论文信息 - Evaluating forecasting algorithm of realistic datasets based on machine learning

Evaluating forecasting algorithm of realistic datasets based on machine learning

Developing a good forecasting model is one of the most important methods to boost profits of enterprises. However, it is also a difficult task to design a well-performed model, especially when using traditional linear statistic tools. There are many useful algorithms in Machine Learning and each has its advantage. This paper begins with two realistic forecasting issues: store sales forecasting and customer satisfaction analyzing, gives a preprocessing scheme to improve the performance of machine leaning algorithms, including Artificial Neural Network (ANN), Support Vector Machine (SVM) and eXtreme Gradient Boosting (XGBoost). We develop schemes respectively for each of the two datasets and it is verified to solve each of the problems. The proposed scheme shows better performance and achieve top 10% rank in the international Machine Learning competition (Kaggle).

[1] Radford M. Neal. Pattern Recognition and Machine Learning , 2007, Technometrics.

[2] Tianqi Chen,et al. XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[3] R. Tibshirani,et al. An introduction to the bootstrap , 1993 .

[4] Haibo He,et al. Learning from Imbalanced Data , 2009, IEEE Transactions on Knowledge and Data Engineering.

[5] Nello Cristianini,et al. An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[6] Prabhat,et al. Artificial Neural Network , 2018, Encyclopedia of GIS.

[7] J. Friedman. Greedy function approximation: A gradient boosting machine. , 2001 .

[8] Tom Fawcett,et al. An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

[9] Nasser M. Nasrabadi,et al. Pattern Recognition and Machine Learning , 2006, Technometrics.

[10] Johan A. K. Suykens,et al. Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[11] Ethem Alpaydin,et al. Introduction to machine learning , 2004, Adaptive computation and machine learning.

[12] G. Box,et al. Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series Models , 1970 .

[13] Geoffrey E. Hinton,et al. Deep Learning , 2015, Nature.

[14] Yong Yu,et al. Sales forecasting using extreme learning machine with applications in fashion retailing , 2008, Decis. Support Syst..