Class imbalance occurs when instances in a class are much higher than in other classes. This machine learning major problem can affect the predicted accuracy. Support Vector Machine (SVM) is robust and precise method in handling class imbalance problem but weak in the bias data distribution, Biased Support Vector Machine (BSVM) became popular choice to solve the problem. BSVM provide better control sensitivity yet lack accuracy compared to general SVM. This study proposes the integration of BSVM and SMOTEBoost to handle class imbalance problem. Non Support Vector (NSV) sets from negative samples and Support Vector (SV) sets from positive samples will undergo a Weighted-SMOTE process. The results indicate that implementation of Biased Support Vector Machine and Weighted-SMOTE achieve better accuracy and sensitivity.
16 March 2018 in Journal of Physics: Conference Series
Skin color segmentation has proven to be useful in various application, such as face detection, hand gesture analysis, image content filtering, etc. There are some challenges on skin-color detection, such as wide variety of illumination condition and skin-like color objects appear in the background of image. Therefore, a method is required to cope with illumination condition while detecting skin color. In our proposed method, we combine the chrominance channels of three color spaces, namely HSV, YCbCr and Normalized RGB, to produce better rate result for skin color segmentation. The result shows that the proposed method is able to detect skin pixel and get up to 91.05% correct result while being tested in ECU and HGR Dataset. This result is significantly higher than the other single color spaces.
25 January 2018 in IOP Conference Series: Materials Science and Engineering
Class imbalance is a situation where instances in one class much higher than instances in other classes. In clustering, this problem not only affects the accuracy of a prediction but also introduces bias in decision-making process. In this case, a machine learning technique will yield a good prediction accuracy from training data class with a large number of instances, but give a poor accuracy in classes with the small number of instances. In this research, we propose an approach for optimizing K-Means clustering in handling class imbalance problem. The approach uses the perceptron feed-forward neural network to determine coordinates of the centroid of a cluster in K-Means clustering processes. Data used in this research are datasets from the UCI Machine Learning Repository. From the experimental results obtained, the proposed approach could optimize the result of K-Means clustering in terms of minimizing class imbalance.
Future of data mining is the social network mining, especially in connection with the web pages, will incorporate the network timeline that are generated with social network extraction. This approach we use to deal with dynamic and process of evolution in networks. Although there is an increasing interest about the social network analysis, but a little of them has a significant impact to social substructures mining. Therefore this paper proposes the mining of social network based on unit analysis in social network analysis to build a network: vertex and edge. In study we used multiple regression to determine the relations of resources of networks, but we explore naturally formal relation of vertices and edges like leadership of an author, and then we explained in experiments. So there is the role of each actor in the social networks.