Fizziology

...

movies did not show up on the logarithmic scale, outliers are to be expected when the data from 300+ movies is taken.

All in all, this is good news for marketing departments in the film industry. As stated before, compiling Twitter data to provide a decent sentiment analysis is tricky, and may not be worth it anyway. On the other hand, it is relatively easy to track the amount of page views and edits a certain Wikipedia entry has. The tough part there is finding a norm from which to compare, as the researchers attained their data via Wikimedia Deutschland. It may not be possible for anyone who asks for it to copy that data.

It is also important to note that this metric measures the amount of people expected to see the movie, not the perceived quality. For that, Twitter may end up being the stronger tool.

With that being said, movie marketing may look past Twitter and to encyclopedic sites like Wikipedia and IMDB to estimate box office revenue a month in advance and adjust their strategy accordingly.

.....................

Page 1 of 1

Transcript: Demo: Movie Review Sentiment Analysis

Presenter: Jayatheerthan

In this course you will learn how to develop a simple sentiment analytics application using IBM

BigInsights Text Analytics.

We will demonstrate how to perform analysis of movie reviews fetched from websites such as

ImDb.com.

Let us assume that the HTML pages of reviews of two famous Bollywood movies, Swades and Ra.One,

have been ripped from the web and stored as files on the local hard disk.

The input to our text analytics system is a collection of snippets from the ripped pages as shown on this

slide. Each of the snippets contains review comments posted by various reviewers. These snippets

contain some words that indicate positive, negative, or neutral sentiments of the reviewer.

This is a sample code snippet that can identify objectives from the input document.

You will learn more about text analytics programming in a future course. For now just understand that

you can use a programming language called AQL to write rules that can extract information from an

unstructured text source.

This is the output of running the extractor code shown on the previous page for the movie Ra.One. You

can notice that the adjectives that are picked up are an indication of positive, negative or neutral

sentiments expressed by the reviewers. Most of the reviewers have used negative adjectives which

explain why the overall rating for the movie is low.

The sentiment extraction result for the movie Swades shows more positive adjectives indicating good

reviews.

What you have seen now is just a tip of the iceberg.

Text analytics engines can perform very complex analytics on very large volumes of unstructured text

and distil relevant information that your business requires.

We hope that you understood how text extractors could be used to perform sentiment analysis.

...............

he days of predicting a box office hit by studying the impromptu reaction of movie-goers and counting how many of them visit the threatre may be numbered.

A new wave of players from the technology sphere is set to transform the age-old science of predicting movie popularity by introducing computer algorithms and search patterns into the equation, a move several new-generation Hollywood producers applaud.

"How much we’re talking about a film online is going to have much more of an impact," says Adi Shankar.

Shankar is executive producer and chief executive of 1984 Private Defense Contractors, the company behind unexpected hits such asThe Grey, starring Liam Neeson and Lone Survivor with Mark Wahlberg.

After the success Shankar’s films have enjoyed on DVD and video-on-demand as well as in cinemas, he says traditional movie "tracking" isn’t enough of a gauge anymore.

"Traditional tracking is flawed because it depends on the kind of movie," the 29-year-old says. "For a certain generation of movie-goers and a certain type of movie, tracking can be incredibly effective, but theatres in general are just outdated."

Now technology giants including Adobe and Google are stepping in to the breach. Last June, Google released the white paper Quantifying Movie Magic with Google Search, in which the company claimed it could predict the opening weekend box office takings of a movie with 92 per cent accuracy. It based its claim on its observation of a 56 per cent increase in online searches for titles, trailers, reviews and other information about a film between 2011 and 2012.

Adobe is using its Marketing Cloud, a suite of products and services including social media and other analytics that provide feedback on digital marketing, to measure the number of online mentions a move title gets even before its release.

The company got Hollywood talking when apart from predicting the films Thor: The Dark World, The Hunger Games: Catching Fire and the second instalment of The Hobbit would all perform well, it predicted – against Hollywood’s better judgement – that Anchorman 2 andJack Ryan: Shadow Recruit would be "surprise hits". Jack Ryan flopped, and Anchorman 2 had a disappointing start.

Tamara Gaffney, principal analyst for Adobe’s Digital Index, stands by the company’s record, saying that when Jack Ryan’s release date changed last year, so did the social chatter that pointed to it being a hit.

"As Jack Ryan’s release date was pushed back it lost momentum," she says. "At the time of our updated prediction, it was only averaging 1410 mentions per day where Anchorman 2 (35,300) and Hunger Games (110,550) averaged much more. Average weekly growth prior to the first prediction was 27 per cent, in line with the profitable