A5016329881- Software Engineering Research
- Software Reliability and Analysis Research
- Topic Modeling
- Adversarial Robustness in Machine Learning
- Artificial Intelligence in Healthcare
- Advanced Clustering Algorithms Research
- Natural Language Processing Techniques
- Explainable Artificial Intelligence (XAI)
- Advanced Malware Detection Techniques
- Sentiment Analysis and Opinion Mining
- Neural Networks and Applications
- Cerebral Palsy and Movement Disorders
- Robotics and Automated Systems
- Teaching and Learning Programming
- Software Testing and Debugging Techniques
- Aeroelasticity and Vibration Control
- Advanced Software Engineering Methodologies
- Software System Performance and Reliability
- Face and Expression Recognition
- Imbalanced Data Classification Techniques
- Intelligent Tutoring Systems and Adaptive Learning
- Emotion and Mood Recognition
- Privacy-Preserving Technologies in Data
- Anomaly Detection Techniques and Applications
- Expert finding and Q&A systems
Birla Institute of Technology, Mesra
2020-2024
Indian Institute of Technology Roorkee
2024
SRM Institute of Science and Technology
2021
Guru Gobind Singh Indraprastha University
2020
Twitter (United States)
2020
Carnegie Mellon University
2019
Graphic Era University
2017
Bryn Mawr College
2008-2009
Atlanta Technical College
2008
Georgia Institute of Technology
2008
Attention mechanisms are ubiquitous components in neural architectures applied to natural language processing. In addition yielding gains predictive accuracy, attention weights often claimed confer interpretability, purportedly useful both for providing insights practitioners and explaining why a model makes its decisions stakeholders. We call the latter use of into question by demonstrating simple method training models produce deceptive masks. Our diminishes total weight assigned...
Every day, thousands of customers post questions on Amazon product pages. After some time, if they are fortunate, a knowledgeable customer might answer their question. Observing that many can be answered based upon the available reviews, we propose task review-based QA. Given corpus reviews and question, QA system synthesizes an answer. To this end, introduce new dataset method combines informational retrieval techniques for selecting relevant (given question) "reading comprehension" models...
An exciting new initiative at Georgia Tech and Bryn Mawr College is using personal robots both to motivate students serve as the primary programming platform for Computer Science 1 curriculum. Here, authors introduce outline plans future.
Active flutter suppression for a two-dimensional typical airfoil in an incompressible flow is studied this paper. The root loci of the systems described by three aerodynamical models, each with single feedback variable, are first investigated to obtain physical understanding these systems. Practical that include actuator and gust models considered next, using linear-quadratic regulator theory. Reduced-order constructed sequence truncations based on modal cost analysis, truncation examined...
The software engineering community is working to develop reliable metrics improve quality. It estimated that understanding the source code accounts for 60% of maintenance effort. Cognitive informatics important in quantifying degree difficulty or efforts made by developers understand code. Several empirical studies were conducted 2003 assign cognitive weights each possible basic control structure software, and these are used several researchers evaluate complexity systems. In this paper, an...
Deep neural network models built by the appropriate design decisions are crucial to obtain desired classifier performance. This is especially when predicting fault proneness of software modules. When correctly identified, this could help in reducing testing cost directing efforts more towards modules identified be prone. To able build an efficient deep model, it important that parameters such as number hidden layers, nodes each layer, and training details learning rate regularization methods...
This paper is presented to improve the eHealth care platform by using Robots which are connected through IoT for providing personalized multiple-care approach particularly a diabetic patient. The architecture of this model such that it makes use Internet Things Disease Management Hub web-centric then wirelessly robot. robot has chain capillary networks in contains medicals sensors and sensor dietary monitoring patient thus provides them with fully multidimensional-care. software built so...
Clustering is a crucial at the same time challenging task in several application domains. It important to incorporate optimum feature finding into our clustering algorithms for getting better prediction accuracy but this difficult when there no or little information about importance relevance of features. To tackle an efficient manner we employ natural evolution process inherent genetic (GA) find features healthy aging dataset. In order empirically verify findings, were combined with number...
Zhengzhong Liu, Guanxiong Ding, Avinash Bukkittu, Mansi Gupta, Pengzhi Gao, Atif Ahmed, Shikun Zhang, Xin Swapnil Singhavi, Linwei Li, Wei Wei, Zecong Hu, Haoran Shi, Xiaodan Liang, Teruko Mitamura, Eric Xing, Zhiting Hu. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. 2020.
The convergence of various multimedia communications that includes voice, video and data presents many opportunities for enabling unified communication but paradoxically leads to inefficiencies the user as may become complex. Model driven technologies such Communication Virtual Machine (CVM) propose reduce complexity through use models define a user's needs. CVM was extended utilize multiple common-off-the-self (COTS) APIs Skype Smack which we refer frameworks, promote service network...
Integration of emotions to develop an integrated emotionally intelligent system is a tricky task. Human face complex entity which displays variety facial expressions. In this research study, Ensemble model using hybridization angles and distances for emotion recognition (HADER) proposed. Proposed HADER forms feature set containing total 37 calculated fiducial points. Hybridized submitted ensemble predict the intensity various emotions. validated on CK+ KDEF datasets. Evaluation shows...
In this evolving era of machine learning security, membership inference attacks have emerged as a potent threat to the confidentiality sensitive data. attack, adversaries aim determine whether particular point was used during training target model. This paper proposes new method gauge data point’s in model’s set. Instead correlating loss with membership, is traditionally done, we leveraged fact that examples generally exhibit higher confidence values when classified into their actual class....
Clustering is a crucial and, at the same time, challenging task in several application domains. It important to incorporate optimum feature finding into our clustering algorithms for better exploration of features and draw meaningful conclusions, but this difficult when there no or little information about importance relevance features. To tackle an efficient manner, we employ natural evolution process inherent genetic (GAs) find healthy aging dataset. empirically verify findings, were...
Vision-Language Models (VLMs) have transformed tasks requiring visual and reasoning abilities, such as image retrieval Visual Question Answering (VQA). Despite their success, VLMs face significant challenges with involving geometric reasoning, algebraic problem-solving, counting. These limitations stem from difficulties effectively integrating multiple modalities accurately interpreting geometry-related tasks. Various works claim that introducing a captioning pipeline before VQA enhances...
Autism is one of the developmental disorders which make major behavioral and communication changes in individual induce a drawback walking pattern. The main aim this study to analyze compare stride patterns an autistic patient healthy individual. With help proposed device, yaw, pitch roll limbs can be easily measured analyzed. These parameters will identify lacking movement physiotherapists improving pattern patients by adopting correct approach remedial actions/treatment. It also used...
In this evolving era of machine learning security, membership inference attacks have emerged as a potent threat to the confidentiality sensitive data. attack, adversaries aim determine whether particular point was used during training target model. This paper proposes new method gauge data point's in model's set. Instead correlating loss with membership, is traditionally done, we leveraged fact that examples generally exhibit higher confidence values when classified into their actual class....