ReviewInsight Core

AI-powered restaurant review analysis engine using Claude Sonnet 4 with prompt caching.

Features

• 🤖 Claude Sonnet 4 analysis with 90% cost reduction via prompt caching
• 🌍 Google Maps review scraping via Outscraper
• 📊 Excel export with multiple sheets (Scores + Keywords)
• 🔄 Resumable analysis with local caching
• ⚡ Parallel processing (configurable workers)
• 💰 Real-time cost tracking

Quick Start

1. Install dependencies

pip install -r requirements.txt

2. Configure environment

cp .env.example .env
# Edit .env with your API keys

3. Run analysis

From CSV:

python cli/main.py analyze --input data/input/reviews.csv --output data/output/

From Google Maps:

python cli/main.py scrape --name "McDonald's Paris Nord" --location "Paris, France" --max-reviews 2000

With competitors:

python cli/main.py scrape --name "McDonald's Paris Nord" \
  --competitors "Burger King Paris" "KFC Paris" \
  --location "Paris, France" \
  --max-reviews 2000

Configuration

Edit config/settings.py or set environment variables:

ANTHROPIC_API_KEY=sk-ant-...
OUTSCRAPER_API_KEY=...
MAX_WORKERS=10
MAX_TOKENS=1000

Cost Estimation

python cli/main.py estimate --reviews 2000

Output:

Estimated costs for 2,000 reviews:
  Outscraper:  $3.00
  Claude API:  $8.00 (with caching)
  Total:       $11.00

Project Structure

• src/analyzers/ - AI analysis logic
• src/scrapers/ - Data collection (Outscraper, CSV)
• src/processors/ - Orchestration & caching
• src/exporters/ - Excel/JSON export
• cli/ - Command-line interface
• tests/ - Unit tests

API Usage

from src.scrapers.outscraper_scraper import OutscraperScraper
from src.analyzers.anthropic_analyzer import AnthropicAnalyzer
from src.processors.orchestrator import Orchestrator
from src.exporters.excel_exporter import ExcelExporter

# 1. Scrape reviews
scraper = OutscraperScraper(api_key="...")
reviews = scraper.scrape("McDonald's Paris", max_reviews=2000)

# 2. Analyze
analyzer = AnthropicAnalyzer(api_key="...")
orchestrator = Orchestrator(analyzer)
results = orchestrator.analyze(reviews, max_workers=10)

# 3. Export
ExcelExporter.export(results, "output.xlsx")

Development

Run tests:

pytest tests/

Run with mock analyzer (no API calls):

python cli/main.py analyze --input data/input/reviews.csv --mock

License

MIT