Welcome to the Restaurant Analysis project! This repository contains the code and data used to analyze various aspects of restaurant performance, customer reviews, and other relevant metrics.
This project aims to provide insights into restaurant performance through data analysis. By leveraging customer reviews, price range , and other relevant information, uncovering trends and patterns that can help restaurants make informed decisions.
- Importing Libraries: First, ensure you have all the necessary libraries imported. This typically includes libraries for data manipulation, and visualization.
- Importing Dataset: Load the dataset into your workspace. Ensure the data file is placed in the 'df' directory.
- Basic Information of the Dataset: Get an overview of the dataset to understand its structure, including the number of rows, columns, and data types.
- Data Cleaning: Clean the dataset to handle missing values, duplicate records, and any inconsistencies.
- Exploratory Data Analysis (EDA): Perform EDA to understand the data and identify any patterns or trends.
- Generate Reports: Generate detailed reports and visualizations of the analysis results.
-
Level 1
- Top cuisines
- City Analysis
- Price range distribution
- Online delivery
-
Level 2
- Restaurant ratings
- Cuisine combinations
- Geographical analysis
- Restaurant chains
-
Level 3
- Restaurant reviews
- Vote analysis
- Price range v/s Online delivery & Table bookings
- North Indian (936), North Indian, Chinese (511), and Chinese (Chinese) are the top 3 Cuisines among all the restaurants. Percentage of restaurants that serve each of the top cuisines are 52.0%, 28.4% and 19.7% respectively.
- New Delhi (5473) is the city with the highest number of restaurants in the dataset. Debriefed on the average rating for restaurants in each city. 'Inner City' stands out on thr top for city with the highest average rating of 4.9.
- Restaurants having least price on there dishes get more orders. Order with price range of 1 have been ordered more, 46.51% of the entire dataset.
- 2,451 restaurants are providing online delivery, 25.7% of the entire dataset. Restaurants providing online delivery are given more average ratings.
- The most common rating range exists from 3 - 4, the most common rating in all the dataset is marked at 3.5 and Average number of votes received by restaurants are 156.77.
- There are 1,731 cuisine combinations, North Indian and Chineese is the most common cuisine combination. Italian & Deli, the cuisine combinations in the dataset having higher ratings.
- Geographical analysis of the restaurants helps in determining the clusters forming in South east Asia, majorly from Idndia the followed by USA and then the rest of the world.
- 734 restaurant chain in the dataset, and Cafe coffe day has most numbers of restaurnats (83). "Talaga Sampireun" the restaurant chain that has total of 5,517 votes with average rating of 4.9.
- The most common positive review is "Average" with 3,734 counts (39.1% of the data). The most common negative review is "Poor" with 180 counts (1.9% of the data). Additionally, there is a negative correlation between review length and rating: as review length increases, the aggregate rating tends to decrease.
- The restaurant "Toit" from Bangalore, India, received the maximum votes (51,705), while "Cantinho da Gula" from Sao Paulo, Brazil, received none. The weak positive correlation between the number of votes and the aggregate rating is confirmed by a correlation coefficient of 0.31.
- The proportion of restaurants offering online delivery rises with higher price ranges, peaking at range 2. Similarly, table booking availability increases with price, highest in ranges 3 and 4. Among 1,991 high-priced restaurants, 464 (23.3%) offer online delivery, while 918 (46.1%) offer table booking.
I welcome contributions to improve this dashboard. If you have suggestions or improvements, please create an issue or submit a pull request.
For any questions or support, please contact me at [rupal95@outlook.com].