Support vector machine (SVM) and twin SVM (TWSVM) are sensitive to the noisy classification, due to the unlimited measures in their losses, especially for imbalanced classification problem. In this paper, by combining the advantages of the correntropy induced loss function (C-Loss) and the hinge loss function (hinge loss), we introduce the rescaled hinge loss function (Rhinge loss), which is a monotonic, bounded, and nonconvex loss, into TWSVM for imbalanced noisy classification, called RTBSVM. We show that the Rhinge loss could approximate the hard margin loss and the hinge loss by adjusting the rescaled parameter, and further, our RTBSVM could improve the stability and performance of TWSVM and it is effective for imbalanced noisy classification. The experimental results show that our method performs better than the compared TWSVMs and robust SVMs on the imbalanced noisy classification.

Stock index price forecasting is a consistent focus of business intelligence. Various factors influence stock index price forecasting, such as technical indicators, financial news, business status, and the macroeconomics situation. In addition, many studies have shown that the exchange rate is related to the stock index price; however, no study has examined whether the exchange rate can be used to forecast stock index prices. Therefore, this paper focuses on this topic and uses exchange rate to predict China stock index price for the first time. Firstly, we compare the association of China stock index price with different data sources to illustrate the feasibility of using the exchange rate to predict stock index prices. Then, we generate some additional technical features of the exchange rate and propose a strategy to predict the stock index price. Finally, we compare the forecast results of China's stock index price based on four data sources, i.e., technical indicators, exchange rate data, US market index data and finance news data from January 3, 2017 to March 20, 2019. Experimental results demonstrate that the performance of exchange rate data for stock index prediction is comparable to other popular data sources and that, in some prediction periods, the exchange rate outperforms such data sources. The results confirm that the exchange rate could be used for forecasting the Shanghai Composite Index prices. INDEX TERMS Stock index prediction, exchange rate, finance news, technical indicators, SZ50, rolling window.

No abstract

Learning from crowds with robust support vector machines SCIENCE CHINA Information Sciences The new interpretation of support vector machines on statistical learning theory SCIENCE CHINA Mathematics 53, 151 (2010); A reliability assessment method based on support vector machines for CNC equipment Science in China Series E-Technological Sciences 52, 1849 (2009); Estimating biophysical parameters of rice with remote sensing data using support vector machines

No abstract