BITS Faculty Publications
Permanent URI for this communityhttp://localhost:4000/handle/123456789/1867
Browse
14 results
Search Results
Item Ensemble Gaussian mixture model-based special voice command cognitive computing intelligent system(Sage, 2020-08) Jangiti, SaikishorDysarthria is a speech disorder caused by stroke, Parkinson’s disease, neurological injury, or tumors that damage the nervous system and weaken the speech quality. Developing a unique voice command system for Dysarthric speech helps to recognize impaired speech and convert them into text or input commands. Hidden Markov Model (HMM) is one of the widely used generative model-based classifiers for Dysarthric speech recognition. But due to insufficient training data, HMM doesn’t provide optimal results on overlapping classes. We propose an ensemble Gaussian mixture model to recognize impaired speech more accurately. Our model converts the sequence of feature vectors into a fixed dimensional representation of patterns with varying lengths. The performance efficiency of the proposed model is evaluated on the Dysarthric UA-speech benchmark dataset. The discriminatory information provided by the proposed approach yields better classification accuracy even for shallow intelligibility words compared to conventional HMM.Item DMK-Medoid Heuristic Product Ranking in Online Market(Research India Publications, 2014) Jangiti, SaikishorFor the past two decades Web Technology (WT) and online shopping has witnessed a monstrous growth. Everyday millions and billions of online shopping websites are growing in the internet for purchasing all types of commodities without disturbing the currency deliberately. Product analysis is an inevitable and umpteen numbers of methodologies are available in the existing world. In this paper, commodity like washing machine is taken as a prime product for analysis in the sites. Semantics has been formulated using different online websites cart. K-medoid clustering (crisp) and fuzzy K-means are employed for analyzing and mining a worthwhile cost and warranty. While making the budget, applying distributed measures and normalization in the above said semantics we have used Distributed Measure (DM) K-Medoid architecture. After the two methodologies crisp and fuzzy has been employed we arrive with that of the actuals. Through variance both budget and actuals are gauged. For attribute measures used gini and theil indexing in the semanticsItem BEST ELECTRONIC SHOPPING TECHNIQUE (BEST)-AN ADHOC COMPONENT USING BAT MODEL(Asian Research Publishing Network, 2015) Jangiti, SaikishorThe rapid development of internet and technology made internet based virtual electronic shops come true. The moral stuff of online shopping is “Avail anything from anywhere at any time”. Online shopping paves a sophisticated way for customers to buy the commodity in less time. It helps the customers to know the feedback about the commodity that makes them to take corrective decisions. In addition to that, it serves the privacy to the customers as the traditional way does not suit this. Comparison shopping have emerged a new path to the online shopping. It assists the customers to compare the ‘N’ commodity simultaneously. However, umpteen number of online shops exists that makes a gap, since no website contribute adequate solution to meet the request of the customers about other aspects of commodity such as warranty, delivery days, review rating, quantity, EMI, COD, shipping cost, compare option etc. In this paper we are proposing BEST - Best Electronic Shopping Technique with a new model called TIM (Training Set - Interface - Model) which is imbibed, considering important attributes from top 20 popular sites to perform evaluation based on Crisp and Fuzzy methodologies. At the same time, details will be hunted and filtered by their demands and sorted them accordingly. Arrived results were obtained using variance, chi square, ANOVA and Theil indexing. As a consequence the customers will be able to yield commodity without spending more time and effort in visiting numerous number of sites.Item Workflow Scheduling In Clouds Using Randomized Scheduling Algorithm(IJPAM, 2018) Jangiti, SaikishorThe provisioning of on-demand resources makes it optimal for executing scientific application workflows in cloud computing. An application starts the process with a small number of resources, and it allocates the resources when required. However, workflow scheduling belongs to NP-hard class of problems, so optimization techniques are preferred for the solution. This paper explores the effect of a Randomized scheduling algorithm in workflow scheduling for the scheduling problem. The use of Randomized scheduling algorithm in comparison with other scheduling algorithms increases the efficiency of workflow scheduling in various scientific workflows and simulators. The experimental result confirms that the Randomized scheduling algorithm well performed than other scheduling approaches and provides better scheduling with reduced makespan.Item Ensemble Gaussian mixture model-based special voice command cognitive computing intelligent system(IOS, 2020-12) Jangiti, SaikishorDysarthria is a speech disorder caused by stroke, Parkinson’s disease, neurological injury, or tumors that damage the nervous system and weaken the speech quality. Developing a unique voice command