Amazon & YouTube Data Tools

This repository contains two separate tools:

1. Amazon Scraper - A Selenium-based scraper for extracting sponsored product data from Amazon.
2. YouTube Video Analyzer - A Streamlit application that finds and recommends the most relevant YouTube videos based on a search query.

Amazon Scraper

Overview

The Amazon Scraper tool uses Selenium to scrape sponsored product information from Amazon's search results. It specifically targets the "soft toys" category and extracts details like:

• Product title
• Brand name
• Price
• Rating
• Number of reviews
• Product image URL
• Product page URL

Files

• amazon_sponsored_softtoys.csv - CSV file containing the scraped data
• app.py - The main Python script for scraping Amazon
• analysis.ipynb - Jupyter notebook for analyzing the scraped data

Setup & Installation

1. Install the required packages:

   pip install selenium pandas webdriver-manager

2. Run the scraper:

   cd amazon-scraper
   python app.py

Features

• Headless browser operation for faster scraping
• Detailed product information extraction
• Brand detection from product pages
• CSV output for easy data analysis

YouTube Video Analyzer

Overview

The YouTube Video Analyzer is a Streamlit application that helps users find the most relevant YouTube videos based on their search query. It uses:

• YouTube Data API for searching videos
• Google Speech Recognition for voice input
• Google's Gemini AI for analyzing video relevancy

Files

• app.py - Main Streamlit application
• requirements.txt - Required Python packages
• .env - Environment variables for API keys (not included in repository)

Setup & Installation

1. Install the required packages:

   cd youtube-assignment
   pip install -r requirements.txt

2. Create a .env file with your API keys:

   YOUTUBE_API_KEY=your_youtube_api_key
   GOOGLE_API_KEY=your_google_api_key
   

3. Run the Streamlit application:

   streamlit run app.py

Features

• Search for YouTube videos published in the last 14 days
• Filter videos by duration (4-20 minutes)
• Voice input for search queries
• AI-powered video relevance ranking
• Embedded video player for the most relevant result

Requirements

Amazon Scraper

• Python 3.7+
• Selenium
• Pandas
• Webdriver-manager

YouTube Analyzer

• Python 3.7+
• Streamlit
• Google API Python Client
• Google Generative AI
• Speech Recognition
• PyAudio
• isodate