News analytics to predict Malaysian telco stock price performance: data visualization

• Main Authors: Nor, Rozi Nor Haizan, Ismarau Tajuddin, Nur Ilyana, Mahathir, Muhammad Akmal, Jusoh, Yusmadi Yah, Aziz, Khairi Azhar
• Format: Conference or Workshop Item
• Published: IEEE 2023
• Online Access: http://psasir.upm.edu.my/id/eprint/37631/; https://ieeexplore.ieee.org/document/10145191

The stock market refers to the collection of markets and exchanges where regular activities of buying, selling and issuance of shares of publicly held companies take place. Stock Market Index is the reading used in comparing the price of the stock market each day and it is very dynamic and susceptible to quick changes therefore a large dataset is needed to create a proper prediction model. Many believe that it follows a random walk pattern. In this project, machine learning is used so that a prediction can be made by monitoring past index patterns and predicting how they would behave in the future. Natural Language Processing is a scoring process of a word in a sentence that translates the word meaning into machine-readable value. With this, we can identify a certain topic of a stock market to be either positive or negative. Knowing these values, we can train the machine learning model to predict the price index trend and see whether news and posts related to stock price affect the market. The result of the prediction is visualized and analyzed to identify the accuracy of the data and view the disparity between the predictions and the actual value.