Tag Archives: statistics

Hunting Political Fear

The R&D team has been preparing an academic basic research talk I will make at the October 30, 2012 Sentiment Analysis Symposium. The title is “Fear and Loathing on the Social Campaign Trail” and we have been working with a … Continue reading

Posted in research, Social Media | Tagged , , , , , , , , | 1 Comment

Heat Maps Arrive

 We are thrilled with the response to v1 of our heat map tools, which provide a unique glimpse into the nature and scope of inter-coder agreement in DiscoverText. This video introduces the beta version of the tool. Overtime, the tool … Continue reading

Posted in DiscoverText, general, product, research | Tagged , , , , , | Comments Off

Coders Needed!

We are building up a pool of coders for some new projects. If you are good at labeling text this may be an excellent opportunity for you. Graduate students are particularly welcome to join the pool. Coders accepted into the … Continue reading

Posted in DiscoverText, general, research | Tagged , , , , , , , , , , , | 1 Comment

Using the CloudExplorer

Tag clouds have gained popularity as a short cut to discovery of frequently occurring terms in a text collection. Much of the inspiration comes from the remarkable Wordle site at http://www.wordle.net. Building on that inspiration, we engineered an interactive word … Continue reading

Posted in DiscoverText, general | Tagged , , , , , , , , , , | Comments Off

Interactive Custom Machine Classifier Histograms

This is the latest DiscoverText filtering feature designed to speed up the creation of accurate custom machine classifiers. This video shows how we use an interactive display of classifier scores to isolate items in a dataset that require further human … Continue reading

Posted in DiscoverText, general, product | Tagged , , , , , , , , , , | Comments Off

Analyzing Co-Occurrences

A common question we hear at Texifter is, “How can one derive co-occurrences in a slice of data?” DiscoverText makes this process relatively easy – when starting with one known variable. For instance, one might like to see which hashtags … Continue reading

Posted in DiscoverText, general, research | Tagged , , , , , , , , , , , , , , , , , , , , , | 1 Comment

Election Analytics Part I: Romney Trends in April

[In the coming  months leading up the U.S. presidential election, the Texifter blog will be featuring a multi-part series on various analytical insights into the election, which can be gained by utilizing a host of social media analytics (SMA) methods within … Continue reading

Posted in DiscoverText, research, Twitter | Tagged , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , | 1 Comment

Klout Marketing by Spotify

In Q1 of this year, serious internet chatter surfaced about the possibility of popular Swedish online streaming music service Spotify launching in the U.S. Spotify is a tantalizing proposition for the American media consumer. While there are costs incurred, the service … Continue reading

Posted in general | Tagged , , , , , , , , , , , , , | Comments Off

Coding Text – Part Three

Researchers interested in large text collections and their itinerant coders tend to muddle through with limited collaborative, cross-disciplinary resources upon which to draw. The generic criteria for high-quality codebook construction and effective coding are underdeveloped, even as the tools and … Continue reading

Posted in general | Tagged , , , , , , , , | 1 Comment

Coding Text Using the “QDAP Method” – Part One

We did it! The free, open source, Web-based, university-hosted, FISMA-compliant “Coding Analysis Toolkit” CAT recorded its one millionth coding choice. Pretty much all the credit goes to Texifter CTO and chief CAT architect Mark Hoy who has put in many … Continue reading

Posted in general | Tagged , , , , , , , , | 1 Comment