Abstract
In recent years, technological developments of electronic circuit have made it possible to detect odors. There has been considerable interest in odor sensors in various fields. In our research, we utilize quartz crystal microbalance sensors as odor sensors because they are inexpensive and have similar properties of human nose. Although an odor is measured by combining a lot of Quartz Crystal Microbalance sensors having different performances for odor detection, it is difficult to detect the specific odor that we want. Additionally, the detection of mixed odor is more difficult since the data of mixed odor is imbalance in practice. To solve this problem, in this paper, we propose a new odor detecting algorithm for imbalanced data improving support vector machine. The characteristic point of the proposed method is it enables us to compensate hyperplane for support vector machine by improving penalty term of it. From the experimental results with some datasets, we achieved that our proposed method allowed us to classify imbalanced data better than support vector machine and we confirmed it was feasible and effective for odor detection.