system for Dysarthric speech helps to recognize impaired speech and convert them into text or input commands. Hidden Markov Model (HMM) is one of the widely used generative model-based classifiers for Dysarthric speech recognition. But due to insufficient training data, HMM doesn’t provide optimal results on overlapping classes. We propose an ensemble Gaussian mixture model to recognize impaired speech more accurately. Our model converts the sequence of feature vectors into a fixed dimensional representation of patterns with varying lengths. The performance efficiency of the proposed model is evaluated on the Dysarthric UA-speech benchmark dataset. The discriminatory information provided by the proposed approach yields better classification accuracy even for shallow intelligibility words compared to conventional HMM.Item Automated question extraction and tagging for cloud-based online communities(Inder Science, 2019) Jangiti, SaikishorCrowd-based question answering forums and cloud-based community question answering platforms provide us with the dais to post questions and answers online. This helps the users to get desired answers from expert users. It is a challenge for a person with mobility needs to go out and explore. The 'wheelchair accessible' information provided by Google Maps is useful to explore before going out. Local guides share this knowledge on Google Maps by answering quick questionnaire. Automated question generation is a key challenge that we face with regard to natural languages in the context of users visited locations, already reviewed places, likes, interests and user experience. In this paper, we have implemented an automatic question generation system that comprises of part-of-speech (POS) tagger, text-to-question generation task using syntactic analysis and a named entity extraction. The proposed system is tested with human effort and is generating valid questionnaireItem Bulk-bin-packing based migration management of reserved virtual machine requests for green cloud computing(European Alliance for Innovation, 2019) Jangiti, SaikishorThe dynamic consolidation of Virtual Machines (VMs) into a minimum number of Physical Machines (PMs) is a key energy-efficient practice in a cloud data centre, to reduce the running PMs and save electricity costs. We proposed a migration based VM consolidation approach for reserved requests. Real Dataset EC2 was used in the simulation experiments. The proposed BBPMM has demonstrated the elastic capability of adjusting the running PMs and it reduced 38% of running PMs in a reservation transition period.Item Resource ratio based virtual machine placement in heterogeneous cloud data centres(Springer, 2019-11) Jangiti, SaikishorServer consolidation through virtualization improves resource utilization significantly in Cloud Data Centres (CDCs). We study the case of a CDC hosting heterogeneous Physical Machines (PMs) as a variable size vector bin-packing problem. The PMs have different configurations of multiple resources like CPU, RAM, Disk Storage and Network Bandwidth. In this paper, we propose PMNeAR-vector heuristic for PM selection in PM-heterogeneity aware Virtual Machine (VM) initial placement. The proposed heuristic is compared with well-known heterogeneity aware FFD-DRR and BFD bin centric heuristics using a dataset with random instances of both VMs and PMs of heterogeneous configurations. Fifty rounds of VM initial placement simulation experiments were conducted to validate the average resource wastage. The results show that on average FFD-DRR and BFD bin centric heuristics are wasting 22.62% and 37.27% more resource units compared to the proposed PMNeAR-vector heuristic.Item Hybrid Best-Fit Heuristic for Energy Efficient Virtual Machine Placement in Cloud Data Centers(EUDL, 2020) Jangiti, SaikishorCloud Service Providers (CSPs) offers Information Technology services like infrastructure and software to users on a pay as you go basis. Energy consumption is one of the significant challenges faced by Cloud Service Providers (CSP). Virtual Machine (VM) placement is an energy-efficient practice performed in the cloud datacenters. Best-Fit Decreasing (BFD) is a VM placement and is known to give a near-optimal solution in a reasonable time by sorting the VMs in decreasing order. We propose a Hybrid Best-Fit (HBF) Heuristic for VM placements. Experimental results show that HBF is consuming 2.516% and 3.392% less energy compared to Best-Fit and BFD heuristics.Item The role of cloud computing infrastructure elasticity in energy efficient management of datacenters(IEEE, 2017) Jangiti, SaikishorCloud Computing is growing its customer base due to its pay per use model of leased computational resources as well as software services. Along with the rapid growth of cloud computing adoption, the energy consumed by cloud computing infrastructure is growing. There is an urgent need for the research on the energy efficiency of cloud computing infrastructure. We reviewed the key technologies and techniques which will support and enhance the energy efficiency of cloud computing infrastructure and makes cloud a sustainable model. Virtualization, elasticity, and energy-efficiency are three important attributes of cloud and its infrastructure; we studied the interdependencies of these techniques and addressed few questions related to their interdependencies.