A5070248228- Data Mining Algorithms and Applications
- Data Stream Mining Techniques
- Topic Modeling
- Data Management and Algorithms
- Rough Sets and Fuzzy Logic
- Advanced Clustering Algorithms Research
- Advanced Text Analysis Techniques
- Complex Network Analysis Techniques
- Natural Language Processing Techniques
- Machine Learning and Data Classification
- Face and Expression Recognition
- Big Data and Business Intelligence
- Time Series Analysis and Forecasting
- Opinion Dynamics and Social Influence
- Cloud Computing and Resource Management
- COVID-19 epidemiological studies
- Advanced Database Systems and Queries
- Imbalanced Data Classification Techniques
- Caching and Content Delivery
- Semantic Web and Ontologies
- Information Retrieval and Search Behavior
- Machine Learning in Bioinformatics
- Anomaly Detection Techniques and Applications
- Social Capital and Networks
- Cloud Data Security Solutions
University of Delhi
2015-2024
South Asian University
2013-2014
We report results from the Supernova Photometric Classification Challenge (SNPCC), a publicly released mix of simulated supernovae (SNe), with types (Ia, Ibc, and II) selected in proportion to their expected rate. The simulation was realized griz filters Dark Energy Survey (DES) realistic observing conditions (sky noise, point-spread function atmospheric transparency) based on years recorded at DES site. Simulations non-Ia type SNe are spectroscopically confirmed light curves that include...
The CODATA Data Science Journal is a peer-reviewed, open access, electronic journal, publishing papers on the management, dissemination, use and reuse of research data databases across all domains, including science, technology, humanities arts. scope journal includes descriptions systems, their implementations publication, applications, infrastructures, software, legal, reproducibility transparency issues, availability usability complex datasets, with particular focus principles,...
We present a parallel version of BIRCH with the objective enhancing scalability without compromising on quality clustering. The incoming data is distributed in cyclic manner (or block if bursty) to balance load among processors. algorithm implemented message passing share-nothing model. Experiments show that for very large sets scales nearly linearly increasing number also clusters obtained by PBIRCH are comparable those using
The emotion of fear related to an infectious disease not only influences individual's behavior but also transmits social contacts. Therefore, modeling human is a precursor reliable estimates epidemic size and duration. In this paper, we present abstract model fear, which realized using Individual-based Fear Model (IBFM). model, coupled with contagion study the influence on dynamics. Since inherent characteristic individual that determines susceptibility infection, discerns between...
Clustering of data streams finds important applications in tracking evolution various phenomena medical, meteorological, astrophysical, seismic studies. Algorithms designed for this purpose are capable adapting the discovered clustering model to changes characteristics but not user’s requirements themselves. Based on previous observation, we perform a comparative study different approaches existing stream algorithms and present parameterized architectural framework that exploits nuances...
Generative language models, such as ChatGPT, have garnered attention for their ability to generate human-like writing in various fields, including academic research. The rapid proliferation of generated texts has bolstered the need automatic identification uphold transparency and trust information. However, these closely resemble human often subtle differences grammatical structure, tones, patterns, which makes systematic scrutinization challenging. In this work, we attempt detect Abstracts...
Knowledge infusion is a promising method for enhancing Large Language Models domain-specific NLP tasks rather than pre-training models over large data from scratch. These augmented LLMs typically depend on additional or knowledge prompts an existing graph, which impractical in many applications. In contrast, directly relevant documents more generalisable and alleviates the need structured graphs while also being useful entities that are usually not found any graph. With this motivation, we...