Bioinformatics is THE field of science which applies computer science and information technology ... more Bioinformatics is THE field of science which applies computer science and information technology to the problems of biological science. One of the most useful applications of bioinformatics is sequence analysis. Sequence analysis, which is the process of subjecting a DNA, RNA to any wide range of analytical approaches, involves methodologies like sequence alignment and searches against biological databases. For the analysis DNA sequences are stored in databases for easy retrieval and comparison. Frequency of pattern occurrence in database may predict the intensity of the disease. When the sequence database is huge, matching a pattern is very time consuming task. This fact leads to the need of utilizing latest complex and expensive hardware like GPU. In this paper, we propose a Parallel string matching algorithm using CUDA (Compute Unified Device Architecture). The focus of the research is the design and implementation of an algorithm by utilizing GPU cores optimally. Our algorithms ...
Bulletin of Electrical Engineering and Informatics, Jun 1, 2022
With the rapidly increasing integration of wind energy into the modern energy grid system, wind e... more With the rapidly increasing integration of wind energy into the modern energy grid system, wind energy prediction (WPP) is playing an important role in the planning and operation of an electrical distribution system. However, the time series data of wind energy always has nonlinear and nonstationary characteristics, which is still a great challenge to be accurately predicted. This paper proposes the intelligent wind power forecast model and evaluates to forecast long term, short term and medium term wind power. It uses statistical and machine learning approach for finding the best model for multiperiod forecasting. The model has been tested on Sotavento wind farm historical data, located in Galicia, Spain. The experimental results show that random forest has better accuracy than other models for long term, short term and medium term forecasting. The power prediction accuracy of the proposed model has been evaluated on RMSE, and MAE metrics. The proposed model has shown better accuracy for medium term and long term forecast. The accuracy is improved by 72.12% in case of medium term and 50.49% in case of long term.
2021 IEEE 20th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC)
With the increased adoption of E-learning platforms, keeping online learners engaged throughout a... more With the increased adoption of E-learning platforms, keeping online learners engaged throughout a lesson is challenging. One approach to tackle this challenge is to probe learners periodically by asking questions. The paper presents an approach to generate questions from a given video lecture automatically. The generated questions are aimed to evaluate learners' lowerlevel cognitive abilities. The approach automatically extracts text from video lectures to generates wh-kinds of questions. When learners respond with an answer, the proposed approach further evaluates the response and provides feedback. Besides enhancing learner's engagement, this approach's main benefits are that it frees instructors from designing questions to check the comprehension of a topic. Thus, instructors can spend this time productively on other activities
Learning and Analytics in Intelligent Systems, 2020
Consumer electronic devices such as mobile handsets, goods tagged with RFID labels, location and ... more Consumer electronic devices such as mobile handsets, goods tagged with RFID labels, location and position sensors are continuously generating a vast amount of location enriched data called geospatial data. Conventionally such geospatial data is used for military applications. In recent times, many useful civilian applications have been designed and deployed around such geospatial data. For example, a recommendation system to suggest restaurants or places of attraction to a tourist visiting a particular locality. At the same time, civic bodies are harnessing geospatial data generated through remote sensing devices to provide better services to citizens such as traffic monitoring, pothole identification, and weather reporting. Typically such applications are leveraged upon non-hierarchical machine learning techniques such as Naive-Bayes Classifiers, Support Vector Machines, and decision trees. Recent advances in the field of deep-learning showed that Neural Network-based techniques outperform conventional techniques and provide effective solutions for many geospatial data analysis tasks such as object recognition, image classification, and scene understanding. The chapter presents a survey on the current state of the applications of deep learning techniques for analyzing geospatial data. The chapter is organized as below: (i) A brief overview of deep learning algorithms. (ii)Geospatial Analysis: a Data Science Perspective (iii) Deep-learning techniques for Remote Sensing data analytics tasks (iv) Deep-learning techniques for GPS data analytics(iv) Deep-learning techniques for RFID data analytics.
