Artificial Intelligence Reveals Hidden Patterns in Indonesian Financial Markets

Strategic Insights -

The rapid growth of textual data has profoundly impacted how we predict and understand financial market behavior. A recent study conducted at Telkom University has unveiled new insights into the connections between news sentiment and stock market behavior of Indonesian companies. By leveraging advanced AI-based sentiment analysis and natural language processing techniques, researchers have explored the influence of news sentiment on stock prices at individual, industry, and media co-occurrence cluster levels. The results have led to the development of a unique analytical framework that offers new perspectives on AI-driven investment strategies and highlights the potential of big data analytics in Indonesia's capital market.

Key Takeaways:

  • The study analyzed a dataset of 192,582 Indonesian financial news articles published between 2018 and 2023 to explore the connections between news sentiment and stock market behavior.
  • The research utilized AI-based sentiment analysis and natural language processing techniques, including identity recognition, network analysis, and correlation assessment.
  • The results showed that sentiment-price correlations strengthened hierarchically, with individual companies at 0.26, industry groupings at 0.30, and news co-occurrence clusters at 0.43.
  • The study concluded that companies frequently mentioned together in the news exhibit stronger and more stable sentiment-price correlations, offering a new analytical perspective for AI-driven investment strategies.
  • The research demonstrated the potential of big data analytics in Indonesia's capital market, highlighting the importance of media co-occurrence clusters in understanding stock market behavior.

Statistics:

  • 192,582 Indonesian financial news articles were analyzed between 2018 and 2023.
  • The study found a sentiment-price correlation of 0.26 at the individual company level.
  • The correlation increased to 0.30 at the industry grouping level.
  • The strongest correlation was observed at the news co-occurrence cluster level, with a value of 0.43.
  • The research references a dataset of 192,582 articles, however, it does not specify the exact number of news articles analyzed.

Sources:

  • "Deciphering news sentiment and stock price relationships in Indonesian companies: an AI-based exploration of industry affiliation and news co-occurrence" published in Discover Artificial Intelligence, 2025, 5(1):1-29.
  • "Study Results from Telkom University Broaden Understanding of Artificial Intelligence" by NewsRx LLC, June 23, 2025; p 818.