Predicting Stock Market Movements Through Multisource Data Fusion Graphs: An Approach Employing Graph Convolutional Neural Network
- Title
- Predicting Stock Market Movements Through Multisource Data Fusion Graphs: An Approach Employing Graph Convolutional Neural Network
- Creator
- Ranjith J.; Chandar S.K.
- Description
- The stock market plays an important role in the capital market, and investigating price fluctuations in the stock market has consistently been a prominent subject for researchers. The application of soft computing techniques to predict and categorize stock market movements is a significant research challenge that has gathered considerable attention from researchers. Although several studies highlight the significance of incorporating information from two sources in stock movement prediction, the potential of advanced graphical techniques for modeling and analyzing multi-source data remains an unattended research area. This study aims to address this gap by introducing a novel model that utilizes multi-source data fusion graphs to predict future market movements. The primary challenge involves establishing a model that can effectively gather the relationships among various data sources and employ this understanding to improve prediction performance. Compared to several existing methods relying only on historical data or sentiment data, which show limited predictive power and lack generality, the proposed approach seeks to overcome these limitations. The proposed model integrates various information sources, including historical prices, news data, Twitter data, and technical indicators for predicting future stock market trends. This presented method involves constructing a subgraph map for each data type to capture events from both rising and falling markets. Then, a Gated Recurrent Unit (GRU) is employed to aggregate the subgraph nodes. These aggregated nodes are then integrated with a Graph Convolutional Neural Network (GCNN) to classify the multi-source graph, therefore achieving stock market trend prediction effectively. To further validate its effectiveness, the presented model is applied to Indian stock market data, demonstrating its feasibility in fusing multi-source stock data and establishing its suitability for effectively predicting stock market movements. 2024 Seventh Sense Research Group
- Source
- International Journal of Engineering Trends and Technology, Vol-72, No. 6, pp. 8-18.
- Date
- 2024-01-01
- Publisher
- Seventh Sense Research Group
- Subject
- Data fusion; Gated Recurrent Unit; Graph Convolution Neural Network; Multi-source data; Multi-source data fusion graphs; Stock market trends; Stock trend prediction
- Coverage
- Ranjith J., School of Business and Management, CHRIST (Deemed to be University), Karnataka, Bengaluru, India; Chandar S.K., School of Business and Management, CHRIST (Deemed to be University), Karnataka, Bengaluru, India
- Rights
- Restricted Access
- Relation
- ISSN: 23490918
- Format
- Online
- Language
- English
- Type
- Article
Collection
Citation
Ranjith J.; Chandar S.K., “Predicting Stock Market Movements Through Multisource Data Fusion Graphs: An Approach Employing Graph Convolutional Neural Network,” CHRIST (Deemed To Be University) Institutional Repository, accessed February 24, 2025, https://archives.christuniversity.in/items/show/13071.