A5101889001- Morphological variations and asymmetry
- Bayesian Methods and Mixture Models
- Advanced Malware Detection Techniques
- Network Security and Intrusion Detection
- Spam and Phishing Detection
- Statistical Methods and Inference
- Soil Geostatistics and Mapping
- Image Retrieval and Classification Techniques
- Gaussian Processes and Bayesian Inference
- Genetic and phenotypic traits in livestock
- Fixed Point Theorems Analysis
- Complex Network Analysis Techniques
- Image and Video Quality Assessment
- Advanced Numerical Analysis Techniques
- EEG and Brain-Computer Interfaces
- Advanced Text Analysis Techniques
- Image Processing and 3D Reconstruction
- Advanced Statistical Methods and Models
- Web Data Mining and Analysis
- Topological and Geometric Data Analysis
- Geometric and Algebraic Topology
- Topic Modeling
- Color perception and design
- Blind Source Separation Techniques
- Neural Networks and Applications
University of Engineering & Management
2014-2023
Athlone Institute of Technology
2020
Birla Institute of Technology, Mesra
2019
Institute of Engineering
2014
Indian Statistical Institute
2011-2013
Florida International University
2011-2012
Jadavpur University
2010
Duke University
2009
University of Arizona
2007-2008
The current work proposes a neural based detection method of two different skin diseases using imaging. Skin images namely Basel Cell Carcinoma and Angioma are utilized. SIFT feature extractor has been employed followed by clustering phase on space in order to reduce the number features suitable for models. extracted bag-of-features modified dataset is used train metaheuristic supported hybrid Artificial Neural Networks classify detect under study. A well-known multi objective optimization...
In this article a nonsingular asymptotic distribution is derived for broad class of underlying distributions on Riemannian manifold in relation to its curvature. Also, the dispersion explicitly related These results are applied and further strengthened planar shape space k-ads.
Abstract Voice communication systems such as Voice-over IP (VoIP), Public Switched Telephone Networks, and Mobile are an integral means of human tele-interaction. These pose distinctive challenges due to their unique characteristics low volume, burstiness stringent delay/loss requirements across heterogeneous underlying network technologies. Effective quality evaluation methodologies important for system development refinement, particularly by adopting user feedback based measurement....
It has become common for data sets to contain large numbers of variables in studies conducted areas such as genetics, machine vision, image analysis and many others. When analyzing data, parametric models are often too inflexible while nonparametric procedures tend be non-robust because insufficient on these high dimensional spaces. This is particularly true when interest lies building efficient classifiers the presence predictor variables. dealing with types it case that most variability...
The amount of malware has been rising drastically as the Android operating system enabled smartphones and tablets are gaining popularity around world in last couple years. One popular methods static detection techniques is permission/feature-based through AndroidManifest.xml file using machine learning classifiers. Ignoring important features or keeping irrelevant may specifically cause mystification to classification algorithms. Therefore, reduce time improve accuracy, different feature...
The voice communication industry is undergoing a rapid phase change with continuously evolving technologies such as cellular, mobile and Internet telephony. Effective evaluation of system performance becoming critical, which will serve an important instrument for service providers to monitor manage quality. Recently, Quality Experience (QoE) metrics are found be more valuable quality assessment mechanism since they closely related human perception compared traditional QoS-based techniques....
Each Android application requires accumulations of permissions in installation time and they are considered as the features which can be utilized permission-based identification malwares. Recently, ensemble feature selection techniques have received increasing attention over conventional different applications. In this work, a cluster based voted technique combining five base wrapper approaches R libraries is projected for identifying most prominent set predictive modeling The proposed...
This project revolves around studying estimators for parameters in different Time Series models and their assymptotic properties. We introduce various bootstrap techniques the obtained. Our special emphasis is on Weighted Bootstrap. establish consistency of this scheme a AR model its variations. Numerical calculations lend further support to our results. Next we analyze ARCH models, study used error distributions. also present resampling estimating distribution estimators. Finally by...