Rajeev Pandey

Content based Structural Recognition for Image Classification using PSO Technique and SVM

The issue of SVMs parameter optimization with particle swarm optimization (pso) provide the optimum solution. This new classification approach may be an efficient alternative, in existing paradigms. PSO technique work with high dimensional datasets and mixed attribute data. The structure of the image is recognized through PSO technique which provide optimized parameter for SVM. This approach determines the performance of image classification after structural recognition based on content of image and comparing the obtained results with those reported for various other classification approaches. PSO-SVM technique can be applied mixed-attribute, hyperspectral data, hyperdimension spaces & problem description spaces and it can also be a competitive alternative to well established classification techniques. The optimized process of data reduces the unclassified region of support vector machine and improves the performance of image classification. The feature of region of image is classified by PSO-SVM technique in inside the image. Cassified features are increase recogniztion ratio because the feature of image is optimized. General Terms Pattern Recognition, high dimensional image classification et. al.

A Hybrid Approach for Sequence Alignment over Genome Data using Compressive Sensing and HBLAST

Medical data is an exponential growth in all the hospitality service area. Genome is an special type of data which deals with the small unit of medical cells. Various matching operation over the genome data is required because of some medical issues arise in various cases. DNA matching, sequence matching, pattern analysis and matching is so called requirement in this area. There are some techniques such as BLAST, HBLAST, RMAP is involved and performed by past researcher. The technique use pre-processing and other filteration , sequence finding is performed. Past approach finds limitation where the large data processing, sequence detection and combine score generation for overall data processing is not performed. In this paper proposed approach is given which work towards the enhancement of previous approach extended with compressive sensing usage for prefetching of data and its filteration. It make use of compressive sensing with which a noise removal, filtering process is executed and thus a refined data is observed for Hadoop processing Mapping approach. Our proposed technique executed with different data set of sequence, count of data present in millions and it gives an effective results while comparing with existing scenario. A further implementation on security usage can performed by us.

Trusit centric approach based on similarity in VANET

VANET (Vehicular ad hoc network) is a network which mainly consists of mobile nodes which are categorized by high speed, dynamically changing topology, lack of any central authority, relationship between nodes are often short lived. It is a network which mainly focus on providing applications which save human lives and provide comfort to drivers during driving. So, Communication which are of two types, vehicle to vehicle (V2V) or Inter-vehicle Communication(IVC) and vehicle to Infrastructure(V2I) or Vehicle to Road Side Units(V2R), plays a major role because there is need that safety message to delivered on time without being any manipulation done by the intermediate nodes. Security plays a major role in VANET for the reliable and efficient delivery of message, Trust a key element of security. It is basically a belief which one node has over the other when it receives information from it. This trust can either be on the node i.e. Entity Centric Trust which provided the information or this trust can be developed on information provided by the node, that is data centric trust or can be on both node as well as on information, that is combined trust. In this, we focus on data centric approach and introduce a similarity metric aided by the infrastructure, to propose our own trust model. Simulation results, shows the accuracy of our trust scheme in respect with the introduction of malicious data.

Phase level energy aware map reduce scheduling for big data applications

The preponderance of large scale data radical applications executed by various business areas for performing data preparation and data analytics based on Map Reduce paradigm which can be better implemented on Hadoop. Such data driven applications which are executed on large clusters set up in data centers hike the energy cost which imposes burden on overall data center cost. Thus minimizing parameter that guides energy consumption becomes paramount requisite to be considered. In this paper we propose a framework for improving energy efficiency of Map Reduce applications. We propose phase level energy aware map reduce scheduling algorithms that assign map and reduce task to system on the basis of maximum node availability. We perform various extensive experiments on Hadoop cluster to determine execution time and energy consumption for several workloads from Hadoop including Terasort and K-means clustering and results evaluated that proposed algorithm consume less energy than various heuristic algorithms and minimizes execution time.