Abstract. As one of the most popular research fields in machine learning, the research on imbalanced dataset receives more and more attentions in recent years. The imbalanced problem usually occurs in when minority classes have extremely fewer samples than the others. Traditional classification algorithms have not taken the distribution of dataset into consideration, thus they fail to deal with the problem of class-imbalanced learning, and the performance of classification tends to be dominated by the majority class. SMOTE is one of the most effective over-sampling methods processing this problem, which changes the distribution of training sets by increasing the size of minority class. However, SMOTE would easily result in over-fitting on account of too many repetitive data samples. According to this issue, this paper proposes an improved method based on sparse representation theory and oversampling technique, named SROT (Sparse Representation-based Over-sampling Technique). The SROT uses a sparse dictionary to create synthetic samples directly for solving the imbalanced problem. The experiments are performed on 10 UCI datasets using C4.5 as the learning algorithm. The experimental results show that compared our algorithm with Random Over-sampling techniques, SMOTE and other methods, SROT can achieve better performance on AUC value.
Abstract. Bio-inspired polarization navigation is a promising navigation method inspired by insects’ autonomous foraging and homing behaviour. Many insects acquire their spatial orientation by sensing the polarization pattern of the skylight. We propose utilization of solar meridian in the polarized skylight as an orientation cue because of its significant features. Using its features, we then design and construct an imaging polarization navigation prototype. The prototype consists of a field-division polarization imaging sensor, the corresponding software, an interface, and the solar-meridian recognizing and measurement algorithm. The field-division polarization imaging sensor is the core component of the prototype and acquires polarized intensity images. To adapt to the demand of real-time on navigation system, we then propose an optimized real-time polarization image processing and pattern recognition algorithm based on Hough transform. The azimuth measurement accuracy of the sensor is then calibrated using a facility that is able to get higher azimuth accuracy by measurement of the star light. To verify the navigation capability of the developed system, we use a dynamic experiment, where the prototype is installed on the top of a vehicle and its navigation performance is compared with GNSS.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.