{"title":"Pruning of Random Forest classifiers: A survey and future directions","authors":"V. Kulkarni, Pradeep Sinha","doi":"10.1109/ICDSE.2012.6282329","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6282329","url":null,"abstract":"Random Forest is an ensemble supervised machine learning technique. Based on bagging and random feature selection, number of decision trees (base classifiers) is generated and majority voting is taken for classification. For effective learning and classification of Random Forest, there is need for reducing number of trees (Pruning) in Random Forest. We have presented here systematic survey of pruning efforts of Random Forest classifier along with the required theoretical background. Most of the work for pruning takes static approach while recently dynamic pruning is being targeted. We have also generated a Comparison Chart by taking relevant parameters. There is research scope for analyzing behavior of Random forest, generating accurate and diverse base decision trees, truly dynamic pruning algorithm for Random Forest classifier, and generating optimal subset of Random forest.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129540312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A novel stereo matching technique for radiometric changes using Normalized Cross Correlation","authors":"M. George, R. R. R. Rejimol","doi":"10.1109/ICDSE.2012.6282305","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6282305","url":null,"abstract":"In real world situations, image color values between two stereo images are often affected by various radiometric factors such as illumination direction, illuminant color, and imaging device changes. So, the assumption of color consistency does not hold well in stereo images. Thus, the performance of most conventional stereo matching algorithms would be severely degraded under radiometric variations. The main focus of this work is on illumination invariant stereo matching by generating illumination invariant images from stereo image data using a non-iterative normalisation in log RGB space. The actual stereo matching is done using a local similarity measure, Normalized Cross-Correlation (NCC) which is the standard statistical method for determining similarity which itself is invariant to linear brightness and contrast variations. Error analysis is done by dividing disparity into uniform and discontinuity regions. Experimental results show that our algorithm outperforms other stereo matching algorithms under severely different radiometric conditions between stereo images.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"333 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123400361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Clustering data set with categorical feature using multi objective genetic algorithm","authors":"D. Dutta, P. Dutta, J. Sil","doi":"10.1109/ICDSE.2012.6281897","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6281897","url":null,"abstract":"In the paper, real coded multi objective genetic algorithm based K-clustering method has been studied where K represents the number of clusters known apriori. The searching power of Genetic Algorithm (GA) is exploited to search for suitable clusters and cluster modes so that intra-cluster distance (Homogeneity, H) and inter-cluster distances (Separation, S) are simultaneously optimized. It is achieved by measuring H and S using Mod distance per feature metric, suitable for categorical features (attributes). We have selected 3 benchmark data sets from UCI Machine Learning Repository containing categorical features only. Here, K-modes is hybridized with GA to combine global searching capabilities of GA with local searching capabilities of K-modes. Considering context sensitivity, we have used a special crossover operator called “pairwise crossover” and “substitution”.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127793114","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"PccP: A model for Preserving cloud computing Privacy","authors":"S. M. Rahaman, M. Farhatullah","doi":"10.1109/ICDSE.2012.6281900","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6281900","url":null,"abstract":"The widespread focus on the Cloud Computing has necessitated the corresponding mechanisms to ensure privacy and security. Various attempts have been made in the past to safeguard the privacy of the individual or agency trying to utilize the services being provided by the cloud. The most challenging task is to provide services to the users while also preserving the privacy of the user's information. In this paper a model that incorporates a three-level architecture, Preserving cloud computing Privacy (PccP) model is proposed which aims to preserve privacy of information pertaining to cloud users. The Consumer Layer deals with all the aspects which relate to enabling the user of the cloud to access the cloud services being provided by the cloud service provider. The Network Interface Layer creates an appropriate mapping between the original IP addresses of the users with a modified IP address, and thereby ensuring the privacy of the IP address of the users. The Privacy Preserved Layer utilizes the functionality of the Unique User Cloud Identity Generator for which an algorithm is proposed in this paper to generate an unique User Service Dependent Identity(USID) with privacy check by establishing mapping among the existing user identity(ID), if any to ID's available in a pool of User ID's to enhance the privacy of sensitive user information. A Privacy check method based on information privacy is being proposed which contributes significantly in maintaining user control over the generated user identities (USID's).","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121931846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
N. Gaur, S. Mahammad, V. Sharma, D. Dhar, R. Ramakrishnan
{"title":"Template based HDF5 satellite digital data product generation software","authors":"N. Gaur, S. Mahammad, V. Sharma, D. Dhar, R. Ramakrishnan","doi":"10.1109/ICDSE.2012.6281903","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6281903","url":null,"abstract":"Satellite data is complex and voluminous and the details are essential. Managing and storing satellite data present many challenges, such as, handling huge volumes of data efficiently, managing quality, proper attribution, citation and integration of different nature of data such as images, ancillary, attitude and orbit related information. HDF5 (Hierarchical Data Format) is a multi object file format for sharing scientific data in a distributed environment. This paper describes a template based approach for standardizing the process of designing and developing software for digital data products using HDF5. The idea to generate HDF product from template is inspired from XML Schema based approach for generating HDF5 products. There is no standard tool available to directly convert XML Schema to HDF5. The XMLtoHDF5 software being developed by HDF group is still in the early stage of development. After getting motivation from this approach, an ASCII based template has been defined specifically for satellite data products, also software that parses the template and generates HDF5 digital data products have been designed and developed. The design of the software and the template are explained in this paper. This approach is used for generating RESOURCESAT-2 digital data products that are voluminous in nature containing images of the scene under consideration and related ancillary information. Validity of generated product is checked by h5check utility and it is in compliance with hdf5 format.