Duties are segregated within a team by using the role-based access control (RBAC) in the Azure In... more Duties are segregated within a team by using the role-based access control (RBAC) in the Azure Internet of Things (IoT) framework, and only an appropriate level of access is granted to users to perform specific tasks, depending on a given situation. However, the same authentication and authorization mechanism is used for "sort of user,", which increases the operation overload on the cloud server. Moreover, due to its RBAC nature, the IoT framework is inefficient in handling a dynamic situation where multiple users request similar kinds of resources, by creating a several repeated roles. This results in inconsistent and inflexible implementation and the loss of the capability to efficiently address policy management, semantics, redundancy issues in roles, dynamic user handling, work delegation issues, scalability, role explosion, individual rights, and security issues in large organizations. In this work, we designed and presented a novel access control model for a significantly large medical scenario with efficient priority-based authentication mechanisms to address the abovementioned problems associated with the Azure IoT cloud. The proposed model encapsulates the enforcement of priority-based resource access rights across multiple users in a large organization, reduces inefficiency and ineffectuality, and supports individuals with the consistent implementation of policies. We evaluated the benefits of the proposed model by comparing it with existing models and the Azure model, using the health care use-case situation. The comparison results show that by incorporating the priority attribute facility in the existing RBAC model, the proposed model classifies the policy mechanism based on priority attributes and proves that the proposed model is capable of handling problems that generally occur when dealing with huge dynamic scenarios in large organizations. Index Term: Priority scheme; role-based access control (RBAC); Azure internet of things cloud; I. INTRODUCTION C URRENTLY, the Azure Internet of Things (IoT) platform utilizes a customized form of role-based access control (RBAC) policy with predefined roles and groups for access control requirements [1].
Extended ECDSR protocol for energy efficient MANET
2015 International Conference on Advanced Computing and Communication Systems, 2015
MANET is a mobile ad-hoc network without base station. MANET is collection of wireless nodes whic... more MANET is a mobile ad-hoc network without base station. MANET is collection of wireless nodes which are connected to each other, but all nodes are independent. They can communicate each other via intermediate nodes or directly. As the nodes are battery operated energy saving is important issue in the MANET. Since all the nodes are battery powered, failure of one node affect the entire network. In order to increase the lifetime of the network, routing must be energy efficient. Routing protocols like DSR, ESDSR, ECDSR, AODV, TORA [13], EEAODR [12], and EPAR [4]are proposed for MANET. Design of energy efficient routing protocol is the key issue for mobile adhoc networks. ECDSR protocol selects nodes on the basis of minimum threshold energy. As ECDSR protocol has overhearing and stale route problem, which leads to packet loss and over energy consumption. In our paper we proposed the solution to address overhearing and stale route problem by suggesting modification in ECDSR protocol. MANET is used in real time critical applications. If MANET is having dense network, then for saving energy timer technique is being proposed by us in this paper. We illustrated and proved that our proposed methodology works efficiently in the line of our solution.
Super-fast parallel eigenface implementation on GPU for face recognition
2014 International Conference on Parallel, Distributed and Grid Computing, 2014
Eigenface is one of the most common appearance based approaches for face recognition. Eigenfaces ... more Eigenface is one of the most common appearance based approaches for face recognition. Eigenfaces are the principal components which represent the training faces. Using Principal Component Analysis, each face is represented by very few parameters called weight vectors or feature vectors. While this makes testing process easy, it also includes cumbersome process of generating eigenspace and projecting every training image onto it to extract weight vectors. This approach works well with small set of images. As number of images to train increases, time taken for generating eigenspace and weight vectors also increases rapidly and it will not be feasible to recognize face in big data or perform real time video analysis. In this paper, we propose a super-fast parallel solution which harnesses the power of GPU and utilizes benefits of the thousands of cores to compute accurate match in fraction of second. We have implemented Parallel Eigenface, the first complete super-fast Parallel Eigenface implementation for face recognition, using CUDA on NVIDIA K20 GPU. Focus of the research has been to gain maximum performance by implementing highly optimized kernels for complete approach and utilizing available fastest library functions. We have used dataset of different size for training and noted very high increase in speedup. We are able to achieve highest 460X speed up for weight vectors generation of 1000 training images. We also get 73X speedup for overall training process on the same dataset. Speedup tends to increase with respect to training data, proving the scalability of solution. Results prove that our parallel implementation is best fit for various video analytics applications and real time face recognition. It also shows strong promise for excessive use of GPUs in face recognition systems.
International Journal of Distributed and Parallel systems, 2014
Association Rule mining is one of the dominant tasks of data mining, which concerns in finding fr... more Association Rule mining is one of the dominant tasks of data mining, which concerns in finding frequent itemsets in large volumes of data in order to produce summarized models of mined rules. These models are extended to generate association rules in various applications such as e-commerce, bio-informatics, associations between image contents and non image features, analysis of effectiveness of sales and retail industry, etc. In the vast increasing databases, the major challenge is the frequent itemsets mining in a very short period of time. In the case of increasing data, the time taken to process the data should be almost constant. Since high performance computing has many processors, and many cores, consistent runtime performance for such very large databases on association rules mining is achieved. We, therefore, must rely on high performance parallel and/or distributed computing. In literature survey, we have studied the sequential Apriori algorithms and identified the fundamental problems in sequential environment and parallel environment. In our proposed ParApriori, we have proposed parallel algorithm for GPGPU, and we have also done the results analysis of our GPU parallel algorithm. We find that proposed algorithm improved the computing time, consistency in performance over the increasing load. The empirical analysis of the algorithm also shows that efficiency and scalability is verified over the series of datasets experimented on many core GPU platform.
