Digital Library[ Search Result ]
Accuracy Evaluation Method for Image Classification Deep Learning Model considering Potentially Misclassified Data
Young-Woo Lee, Min-Ju Song, Heung-Seok Chae
http://doi.org/10.5626/JOK.2021.48.6.657
Image classification deep learning models have a problem of misclassifying the type of image when the image is modified. As the existing accuracy evaluation methods do not take into account images that can potentially be misclassified due to modification of the image, it becomes to trust the evaluation result even if the type of image is accurately classified. In this study, we have proposed a method for evaluating the accuracy of image classification deep learning models considering the potentially misclassified data. We have measured the boundary cost to identify potentially misclassified data for each model and data set and identified potentially misclassified data based on the boundary cost. Also, we have measured the accuracy considering the potentially misclassified data. As a result of identifying potentially misclassified data, 0.1～4.2% of the data were identified as potentially misclassified data, of which approximately 18～60% were actually misclassified. As a result of the accuracy evaluation, it was estimated that the higher-accuracy model was more robust to image modification, and the lower-accuracy model was less robust to image modification.
Patterns of Detecting Feature Interaction in Autonomous Car
Young Woo Lee, Heung Seok Chae
http://doi.org/10.5626/JOK.2019.46.10.1001
In a system with multiple features working together, unexpected behaviors may arise due to feature interaction. An interaction detection method that only considers the dependencies between a system"s components can cause false positives, because interactions are considered to occur between features that are not actually performed simultaneously. In addition, it does not take into account for the interactions that result from the association, such as speed and direction. This paper proposes a pattern with which to detect interaction based on time-series data of the system. In case studies, I classified interaction types by combining interaction attributes, mapped the patterns to identify each interaction, and then performed interaction detection based on the time series data of an autonomous car. For the ACC, OA, LKA, and EVA features of the autonomous car, interactions between speed and direction variables were detected using a non-continuous partitioning pattern and a repetitive partitioning pattern. The interaction resulting from the association between direction and speed variables was detected using a partition conflict pattern.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr