Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506
Browse
10 results
Search Results
Item Investigating the "wisdom of crowds" at scale(Association for Computing Machinery, Inc acmhelp@acm.org, 2015) Mysore, A.S.; Yaligar, V.S.; Ibarra, I.A.; Simoiu, C.; Goel, S.; Arvind, R.; Sumanth, C.; Srikantan, A.; Bhargav, H.S.; Pahadia, M.; Dobhal, T.; Ahmed, A.; Shankar, M.; Agarwal, H.; Agarwal, R.; Anirudh-Kondaveeti, S.; Arun-Gokhale, S.; Attri, A.; Chandra, A.; Chilukuri, Y.; Dharmaji, S.; Garg, D.; Gupta, N.; Gupta, P.; Jacob, G.M.; Jain, S.; Joshi, S.; Khajuria, T.; Khillan, S.; Konam, S.; Kumar-Kolla, P.; Loomba, S.; Madan, R.; Maharaja, A.; Mathur, V.; Munshi, B.; Nawazish, M.; Neehar-Kurukunda, V.; Nirmal-Gavarraju, V.; Parashar, S.; Parikh, H.; Paritala, A.; Patil, A.; Phatak, R.; Pradhan, M.; Ravichander, A.; Sangeeth, K.; Sankaranarayanan, S.; Sehgal, V.; Sheshan, A.; Shibiraj, S.; Singh, A.; Singh, A.; Sinha, P.; Soni, P.; Thomas, B.; Tuteja, L.; Varma-Dattada, K.; Venkataraman, S.; Verma, P.; Yelurwar, I.In a variety of problem domains, it has been observed that the aggregate opinions of groups are often more accurate than those of the constituent individuals, a phenomenon that has been termed the "wisdom of the crowd." Yet, perhaps surprisingly, there is still little consensus on how generally the phenomenon holds, how best to aggregate crowd judgements, and how social influence affects estimates. We investigate these questions by taking a meta wisdom of crowds approach. With a distributed team of over 100 student researchers across 17 institutions in the United States and India, we develop a large-scale online experiment to systematically study the wisdom of crowds effect for 1,000 different tasks in 50 subject domains. These tasks involve various types of knowledge (e.g., explicit knowledge, tacit knowledge, and prediction), question formats (e.g., multiple choice and point estimation), and inputs (e.g., text, audio, and video). To examine the effect of social influence, participants are randomly assigned to one of three different experiment conditions in which they see varying degrees of information on the responses of others. In this ongoing project, we are now preparing to recruit participants via Amazon's Mechanical Turk.Item An index coded approach for reducing number of broadcasts in vehicular networks(Institute of Electrical and Electronics Engineers Inc., 2017) Patil, A.; Shekar, N.; Shet, N.S.V.Vehicular network standard demands a short range reliable vehicle-to-vehicle (V2V) communication with high mobility, also increasing amount of downloaded data in short encounters. Satisfying demands of clients with minimum number of transmissions using index coding has proven notable significance in wireless ad-hoc networks. The proper encoding at the server with side information leads to simultaneous satisfaction of clients with reduced amount of broadcasts. MERLIN (Maximum Expected download over Random LINks) is one of the fundamental protocols developed for optimizing downloads in vehicular networks over random encounter durations [1]. In this paper, we present I-MERLIN (Index coding based MERLIN) protocol and analyze its performance over MERLIN. We show that same amount of data can be downloaded from server in reduced number of transmissions, for single server multiple clients scenario. We also show that, I-MERLIN works well to satisfy multiple clients in minimum number of requests, which can be applicable for real road scenarios. © 2017 IEEE.Item Clustering and cooperative data transfer in VANETs using index coding(Inderscience Publishers, 2018) Patil, A.; Shet, N.S.V.Along with the prime aim of providing on-road safety, vehicular ad hoc networks have enabled the infotainment applications due to increasing demands from various travellers. The high mobility of nodes demands fast and efficient broadcasting of data. In this paper, we use index coding with clustering and cooperative downloading to satisfy various demands of travelling vehicles in reduced number of transmissions. The proposed index coding based protocol achieves better download throughput through V2V data exchange between clusters. Simulation results show that the proposed design achieves higher throughput than previously defined IEEE 1609.4 and Vehicular Enhanced Multichannel MAC (VEMMAC) protocols by reducing the total number of transmissions at the server. © © 2018 Inderscience Enterprises Ltd.Item Challenges for Decentralized Congestion Control Mechanisms in Vehicular Ad Hoc Networks(Institute of Electrical and Electronics Engineers Inc., 2018) Patil, A.; Shet, N.S.V.Vehicular Ad Hoc Networking (VANET) is an emerging technology defined with main motive of avoiding on road accidents and provide safety and life critical applications to users in a well-organized manner. We present current challenges in designing VANET protocols through extensive survey from the perspective of decentralized congestion control (DCC) strategies. The challenges in routing and security applications are also outlined briefly. Internet of things (IoT) will be the future networking system in which all things will be connected to interact. In the near future, VANETs will be the main part of IoT networks because of its safety and infotainment applications. A lot of work is done in this area, still, the industry thrusts for more efficient and robust design. This paper highlights various research challenges, which will help other researchers who wish to work in this area. © 2018 IEEE.Item Transmit Data Rate Control Based Decentralized Congestion Control Mechanism for VANETs(Institute of Electrical and Electronics Engineers Inc., 2019) Patil, A.; Deeksha, M.; Shekar, N.; Shet, N.S.V.; Kulkarni, M.Vehicular Ad Hoc Networking (VANET)technology is becoming prominent in this era. It is defined to improve on-road conditions