Modeling Rainfall Prediction Using Data Mining Method: A Bayesian Approach
2013 Fifth International Conference on Computational Intelligence, Modelling and Simulation, 2013
Weather forecasting has been one of the most scientifically and technologically challenging probl... more Weather forecasting has been one of the most scientifically and technologically challenging problem around the world. Weather data is one of the meteorological data that is rich with important information, which can be used for weather prediction We extract knowledge from weather historical data collected from Indian Meteorological Department (IMD) Pune. From the collected weather data comprising of 36 attributes, only 7 attributes are most relevant to rainfall prediction. We made data preprocessing and data transformation on raw weather data set, so that it shall be possible to work on Bayesian, the data mining, prediction model used for rainfall prediction. The model is trained using the training data set and has been tested for accuracy on available test data. The meteorological centers uses high performance computing and supercomputing power to run weather prediction model. To address the issue of compute intensive rainfall prediction model, we proposed and implemented data intensive model using data mining technique. Our model works with good accuracy and takes moderate compute resources to predict the rainfall. We have used Bayesian approach to prove our model for rainfall prediction, and found to be working well with good accuracy.
Optimal Positioning of Small Cells for Coverage and Cost Efficient 5G Network Deployment: A Smart Simulated Annealing Approach
2020 IEEE 3rd 5G World Forum (5GWF), 2020
With the escalating load on the current 4G network, the requirement for moving up to 5G has shown... more With the escalating load on the current 4G network, the requirement for moving up to 5G has shown up. The deployment of a 5G network would deliver the end-users a wide spectrum of technologies. But a random deployment of 5G small cell towers to attain better coverage leads to a high increase in cost, an increase in interference, and also a decrease in resource utilization. To solve this Hyper Dense Deployment Problem (HDDP), a Smart Simulated Annealing algorithm with a heuristic to remove excessive small cells and a heuristic to displace the small cells to achieve greater coverage, is adopted. Two different approaches of displacement are considered, Random Displacement Approach and Probed Displacement Approach. The methodology enhances and optimizes the deployment while taking geospatial entities like vegetation, buildings, road networks, etc. into consideration. The newly devised strategy minimizes the cost, optimizes coverage, and thus enables greater resource utilization. The two...
This paper contains the overview of various parallelization techniques to improve the performance... more This paper contains the overview of various parallelization techniques to improve the performance of existing data mining algorithms and make the capable of handling large amount of data. There are variety of techniques to achieve the parallelization in data mining field, in this paper a brief introduction to few of the popular techniques is presented. The second part of this paper contains information regarding various data algorithms that are proposed by various authors based on these techniques. In Introduction various results corresponding to a survey are provided.
Rabi crops play a major role to meet the foodgrains requirement of ever growing population. In In... more Rabi crops play a major role to meet the foodgrains requirement of ever growing population. In India, out of total foodgrain production, rabi crop production is nearly half. Rabi jowar, wheat and gram are the major rabi crops. Though there is significant increase in area and production of rabi crops, the productivity showed decreasing trend in last two decades. In general, most of the farmers are not using the recommended levels of inputs. Therefore, there exists a gap between the recommended and actual use levels of input mix. This leads to a gap in the potential yield and the actual yield of the rabi crops (Jowar, wheat and gram), which is called yield gap. The present investigation was attempted to examine the input use and output levels, to estimate the yield alongwith factors responsible for the yield gap and the constraints in cultivation of jowar, wheat and gram. Jowar, wheat and gram are the major rabi crops of Maharashtra. Solapur is one of the important district rabi jowar...
Data mining is the process of extracting interesting, non-trivial, implicit, previously, unknown ... more Data mining is the process of extracting interesting, non-trivial, implicit, previously, unknown and potentially useful information or patterns from large information repositories. This paper focuses on Association Rule Mining on large image datasets. ARM is largely applied on datasets containing text, but we shall exploit its capabilities to mine images to get interesting and useful correlations and determine the degree of togetherness among faces in the video. Video processing generates a very large dataset which makes it difficult to analyze it manually. Our research model presented in this paper combines two of the most actively researched areas of computer science: Computer Vision and Data Mining.
Uploads
Papers by V B Nikam