Stock Transaction Analysis System based on Hadoop and Capital flow

Stock transaction analysis is an interesting topic to be investigated both for financial economist and data processing expert. In practice, various analysis systems have been studied to gain excellent result for prediction. However, few jobs have been done on the combination of financial index analysis and data processing techniques. The purpose of this paper is to test if the stock transaction behavior can be detected by using data processing technique on a synthetic index of capital flow. In this paper, the overall logic architecture and functional framework of analysis system have been introduced detailedly. In addition, a developed prototype system has been tested and the way how the main components are implemented is also illustrated. The experiments show that the prototypes system is capable of achieving better trade-off regarding the accuracy, availability and flexibility.