Rule-selection through social computing : a stock-trading rule classification system using the Ant Miner algorithm

• Main Author: Aditya Malla
• Other Authors: Ng Geok See
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer applications::Computers in other systems
• Online Access: http://hdl.handle.net/10356/41826
• Institution: Nanyang Technological University

Stock market has been the centre of attraction for researchers and practitioners in the recent years. Different techniques have been used in the trading community for prediction tasks and recently the concept of Nature‐Inspired Social Algorithms (E.g. Ant Colony Optimization metaheuristic) has emerged out as one of them. Being a relatively new technique, there is a lack of research focusing on application of ACO or ACO based techniques to Stock Market price predictions. Initially, we propose the usage of statistical and technical analysis techniques to estimate the optimum input parameters for our prediction model. This affects the efficiency and speed of the system while making predictions. In our benchmark comparisons, the usage of these input dimensions has shown excellent results by increasing the prediction accuracy by more than 7 – 8%.