Analyzed restaurant data in Bangalore to determine food preferences in specific neighborhoods. The project involved preprocessing the dataset by identifying name and address as primary keys to eliminate duplicates. A food corpus was created from restaurant menus, and reviews with ratings ≥ 3 were analyzed to extract the most liked food items for each restaurant.
To ensure accurate insights, unstructured menu data was carefully refined to remove noise and non-food items. By linking review data with the refined menu corpus, the project successfully identified popular food trends across various localities, providing valuable insights for targeted restaurant menu optimization.