Yogesh Singh Kharkwal

Sentiment Analysis Model:

• Developed a sentiment analysis model to classify text reviews as positive or negative using Recurrent Neural Networks (RNN).

• Implemented text preprocessing techniques including tokenization, padding, and sequence encoding to prepare data for model training.

• Designed and trained an RNN-based architecture for effective sequence modeling and sentiment classification.

• Improved model performance through hyperparameter tuning and optimization techniques.

Text Summarizer App:

Developed an end-to-end text summarization application using a fine-tuned pre-trained T5 Small Transformer model from the Hugging Face platform to generate concise and meaningful summaries from large text inputs

• Built a scalable backend using FastAPI for efficient model inference and API handling.

• Designed a responsive and user-friendly frontend using HTML, CSS, and JavaScript.

• Implemented data preprocessing and cleaning pipelines using Python and Pandas.

• Integrated PyTorch for model training, evaluation, and deployment.

Technologies Used:

• Python, Pandas, PyTorch, Hugging Face (T5 Pre-trained Transformer), FastAPI, HTML, CSS, JavaScript

smart card costomer insights

• Performed exploratory data analysis on smart card customer transaction data

• to identify usage patterns and behavioral insights. Implemented data cleaning, visualization, and statistical analysis to understand customer segmentation and trends.

• Genrate Final Output and summary