This Halloween, we built something spooky using a new product of ours.
It’s diff(erent) then anything we’ve ever done before.
Diff is an API for scraping and monitoring changes to website data and content. It’s used to make scraping data easier, and we created it so we could use it on our internal products. But we thought it could be super useful for other companies too — especially data scientists, journalists and anyone who needs to monitor data changes or use large sets of data.
We made a horror movie data visualization with Diff.
In the spirit of Halloween, we decided to show off what our Diff Web Scraping API can do by taking the scripts from our favorite horror movies and scraping that data to discover the emotional arc of each film. We used the Diff API to analyze the rollercoaster of emotions our favorite horror films inflict on our psyches and put this data in graph form for visualization. We’ve got Halloween, Silence of the Lambs, Carrie and more.
Experience it yourself…if you DARE
Sink Your Teeth Into the Data
Our Halloween-themed adventure also included a word cloud and bigrams to analyze the overarching language of the film. This program uses natural language processing and sentiment analysis of horror movie scripts. Essentially, the emotion of language throughout the film ranges from negative emotion to positive emotion. In The Silence of The Lambs, when Precious falls into the well, that’s where you see the lowest point — the valley. The FBI graduation ceremony is the peak.
At Dirigible Studio, we pride ourselves on being cross-disciplinary, having a deep bench of talent, and fostering that talent. Jake Casey, partner and developer, is the perfect example of this. He’s interested in the intersection of coding, artificial intelligence and humans, and Diff was born as one of his passion projects.
Think Diff can make a difference with your data?
Published on Oct 27 2021
Last Updated on Jan 23 2022