All posts by Karthick R
ML-POWERED UPI FRAUD DETECTION
ML-POWERED UPI FRAUD ML-POWERED UPI FRAUD XGBoost (Extreme Gradient Boosting) is chosen for its efficiency, scalability, and ability tohandle large datasets while preventing overfitting. The model is designed to analyzetransactional patterns, detectRead More…
CARDIAC ARRYTHMIA DETECTION USING AI/ML
CARDIAC ARRYTHMIA DETECTION USING AI/ML CARDIAC ARRYTHMIA DETECTION USING AI/ML Historically, the detection of arrhythmia relied on human clinicians interpretingelectrocardiograms (ECGs) themselves. Though effective, this practice is labor-intensive,susceptible to inter-observer variation, andRead More…
Smart Environmental Sensing: Outdoor Pollution Detection Using IoT and Cloud-Based Predictive Models
Smart Environmental Sensing: Outdoor Smart Environmental Sensing: Outdoor Extensive field testing was conducted across multiple urban zones with varying pollution profiles tovalidate the system’s effectiveness. The IoT sensors demonstrated high sensitivity andRead More…
Next-Gen Outdoor Air Pollution Monitoring Using Smart IoT Sensors and Deep Learning Models
Next-Gen Outdoor Air Pollution Next-Gen Outdoor Air Pollution To overcome these obstacles, integrating IoT sensor networks with advanced data processingtechniques, particularly deep learning, has emerged as a cutting-edge approach. Deep learningmodels haveRead More…
Transformer-Enhanced Text Mining for Scalable Topic Modeling in Big Data Environments
Transformer-Enhanced Text Mining Transformer-Enhanced Text Mining A primary obstacle in utilizing transformers directly for topic modeling is their computationalintensity and the difficulty in interpreting the resulting embeddings in terms of explicit topics.Additionally,Read More…
Zero-Shot Text Extraction and Topic Modeling with Prompt-Driven Large Language Models
Zero-Shot Text Extraction Zero-Shot Text Extraction, in this context, refers to a model’s ability to perform a task it has not beenexplicitly trained on, simply by interpreting instructions embedded in the prompt.Read More…
Multimodal Text Mining and Aspect-Oriented Topic Modeling Using Foundation Models
Multimodal Text Mining and Aspect The advent of deep learning and particularly the rise of transformer-based models haverevolutionized natural language processing (NLP). Embedding techniques, which encode words,sentences, or entire documents as denseRead More…
Unsupervised Topic Discovery and Text Extraction with Generative AI and Embedding Techniques
Unsupervised Topic Discovery and Text With the emergence of neural networks and representation learning, embedding techniquesrevolutionized textual analysis. Word2Vec [Mikolov et al., 2013] introduced dense wordembeddings that preserved semantic relationships, enabling wordsRead More…
End-to-End Information Extraction and Dynamic Topic Modeling Using Deep NLP Pipelines
End-to-End Information Extraction End-to-End Information Extraction In today’s data-driven world, the proliferation of unstructured textual data presents bothtremendous opportunities and significant challenges. Vast amounts of text are continuouslygenerated from diverse sources suchRead More…