| Summary: | 博士 === 國立清華大學 === 工業工程與工程管理學系 === 100 === Searching for an optimal feature subset from a high-dimensional feature space is an NP-complete problem; hence, traditional optimization algorithms are inefficient when solving large-scale feature selection problems. Therefore, meta-heuristic algorithms are extensively adopted to solve such problems efficiently. This study proposes two approaches: opposite sign test and regression-based particle swarm optimization for feature selection problem. The proposed algorithms can increase population diversity and avoid local optimal trapping by improving the jump ability of flying particles. The data sets collected from UCI machine learning data bases are used to evaluate the effectiveness of the proposed approaches. Classification accuracy is used as a criterion to evaluate classifier performance. Results show that our proposed approaches outperform both genetic algorithms and sequential search algorithms.
In addition, a real case about the diagnosis of obstructive sleep apnea (OSA) using the proposed approach is presented. Through the implementation of this real case study, we found that the proposed approach could be applied as a screening tool for early OSA diagnosis. As a result, PSO can be applied to assist doctors in foreseeing the diagnosis of OSA before running the PSG test, allowing the medical resources to be used more effectively.
|
|---|
