anqi-lu / TMDB-keywords

Movie plot keywords visualization.
https://anqi-lu.github.io/TMDB-keywords/
MIT License
2 stars 0 forks source link

TMDB Movie Plot Keywords Visualization

A project that visualizes most frequent plot keywords in movies from TMDB 5000 Movie Dataset on Kaggle.

Deployed on github pages at https://anqi-lu.github.io/TMDB-keywords/.

The TMDB 5000 Movie Dataset contains information for about 5000 movies from The Movie Database(TMDB), a crowd-sourced movie information database. The movie information includes movie genres, country, actors, directors, plot keywords, gross profit, and much more.

Prototyping

The first visualization I made on this dataset was a donut chart, illustrating the number of movies by country. It shows the which countries are the movies from. Because TMDB is built entirely by the user so there could be a lot of bias in the data (Most movies are from western countries). Therefore, I decided that I was not going to do much with the “country” attribute of the data. Instead, I decided to focus on the “plot_keywords”, “genres”, and “year”.

The questions about this dataset became:

According to the questions, I listed the following tasks:

  1. Word cloud of top 20 key words for all genres
  2. Genre filter — display top 20 keywords for each genre 3. Line chart of count by year
  3. Line chart of count by year

Sketch

sketch

Source of Inspiration: 60 years of french first names and stream-graph explorer

This sketch brings the additional tasks:

  1. Select multiple keywords and assign different color for each line in the line chart
  2. Hover to select a year and corresponding tooltip
  3. Zoom by brushing on the year

Interactions

wordcloud genre screen shot 2017-11-06 at 8 26 05 pm tooltip

Future Work

Attribution

This D3.js web project is forked from curran's this template project.