Pages Navigation Menu

Applying Sentiment Analysis to Star Wars: The Force Awakens

Posted by on Jan 20, 2016 in Blog, Data Visualization, Datamining, Geolocation and Psychogeography | Comments Off on Applying Sentiment Analysis to Star Wars: The Force Awakens

One of the more influential sites for data scientists, KDNuggets recently published a case study showing how sentiment analysis could be applied to track the reaction around a film’s early release cycle.  In this case, the film was the 2015 holiday blockbuster Star Wars: The Force Awakens.

10 milliostarwarsSA-1n tweets were collected through the Twitter API, between 12/4/15 and 12/29/15, with the release date on 12/17/15.  About 2.5% contained geolocation data either in form of direct coordinates or human readable location (e.g. New York). The researchers said “…the first thing we looked at was the frequency of Star Wars related tweets in time. It is clearly visible that most of the tweets came from US and UK, which can be easily explained by popularity of Twitter itself in these countries. Next thing to see is the periodicity of day and night, where people tweet more at night than during the day. Also the timezone shift is clearly visible.  More interestingly, we can see the build up before the release, as the number of tweets is increasing for a few days before the world premiere and sky rocketing on this day…”

starwarsSA-2Each tweet was assigned a score between -1 and +1 (-1 being highly negative, +1 highly positive). Results were plotted in a hexbin map, visualizing global sentiment and aggregating by mean within the cell.  Interestingly, average sentiment shows a steady decline as the time passes. There is an observable dip on the day of world premiere but “sentiments keep steadily low the whole time.” The researchers make several interesting observations concerning the results.  Since worldwide interest in the film, at least as reported in the media, approached general hysteria, why doesn’t the Twitter analysis parallel this?

One possible explanation is the inherent sampling bias when working with social network data.  After all, data is derived only from those who voluntarily decide to share. These are usually the ones with stronger opinions – either highly positive or negative, producing a somewhat polarizing effect.  Next,  sentiment analysis is constrained by the modeling methods and tools available for Natural Language Processing (NLP), and one of these constraints is that the algorithms require a data corpus in the English language.  Sentiment analysis that proposes a global sampling plan will necessarily have gaps in its dataset, since non-English texts will be omitted from the analysis.

Read More

A Year in Snapshots: Photos around the World for each day of the year

Posted by on Jan 3, 2012 in Blog, Data Visualization, Geolocation and Psychogeography | Comments Off on A Year in Snapshots: Photos around the World for each day of the year

triposo_photos.jpg
Mobile travel guide provider Triposo provides free travel guides, augmented by location based data. ? With the abundance of data from a variety of open datasets, ? Triposo has created a remarkable short video showing photos taken around the world, on each day of the year. ? On a world map, ? the relative density of the photo data is shown.

Notice the brighter dots at specific points in time in several regions, many probably related to holidays. For example, around the beginning of October in China or May 1, in Europe. The October surge could also have something to do, though, with the “default” date of many digital cameras out of the box being preset for October 1.

See the clip on Youtube

 

Read More

Embedded Camera-Eye

Posted by on Jul 10, 2011 in Bioreactive Media, Blog, Emerging Science and Technology, Geolocation and Psychogeography | Comments Off on Embedded Camera-Eye

After losing an eye in a car accident, Tanya Vlach is trying to raise money to have a webcam fashioned like an eye installed in the non-functional eye socket. The prosthetic camera-eye is designed to be waterproof, capable of wireless HD video transmission at 720p, and will include zoom and still-capture features, activated by blinking. On Vlach’s wishlist for the eye — expected to cost at least $15,000 to design and embed — is facial recognition, geotagging, and infrared viewing.

Read More

Visualizing Science Readers

Posted by on Dec 8, 2010 in Blog, Data Visualization, Datamining, Geolocation and Psychogeography, Media and Markets | Comments Off on Visualizing Science Readers

Curious about what scientists might be reading?? Springer (noted publisher of more than 5 million scientific and academic titles) has launched a new analytics tool that reveals how its users and subscribers are downloading its content.

There are a number of interactive visualization tools at the site,? including a world map illustrating the origin of download requests,? an updating topical/keyword tag cloud,? and displays of real time downloads.?? This type of information could be used in many, many ways.?? On the commercial level,? authors and editors can get a fascinating view of which topics are emerging and in what geographic markets.? Scholarly and research applications abound, as well.? As a community,? scientists have an outsized impact on society and understanding trends in their work and interests could be useful.

Read More

The Marfa Ring

Posted by on Dec 11, 2009 in Blog, Geolocation and Psychogeography, Media and Markets, Technology and Art | Comments Off on The Marfa Ring

The Marfa Ring project is an experiment in colonizing the virtual geography of the small town of Marfa, Texas by creating a “Web Ring” of sites about it. Due to the Ring’s interlinking, Google search results are skewed in favor of the ring sites (which vary in levels of veracity and intent) as opposed to Marfa’s legitimate web presence.

Read More

We’re watching

Posted by on Oct 18, 2008 in Bioreactive Media, Blog, Data Visualization, Emerging Science and Technology, Geolocation and Psychogeography, Media and Markets, Technology and Privacy | Comments Off on We’re watching

A Canadian company has developed a technology called Eyebox2 that it says can monitor the gaze of passersby and respond to shifts in attention,  even tracking multiple people at once,  and even from more than 30 feet away.

The company says the advertising potential is large, wherein outdoor or unconventional ad space could be sold "by the eyeball." In addition, the technology could provide a new kind of test lab for evaluating and refining ads in unusual contexts.

Read More