Star Wars & Sentiment Analysis

  2 m, 51 s

Since everyone’s favorite galaxy far, far away returned to the big-screen in 1999 with the first of George Lucas’s much-maligned Star Wars prequels, the release of a Star Wars film is bound to cause some controversy. However, the release of Star Wars The Last Jedi on 12/15 was followed by perhaps the most intense and immediate fan backlash to any of the previous films. Perhaps I’m biased, but this seems like a galactic crisis that only natural language processing, machine learning, and sentiment analysis can truly solve.

In the week preceding the wide release, Star Wars The Last Jedi screened for Hollywood insiders and reviewers around the country. Once that happened, plot spoilers inevitably made their way to the internet. Director Rian Johnson truly made a unique Star Wars film, and some of what makes it so unique was lost in translation from screen to online forums like Twitter, Reddit, and the internet’s most wretched hive of scum and villainy 4Chan. Cries over the ruination of Star Wars took over, and soon hashtags like “#TheLastJediAwful” were trending higher than the usual “#TheLastJedi” (that came complete with a BB-8 emoji). To be clear, none of this was out of the ordinary because no one hates Star Wars like Star Wars fans. But then something strange happened on Rotten Tomatoes, a site whose ratings are a go-to source for film fans.

At the time of this writing, only 55 percent of users reviewing the film liked it. This contrasts greatly with the 93 percent “fresh” rating given to the film by critics. It also marks the lowest audience score for any Star Wars film, including Star Wars Attack of the Clones which many regard as the “worst” film in the whole series. What makes this even more interesting is that the disparity is usually the other way around for blockbusters. Typically, critics hate the film, while audiences give it a much more favorable score.

To be clear, other groups who measure audience sentiment don’t show the same sort of disparity. Further complicating things, a Facebook user who goes by “Down With Disney” took credit for using social media bots to artificially decrease the audience score to hurt the studio. This is just the latest in a series of incidents that call into question the Rotten Tomatoes method of aggregating user data to effectively measure how they feel about a film.

That’s where sentiment analysis and text analytics comes in. Using Lexalytics’ tools, it’s possible to mine the data on social media and reviews from sites like Rotten Tomatoes, but also IMBD and other forums to get a clearer picture of how these folks (or, in some cases, bots) feel about the film. Are people angry about the story itself? Are they angry about the perceived progressive politics in the film? Are they just mad it’s not “better than The Empire Strikes Back?” We can find out.

People often think of text analytics as only useful for service-oriented industries like airports or hospitality. However, if studios want to truly understand what their audiences think of their films, natural language processing solutions like Salience, or the Semantria suite, are the perfect tools to use. And we’re going to show you how. Over the next few weeks we are going to run an analysis of reaction to the Star Wars film, to see if the so-called “backlash” is really happening at all.

Categories: Insights, Sentiment Analysis

Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>