IMDb data from Two Generations, from 1979 to 2019; Part one, Dataset Introduction and Preliminary Analysis

TL;DR

This paper introduces a comprehensive IMDb dataset covering 79,000 titles from 1979 to 2019 and provides preliminary analysis on trends, demographics, and factors influencing movie success.

Contribution

It creates the largest IMDb dataset to date and offers initial insights into data trends, demographics, and success factors using statistical and machine learning methods.

Findings

Identified trends in IMDb data over four decades

Analyzed demographic patterns of IMDb scores

Explored relationships between genre, ratings, and success factors

Abstract

"IMDb" as a user-regulating and one the most-visited portal has provided an opportunity to create an enormous database. Analysis of the information on Internet Movie Database - IMDb, either those related to the movie or provided by users would help to reveal the determinative factors in the route of success for each movie. As the lack of a comprehensive dataset was felt, we determined to do create a compendious dataset for the later analysis using the statistical methods and machine learning models; It comprises of various information provided on IMDb such as rating data, genre, cast and crew, MPAA rating certificate, parental guide details, related movie information, posters, etc, for over 79k titles which is the largest dataset by this date. The present paper is the first paper in a series of papers aiming at the mentioned goals, by a description of the created dataset and a preliminary analysis including some trend in data, demographic analysis of IMDb scores and their relation of genre MPAA rating certificate has been investigated.