Abstract
Stock investment decisions are often made based on current events of the global economy and the analysis of historical data. Conversely, visual representations may assist investors gain a deeper understanding not only on the market overview structure but detailed information on specified targets including useful insights on stock market trends. The trend analysis is based on long-term data collection. This study adopts a hybrid method that combines both clustering algorithms and force-directed algorithms to overcome the scalability problem when visualizing large datasets. This methodology exemplifies the potential relationships and interaction among each individual stock, as well as determining the strength of connectivity, which in turn will provide investors adifferent angle of view on the stock relationships. Information derived from visualization will also assist investors to make better informed decisions with less human disturbance due to pure mathematic calculations. The results of the experiments reflect that the proposed method can produce visualized data aesthetically by providing clearer views on an entire structure and specific connections.
Original language | English |
---|---|
Pages (from-to) | 18-27 |
Number of pages | 10 |
Journal | International Journal of Engineering Science and Innovative Technology |
Volume | 6 |
Issue number | 1 |
Publication status | Published - 2017 |
Keywords
- information visualization
- investments
- stock exchanges
- stocks
- prices