Sentiment analysis for financial news headlines using machine learning algorithm
Abstract
The study covers the implementation of machine learning algorithm approaches in sentiment analysis of Malaysia financial news headlines. This study can be used for stakeholders who want to know about the financial news and seek knowledge or data in the financial world. The data are gained from Malaysia online financial news, which are from Business section of New Straits Times. Our study applies Opinion Lexicon-based algorithm and Naïve Bayes algorithm as the method to perform sentiment analysis. This study consists of several phases in pre-processing such as extract data, stop word removal, and stemming to clean the dataset and make it as data preparation before performing the sentiment analysis with the selected machine learning algorithms. In the stop word removal, tm package in R is used to clean the dataset while for stemming process, Snowball stemmer is used to set the data to its root word. Sample outcomes of analysis are explained for both algorithms. The conclusion describes the summation of the study and future works.
Document Type
Conference Proceeding
Volume
739
School
School of Business and Law
RAS ID
44720
Copyright
subscription content
Publisher
Springer
Comments
Shuhidan, S. M., Hamidi, S. R., Kazemian, S., Shuhidan, S. M., & Ismail, M. A. (2018). Sentiment analysis for financial news headlines using machine learning algorithm. In Proceedings of the 7th International Conference on Kansei Engineering and Emotion Research 2018 (pp. 64-72). Springer, Singapore. https://doi.org/10.1007/978-981-10-8612-0_8