Built with datascienceportfol.io

IMDB Movie Data Analysis

Project photo

Project links

Skills

About this project

Project Overview

This project analyzed a dataset of over 5,000 movies from IMDB to uncover insights about film profitability, director performance, audience preferences, and long-term voting trends. The study focused on data cleaning, exploratory data analysis (EDA), and visualization to identify factors contributing to movie success and popularity.

Key Responsibilities

Data Extraction & Cleaning

  • Worked with a dataset of 5,043 movies containing 28 features.

  • Handled missing values using mode/median imputation (e.g., replacing missing language with “English”).

  • Removed duplicate rows and irrelevant entries.

  • Treated outliers in budget, gross revenue, and reviews.

Feature Engineering

  • Extracted relevant features: director, actors, budget, revenue, genres, IMDB score, votes.

  • Created a new profit feature (gross – budget).

  • Grouped movies by decades to analyze audience engagement trends.

Exploratory Data Analysis (EDA)

  • Identified the most profitable films (e.g., Avatar, Titanic, Jurassic World, Avengers).

  • Constructed IMDB Top 250 rankings using ratings and audience votes.

  • Analyzed top directors with highest average IMDB scores (Christopher Nolan, Hitchcock, etc.).

  • Explored popular genres based on mean IMDB ratings (Crime, Drama, Fantasy, Mystery).

  • Compared films starring Leonardo DiCaprio, Brad Pitt, and Meryl Streep on critic and user reviews.

  • Visualized audience voting patterns across decades, showing rapid growth from 2000s onwards.

Visualization & Insights

  • Built bar plots, box plots, and trend charts to highlight findings.

  • Designed clear visual narratives linking movie profitability, critical reception, and audience demand.

Key Achievements

  • Successfully cleaned and prepared a complex dataset with 5,000+ entries.

  • Built actionable insights into profitability, audience voting trends, and popular genres.

  • Demonstrated strong storytelling with visualizations for business and research impact.

  • Delivered a scalable, repeatable workflow for entertainment data analytics.

Challenges Addressed

  • Extensive missing values across multiple columns.

  • Handling duplicates and inconsistent data formats.

  • Detecting and managing outliers in revenue and budget.

  • Translating raw movie metadata into meaningful business insights.

Future Scope

  • Apply machine learning models to predict movie profitability and ratings.

  • Integrate social media sentiment (e.g., Twitter, YouTube reviews) into the analysis.

  • Extend analysis to OTT platforms (Netflix, Prime, etc.) for modern relevance.

  • Incorporate actor/crew-level performance modeling for deeper insights.