movie-metadata

Visualizing Movie Metadata

david lu (dlu1)

--

Introduction

Project for an Independent Study, CMU Spring 2018.

This project examines different data visualization methods for exploring a collection of data, for easier human consumption and processing of said data.

This specific case uses Movie Metadata and tries to answer the question "What are the top rated movies of a specific genre or collection of genres, with various other limiters?"

Goals of this project:

• Make it easier for a user to process and understand than a text-based list.
• Make it easier for a user to use and navigate for a specific query.
• Make it intuitive for a user to use, even without instructions.

--

Future Work

There's a lot to improve upon here, but here are some immediate things of note:

• Make the website responsive.
• Make it easier for users to realize how to deselect a selection
• Explore different graph layouts
• Explore more meaningful metadata sorting graphs/fields (maybe month isn't as meaningful)
• More meaningful filtering of movies (e.g. not enough votes / not popular enough, maybe follow IMDb's list)
• Better data collection methods (IMDb's top lists per year?)

--

Data

Data collected from MovieLens and IMDb.

Trying to transition to solely IMDb.

The movies object is organized as such: It's an array of all movies in the MovieLens database subset, with its MovieLens ID being the index position in the array. (MovieLens Movie ID 423 = movies[423])

Each movie object in this array then contains the following fields:
[budget, country, director, durationMin, genres[], gross, id, imdbID, imdbRatingAvg, imdbRatingNum, language, mpaaRating, ratingAvg, ratingNum, releaseDate, releaseYear, title].

As a reference, here is one element (MovieLens ID #1, Toy Story):

1: Object
  budget:"$30,000,000"
  country:"USA"
  director:"John Lasseter"
  durationMin:"81"
  genres:Array(5)
    0:"Adventure"
    1:"Animation"
    2:"Children"
    3:"Comedy"
    4:"Fantasy"
  gross:"$191,796,233"
  id:"1"
  imdbID:"114709"
  imdbRatingAvg:"8.3"
  imdbRatingNum:"666,965"
  language:"English"
  mpaaRating:"G"
  ratingAvg:"3.921239561324077"
  ratingNum:"49695"
  releaseDate:"22 November 1995 (USA)"
  releaseYear:"1995"
  title:"Toy Story (1995)"

With this data organized as such, I should be able to easily graph out relationships between various fields, into the various module concepts and link them appropriately.