and provide safety and different entertainment applications to end users in orderly and efficient manner. The idea of making everything connected in Internet of things (IoT)has evolved as a promising networking system and VANETs are one of the components of it. In this paper, we present a multistate active transmit data rate control based decentralized congestion control mechanism. The real world scenario generated from SUMO is considered for simulation. Simulation results show that, data rate control mechanism performs better over transmit power control and without adaptation mechanisms. © 2019 IEEE.Item Effect of ECAP on sliding wear behaviour of Mg-Zn-Gd-Zr alloy(Elsevier Ltd, 2020) Patil, A.; Bontha, S.; Ramesh, M.R.Magnesium is a lightweight, recyclable, and biocompatible material. However, the extensive commercial use of Magnesium and its alloys is hindered by their poor wear behaviour and mechanical properties. Equal Channel Angular Pressing (ECAP) is a severe plastic deformation technique which improves the material properties through grain refinement. In the present study, wear behaviour of ECAP processed Mg-Zn-Gd-Zr alloy was investigated. ECAP process was carried out up to 3 passes at a temperature of 380 °C. Wear testing of as-cast and ECAP processed alloy were carried out using dry sliding wear method on a pin on disk tribometer by varying loads. The wear mechanism was analysed using Scanning Electron Microscope (SEM) and Energy Dispersive X-ray Spectroscopy (EDS). Average Coefficient of Friction (COF) increased after each pass of ECAP. Wear rate increased with the applied load. Despite severe plastic deformation, wear resistance of ECAP processed samples was found to be lower than that of as-cast samples at higher loads. Abrasive and oxidation wear mechanisms were found in both as-cast and ECAP processed samples. © 2019 Elsevier Ltd. All rights reserved.Item Multi-device Login Monitoring for Google Meet Using Path Compressed Double-Trie and User Location(Springer Science and Business Media Deutschland GmbH, 2021) Patil, A.; Singh, A.; Chauhan, N.Google Meet, much like other online video-conferencing platforms, has seen a surge in popularity for its reliability and convenience, which also makes it necessary to be analysed for vulnerabilities owing to its large user-base. This paper focuses on an observational study on Google Meet to find certain shortcomings, specifically the freedom users have to login through multiple devices and how it can aid attackers in certain scenarios. A simulated back-end architecture is developed to propose a proof of concept on tackling the explored issues using a path compressed double-trie structure and the location of user from his public IP address. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.Item Attentiveness monitoring and user record maintenance in virtual classrooms(Institute of Electrical and Electronics Engineers Inc., 2021) Patil, A.; Singh, A.; Chauhan, N.Virtual classrooms rely on video conferencing tools like Google Meet, Microsoft Teams or Zoom, to carry out lecture sessions. Although previously it was an additional tool, the pandemic scenario made it the primary mode of taking classes. It is observed that lecturers find it hard to monitor and look after every student in such a scenario due to divided webcam feeds, simultaneous presentation and replying on chats. An automated monitoring system can aid in such a scenario where the user's local system can be used for localized calculation of attentiveness using face and eye detection. A simulated back-end architecture is developed to propose a proof-of-concept for the same using a path compressed master-Trie structure. © 2021 IEEE.Item Multistate Active Combined Power and Message/Data Rate Adaptive Decentralized Congestion Control Mechanisms for Vehicular Ad Hoc Networks(IOP Publishing Ltd, 2022) Deeksha, M.; Patil, A.; Kulkarni, M.; Shet, N.V.S.; Muthuchidambaranathan, P.Vehicular ad hoc networks (VANETs) have emerged in time to reduce on-road fatalities and provide efficient information exchange for entertainment-related applications to users in a well-organized manner. VANETs are the most instrumental elements in the Internet of Things (IoT). The objective lies in connecting every vehicle to every other vehicle to improve the user's quality of life. This aim of continuous connectivity and information exchange leads to the generation of more information in the medium, which could congest the medium to a larger extent. Decentralized congestion control (DCC) techniques are specified to reduce medium congestion and provide various safety applications. This article presents two DCC mechanisms that adapt message rate and data rate combined with transmit power control mechanism. These mechanisms are developed under multi-state active design proposed by the standard. The proposed methods deliver better performance over other mechanisms in terms of power, channel load, and channel utilization using real-time-based scenarios by simulation in SUMO. © 2022 Institute of Physics Publishing. All rights reserved.Item Hardware based Analysis of Deep Neural Networks(Institute of Electrical and Electronics Engineers Inc., 2023) Karia, A.; Patil, A.; Apoorva, M.K.; Varsha, P.; Pillay, S.; Mohan, B.R.The advent of the Machine Learning (ML) era is now evident and its inculcation into early education requires students to have feasible options to work with the field. The proposed comparative analysis tests different frameworks namely sci-kit with Keras, Tensor-flow and PyTorch on various available processors and discussed further using a range of standard metrics to evaluate the model and well as the underlying hardware that they will run upon. Based on the study, the most viable combination of framework and hardware for educational purpose shall be found out. © 2023 IEEE.