Abstract Emerging multimedia applications demands content-based video processing. Video has to be segmented into objects for content-based processing. A number of video object segmentation algorithms have been proposed such as semiautomatic and automatic. Semiautomatic methods adds burden to users and also not suitable for some applications. Automatic segmentation systems are still a challenge, although they are required by many applications. The proposed... Emerging multimedia applications demands content-based video processing. Video has to be segmented into objects for content-based processing. A number of video object segmentation algorithms have been proposed such as semiautomatic and automatic. Semiautomatic methods adds burden to users and also not suitable for some applications. Automatic segmentation systems are still a challenge, although they are required by many applications. The proposed work aims at contributing to identify the gaps that are present in the current segmentation system and also to give the possible solutions to overcome those gaps so that the accurate and efficient video segmentation system can be developed. Read More Read Less

Abstract Outlier detection always plays a major role in different industrial domains ranging from major financial fraud detection to network intrusions and also major clinical diagnosis of different diseases. Over the years of research, distance based outlier detection algorithm and density based outlier detection algorithm have emerged as a viable, scalable, parameter free alternative way to the very long before the traditional... Outlier detection always plays a major role in different industrial domains ranging from major financial fraud detection to network intrusions and also major clinical diagnosis of different diseases. Over the years of research, distance based outlier detection algorithm and density based outlier detection algorithm have emerged as a viable, scalable, parameter free alternative way to the very long before the traditional approaches. In this paper, we evaluate both the distance and density based outlier detection approach applying on a time-series data i.e. stock market data traded for three months. We begin by surveying and examining the design landscape of the above outlier detection approaches. Also implemented the outlier detection framework for the said design and analysis to understand the various pros and cons of various optimizations proposed by us on a real-time data set i.e. the current stock market data set. The combinations of optimization techniques (factors) and strategies through distance and density based outlier approaches always dominate on various types of data sets. So plays a dominant role on this real-time stock market data set. Read More Read Less

Abstract Nowadays, handling with massive data at big rate or large-scale data mining has become new task. Rough set theory for knowledge learning or developing is applied in data mining. In this project we are going to present a serial as well as a parallel method [1]. These serial and parallel methods are used for computing rough set approximations. We will implement the... Nowadays, handling with massive data at big rate or large-scale data mining has become new task. Rough set theory for knowledge learning or developing is applied in data mining. In this project we are going to present a serial as well as a parallel method [1]. These serial and parallel methods are used for computing rough set approximations. We will implement the parallel algorithm on Hadoop Map Reduce platform [3, 5]. We will evaluate the performance of three parallel algorithms with respect to its speedup, scale up and size up[7]. The Hadoop Map Reduce platform is a programming model and also it is a software framework for developing applications. The Hadoop Map Reduce platform works on massive data and also it the data gets processed rapidly in parallel on large clusters of compute notes. In this project, we will evaluate only the efficiency of the algorithms and not its accuracy. By using parallel method on Hadoop platform, due to transparent spilling, the data size limit will be get removed. However, algorithms based on rough set theory are quite a challenging task for the applications of enlarged data. Read More Read Less