it till you make it: Reputation, competition, and Yelp review fraud. Management Science, 62(12), 3412-3427.

Martens, D., & Maalej, W. (2019). Towards understanding and detecting fake reviews in app stores. Empirical Software Engineering, 24, 3316-3355.

Mayzlin, D., Dover, Y., & Chevalier, J. (2014). Promotional reviews: An empirical investigation of online review manipulation. American Economic Review, 104(8), 2421-2455.

Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781.

Mukherjee, A., Venkataraman, V., Liu, B., & Glance, N. S. (2013). What yelp fake review filter might be doing? Paper presented at the Seventh international AAAI conference on weblogs and social media, Palo Alto, California.

Muralidhar, N., Islam, M. R., Marwah, M., Karpatne, A., & Ramakrishnan, N. (2018). Incorporating prior domain knowledge into deep neural networks. Paper presented at the 2018 IEEE International Conference on Big Data (Big Data), Seattle, WA, USA.

Nandwani, Y., Pathak, A., & Singla, P. (2019). A Primal Dual Formulation For Deep Learning With Constraints. Paper presented at the Advances in Neural Information Processing Systems, Vancouver, Canada.

Ott, M., Choi, Y., Cardie, C., & Hancock, J. T. (2011). Finding deceptive opinion spam by any stretch of the imagination. arXiv preprint arXiv:1107.4557.

Pang, B., & Lee, L. (2004). A sentimental education: Sentiment analysis using subjectivity summarization based on minimum cuts. Paper presented at the Proceedings of the 42nd annual meeting on Association for Computational Linguistics, Barcelona, Spain.

Rademaker, M., & Baets, B. D. (2014). A ranking procedure based on a natural monotonicity constraint. Information Fusion, 17, 74-82.

Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). " Why should I trust you?" Explaining the predictions of any classifier. Paper presented at the Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, San Francisco, California, USA.

Roscher, R., Bohn, B., Duarte, M. F., & Garcke, J. (2020). Explainable machine learning for scientific insights and discoveries. IEEE Access, 8, 42200-42216.

Vo, K., Pham, D., Nguyen, M., Mai, T., & Quan, T. (2017). Combination of domain knowledge and deep learning for sentiment analysis. Paper presented at the International Workshop on Multi-disciplinary Trends in Artificial Intelligence, Gadong, Brunei.

Rueden, L. V., Mayer, S., Beckh, K., Georgiev, B., Giesselbach, S., Heese, R., . . . Ramamurthy, R. (2019). Informed Machine Learning--A Taxonomy and Survey of Integrating Knowledge into Learning Systems. arXiv preprint arXiv:1903.12394.

Taboada, M., Anthony, C., & Voll, K. D. (2006). Methods for Creating Semantic Orientation Dictionaries. Paper presented at the LREC.

Wu, Y., Ngai, E. W. T., Wu, P., & Wu, C. (2020). Fake online reviews: Literature review, synthesis, and directions for future research. Decision Support Systems, 113280.

Xu, J., Zhang, Z., Friedman, T., Liang, Y., & Broeck, G. (2018). A semantic loss function for deep learning with symbolic knowledge. Paper presented at the International Conference on Machine Learning, Stockholm, Sweden.

Yang, C.-L., Yang, C.-Y., Chen, Z.-X., & Lo, N.-W. (2019). Multivariate time series data transformation for convolutional neural network. Paper presented at the 2019 IEEE/SICE International Symposium on System Integration (SII), Paris, France.

Zhang, D., Zhou, L., Kehoe, J. L., & Kilic, I. Y. (2016). What online reviewer behaviors really matter? Effects of verbal and nonverbal behaviors on detection of fake online reviews. Journal of Management Information Systems, 33(2), 456-481.

Zhang, H., & Zhang, Z. (1999). Feedforward networks with monotone constraints. Paper presented at the IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No. 99CH36339), Washington, DC, USA.