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122025131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Participatory sensing for emergency communication via MANET","authors":"M. Ramesh, A. Jacob, R. D. Aryadevi","doi":"10.1109/ICDSE.2012.6282316","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6282316","url":null,"abstract":"Most large-scale disasters are characterized by breakdown of essential communication systems. The failure of telecommunication system reduces the pace of emergency response and relief efforts. The paper is confined mainly on how communication can be carried even when a cellular network is down. The proposed system designs a participatory system for emergency communication. The existing participatory sensing systems have not addressed the issues arising when an area is hit with a disaster. By creating mobile ad hoc network of smart phones, it would be possible to allow messages requesting help to pass between phones until they are intercepted by a node which can directly connect to the cellular network or internet. An algorithm for routing the message in such a network is presented. It also showcases an android application called Android app for Emergency Communication (AEC) which would perform this task.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129593215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Adaptive controller for improved performance of brushless DC motor","authors":"N. Leena, R. Shanmugasundaram","doi":"10.1109/ICDSE.2012.6281896","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6281896","url":null,"abstract":"This paper presents the development and performance analysis of model reference adaptive controller using Artificial Neural Network (ANN) for Brushless DC motor (BLDC) drives. The model reference adaptive systems (MRAS) have a parameter adjustment mechanism along with the normal feedback loop and hence give better solutions when there are variations in process parameters. Neural networks (NNs) with their inherent parallelism, learning capabilities and fault tolerance have proven to be a promising solution in estimating and controlling nonlinear systems. This paper combines a MRAS with ANN to solve the problems of non-linearity, parameter variations and load excursions that occur in BLDC motor drive systems. The performance of the traditional PID controller based speed control method is compared with the model reference based speed control for BLDC motor drive system using MATLAB Simulink software. Simulation results are presented to prove that the MRAC based model is capable of speed tracking as well as reduce the effect of parameter variations.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"517 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134091451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Real time computer assisted surgical navigation","authors":"N. T. Pallath, T. Thomas","doi":"10.1109/ICDSE.2012.6282322","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6282322","url":null,"abstract":"Several studies have proved the efficacy of computer assistance in spinal surgery, especially in the technique of pedicle screw insertion for vertebral fusion and correction of deformities. Computer assisted surgical navigation systems have proven success in the surgical treatment of spinal fractures and disorders. We have developed a novel method for real time computer assisted spinal surgery, based on online video captured by using CCD camera. The paper presents a real time computer assisted spinal surgery for pedicle screw insertion. A video is considered to be a sequence of frames. Image processing techniques applies to video processing as well, however speed of processing is crucial for real time video processing. A system consisting of workstation computer, image sensors or cameras, digitizer and image processing hardware have been used for real time video processing.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121696225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Prediction on proactive channel sensing strategy for Cognitive Radio","authors":"J. Loganathan, S. Sarang, T. Geetha","doi":"10.1109/ICDSE.2012.6282315","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6282315","url":null,"abstract":"Sensing optimal idle channel for future transmission is the hot research topic in Cognitive radio (CR) network environment. CR Network performance greatly depends upon the waiting time of the Cognitive User (CU) for getting free channel from licensed band. Probability of available free channel will be less when arrival rate of primary user to their allocated licensed band is high. So the effective channel selection and switching strategy needed to overcome this issue and increasing throughput of the data transmission over cognitive radio network. In this paper focusing prediction on channel selection scheme pro-actively and comparing the performance with existing reactive prediction polices.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130470687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
V. Preeja, K. A. Abdul Nazeer, S. S. Vinod Chandra
{"title":"Common structural motif identification in genomic sequences","authors":"V. Preeja, K. A. Abdul Nazeer, S. S. Vinod Chandra","doi":"10.1109/ICDSE.2012.6282301","DOIUrl":"https://doi.org/10.1109/ICDSE.2012.6282301","url":null,"abstract":"Identification of structural and sequence motifs in genomic sequences is gaining much attention now a days. Ribonucleic acid or RNA is one of the important biomolecule whose secondary structure defines its functionality. Soft computing techniques like genetic programming have been used for motif identification. In this paper, we propose a method for identifying common structural motifs in a set of RNA sequences by particle swarm optimization technique. The common motif identification finds its application in drug design. To prove the correctness of our approach, we tested it on data sets previously used to find common structural motif using genetic programming approaches. Several experiments were conducted using Signal Recognition Particle (SRP), Ferritin Iron Response Element (IRE) and microRNA sequences. Our results are comparable to the latest genetic programming approach. Some of the results indicate that our method out performs to most of the genetic programming approaches.","PeriodicalId":293272,"journal":{"name":"2012 International Conference on Data Science & Engineering (ICDSE)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117117817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}