HARSH TIWARI

This project focuses on Sentiment Analysis using Natural Language Processing (NLP) and Long Short-Term Memory (LSTM) networks. The text data is preprocessed through cleaning, tokenization, stop-word removal, and sequence padding before being fed into the LSTM model. The model helps automate opinion mining and can be applied to customer feedback, product reviews, and social media analysis.

Designed and developed interactive Power BI dashboards to analyze UPI transaction

trends, volumes, and user behavior. Implemented ETL workflows for cleaning, transforming, and modeling raw transaction data into

actionable insights. Delivered visual analytics highlighting peak usage patterns, fraud detection indicators, and business growth

opportunities.

Conducted an end-to-end data analytics project analyzing student

depression and mental health trends using a structured dataset. Used SQL to perform data cleaning, filtering, and transformation,

including handling missing values, creating derived columns, and applying aggregate functions (COUNT, AVG, GROUP BY, HAVING) to

identify key patterns. Designed optimized SQL queries to analyze relationships between academic pressure, sleep duration, lifestyle

habits, gender, and depression levels. Extracted insights such as high-risk student groups, contributing factors to depression, and

correlations between study load and mental well-being. Imported cleaned data into Tableau to create interactive dashboards and

visualizations. Developed an interactive Tableau dashboard with filters (gender, age group, academic year, stress level) to allow dynamic

exploration of data.