Context-Aware Systems and Applications, and Nature of Computation and Communication. 6th International Conference, ICCASA 2017, and 3rd International Conference, ICTCC 2017, Tam Ky, Vietnam, November 23-24, 2017, Proceedings

Research Article

Hybrid Classifier by Integrating Sentiment and Technical Indicator Classifiers

  • @INPROCEEDINGS{10.1007/978-3-319-77818-1_3,
        author={Nguyen Van and Nguyen Doanh and Nguyen Khanh and Nguyen Anh},
        title={Hybrid Classifier by Integrating Sentiment and Technical Indicator Classifiers},
        proceedings={Context-Aware Systems and Applications, and Nature of Computation and Communication. 6th International Conference, ICCASA 2017, and 3rd International Conference, ICTCC 2017, Tam Ky, Vietnam, November 23-24, 2017, Proceedings},
        proceedings_a={ICCASA \& ICTCC},
        year={2018},
        month={3},
        keywords={Machine learning Stock market Classifier Sentiment analysis Hybrid classifier Technical indicator},
        doi={10.1007/978-3-319-77818-1_3}
    }
    
  • Nguyen Van
    Nguyen Doanh
    Nguyen Khanh
    Nguyen Anh
    Year: 2018
    Hybrid Classifier by Integrating Sentiment and Technical Indicator Classifiers
    ICCASA & ICTCC
    Springer
    DOI: 10.1007/978-3-319-77818-1_3
Nguyen Van1,*, Nguyen Doanh, Nguyen Khanh, Nguyen Anh,*
  • 1: Hanoi University of Science and Technology
*Contact email: vanndkstnk57@gmail.com, anh.nguyenthingoc@hust.edu.vn

Abstract

Classifiers in stock market are an interesting and challenging research topic in machine learning. A large research has been conducted for classifying in stock market by using different approaches in machine learning. This research paper presents a detail study on integrating sentiment classifier and technical indicator classifier. The research subject is investigated to classify a stock into one of three labels being top, neutral or bottom. First, using technical indicators such as relative strength index (RSI), money flow index (MFI) and relative volatility index (RVI) to classify stock, then using bagging of learning machine to classify the stock. Second, using sentiment data to classify the stock. Third, integrating technical indicator and sentiment classifiers to build hybrid classifier. In this study, hybrid machine learning by combining sentiment and technical indicator classifiers is proposed. We applied this proposal hybrid classifier for five stocks in VN30. The empirical results show hybrid classifier stock has more power than single technical indicator classifier or sentiment classifier.