This is a repository for TKU-PSO, a particle swarm optimization algorithm for top-k high-utility itemset mining in quantitative transactional data.
How to run:
- Select algorithm parameters in TKU_PSO.java (k, pop_size, and iterations).
- Set the "input" string in TKU_PSO.java to the path of the database file. The file must be in SPMF format, here are some example datasets.
- Set the "output" string to any .txt file path. The discovered patterns are written to this file during execution.
- Run main.java
The project also contains logic for validating algorithm accuracy in FindAcc.java.