Vo, NNY, Xu, G, Liu, S, Brownlow, EJ, Culbert, B & Chu, C 2018, 'Client Churn Prediction with Call Log Analysis', Database Systems for Advanced Applications, International Conference on Database Systems for Advanced Applications, Springer, Gold Coast, Australia, pp. 752-763.View/Download from: UTS OPUS or Publisher's site
Vo, NNY, Liu, S, He, X & Xu, G 2018, 'Multimodal Mixture Density Boosting Network for Personality Mining', Advances in Knowledge Discovery and Data Mining (LNCS), Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, Melbourne, Australia, pp. 644-655.View/Download from: UTS OPUS or Publisher's site
Knowing people's personalities is useful in various real-world applications, such as personnel selection. Traditionally, we have to rely on qualitative methodologies, e.g. surveys or psychology tests to determine a person's traits. However, recent advances in machine learning have it possible to automate this process by inferring personalities from textual data. Despite of its success, text-based method ignores the facial expression and the way people speak, which can also carry important information about human characteristics. In this work, a personality mining framework is proposed to exploit all the information from videos, including visual, auditory, and textual perspectives. Using a state-of-art cascade network built on advanced gradient boosting algorithms, the result produced by our proposed methodology can achieve lower the prediction errors than most current machine learning algorithms. Our multimodal mixture density boosting network especially perform well with small sample size datasets, which is useful for learning problems in psychology fields where big data is often not available.
Vo, NNY & Xu, G 2017, 'The volatility of Bitcoin returns and its correlation to financial markets', Proceedings of 4th International Conference on Behavioral, Economic, and Socio-Cultural Computing (BESC 2017), Krakow, Poland, 16-18 October 2017, International Conference on Behavioral, Economic, Socio-cultural Computing, IEEE, Krakow, Poland.View/Download from: UTS OPUS or Publisher's site
The 2008 financial crisis had scattered incredulity around the globe regarding traditional financial systems, which made investors and non-financial customers turn to other alternative such as digital banking systems. The existence and development of blockchain technology make cryptocurrency in recent years believably become a complete alternative to traditional ones. Bitcoin is the world's first peer-to-peer and decentralized digital cash system initiated by Nakamoto . Though being the most prominent cryptocurrency, Bitcoin has not been a legal trading currency in various countries. Its exchange rate has appeared to be an exceptionally high-risk portfolio with extreme volatility, which requires a more detailed evaluation before making any decision. This paper utilizes knowledge of statistics for financial time series and machine learning to (i) fit the parametric distribution and (ii) model and forecast the volatility of Bitcoin returns, and (iii) analyze its correlation to other financial market indicators. The fitted parametric time series model significantly outperforms other standard models in explaining the stylized facts and statistical variances in the behavior of Bitcoin returns. The model forecast also outperforms some machine learning methodologies, which would benefit policy makers, banks and financial investors in trading activities for both long-term and short-term strategies.