Prashant Verma's Projects
Code repository for automating company data extraction from the Ministry of Corporate Affairs site using Python, Selenium, and Pandas. Results are saved locally in CSV format.
A conversational chatbot designed for currency conversion tasks using Flask, Python, and Dialogflow. Integrating APIs for efficient conversions, users can specify currencies and values effortlessly.
This project utilizes unsupervised machine learning to segment bank customers for targeted marketing campaigns. It covers tasks like data exploration, determining optimal clusters, and applying k-means for segmentation. Ideal for marketing departments in banking and retail industries.
These dashboards provide insights across diverse domains, including cryptocurrency sales, workforce challenges, disease impact analysis, and retail trends. Leveraging tools like Power BI and Excel, they offer actionable insights for decision-making.
This Django blog project allows users to publish and manage blogs after registration and login. Features include user profile updates, blog search, and comment management. The application provides a simple and intuitive platform for blogging.
This project is a streamlined Streamlit web app for easy image scraping from Google Images. Enter your search query, fetch, and download images locally in a zip file. Simple setup and customization for tailored results.
Student Performance Analysis with Machine Learning analyzes factors impacting student outcomes using a robust machine learning pipeline. Achieving an impressive R2 score, it predicts student performance effectively. With extensive data preprocessing and deployment on AWS Elastic Beanstalk, it ensures scalability and high availability.
This Python project utilizes MediaPipe to recognize hand landmarks in images, videos, and webcam streams. It detects and locates 21 key points on the hand, offering a simple and efficient solution for various applications requiring hand gesture analysis.
Utilizing the House Prices Dataset , this project predicts home prices through a Jupyter notebook-based data science pipeline. It includes exploratory data analysis, cleaning, feature engineering, and modeling. The project explores diverse aspects of residential homes to understand price influences beyond traditional factors.
A conversational chatbot powered by OpenAI's Large Language Model (LLM) and built using Streamlit for interactive user interactions. The chatbot utilizes advanced natural language processing models and techniques for dynamic message handling and real-time response generation.
This recommendation system employs content-based filtering and NLP preprocessing to suggest similar movies based on user preferences and movie data. It fetches movie posters via APIs and is deployed on Streamlit for easy access.
This is a Streamlit web app leveraging EasyOCR to extract text from uploaded images, presenting it with confidence scores and visual annotations, simplifying Optical Character Recognition tasks.
Profile's main README.md resource.
This repository contains the source code of my personal portfolio, showcasing my skills, projects, and experiences.
This tool predicts data science salaries by scraping Glassdoor job descriptions. Features quantify company valuation of key skills. Regression models include Multiple Linear Regression, Lasso Regression, and Random Forest. Random Forest outperformed others. Technologies/libraries used: Python, Pandas, NumPy, Scikit-learn, Seaborn, Matplotlib.
SQL analysis followed by a Tableau dashboard creation was used to uncover the reasons behind declining sales.
Using Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM), this project predicts Nestle India stock prices. The dataset spans from May 18, 2018 to May 20, 2022, and results are visualized using matplotlib.
This project is a Streamlit application powered by OpenAI's DALL-E, offering seamless text-to-image generation. Users can input textual prompts to generate high-quality images, while rate limiting ensures controlled API usage. Visit the live demo for a firsthand experience.
This project tackles the Titanic challenge on Kaggle, predicting passenger survival based on variables like age, sex, and passenger class. The Jupyter notebook covers essential steps of a data science pipeline, including exploratory data analysis, data cleaning, feature engineering, and modeling. The dataset used is the Titanic dataset.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent