PBSVM: Partitioning and biased support vector machine for vocal fold pathology assessment using labeled and unlabeled data sets |
| |
| Authors: | Tahereh Emami Azadi Farshad Almasganj |
| |
| Affiliation: | 1. The key laboratory for advanced food manufacturing equipment technology of Jiangsu province, Wuxi, 214122, China;2. School of Mechanical Engineering, Jiangnan University, Wuxi, 214122, China;3. College of Electrical and Information Engineering, Hunan University, Changsha, 410082, China;4. National Center of Quality Supervision and Inspection for Commodity, Yiwu, 322000, China;1. W. Szafer Institute of Botany, Polish Academy of Sciences, Lubicz 46, 31-512 Kraków, Poland;2. Real Jardín Botánico, CSIC, Plaza de Murillo 2, 28014 Madrid, Spain;1. State Key Laboratory for Strength and Vibration of Mechanical Structures, Shaanxi Engineering Research Centre of NDT and Structural Integrity Evaluation, School of Aerospace Engineering, Xi’an Jiaotong University, Xi’an 710049, Shaanxi, China;2. Leading Edge NDT Technology (LENDT) Group, Malaysian Nuclear Agency, 43000 Bangi, Kajang, Selangor, Malaysia;1. Irish Sea Fisheries Board (BIM), New Docks, Galway, Ireland, UK;2. Coastal & Marine Research Centre (CMRC), University College, Cork, Ireland, UK;3. Marine Institute (MI), Oranmore, Galway, Ireland, UK |
| |
| Abstract: | Most of the existing classification methods, used for voice pathology assessment, are built based on labeled pathological and normal voice signals. This paper studies the problem of building a classifier using labeled and unlabeled data. We propose a novel learning technique, called Partitioning and Biased Support Vector Machine Classification (PBSVM), which tries to utilize all the available data in two steps: (1) a new heuristically partition-based algorithm, which extracts high quality pathological and normal samples from an unlabeled set, and (2) a more principle approach based on biased formulation of support vector machine, which is fairly robust to mislabeling and unbalance data problem. Experiments with wavelet-based energy features extracted from sustained vowels show that the new recognition scheme is highly feasible and significantly outperform the baseline classical SVM classifier, especially in the situation where the labeled training data is small. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
