Monday, January 17, 2011

How To "Engineer" Your Unstructured Data

Engineers are all about  structure. Especially civil engineers. I should know. I'm married to one. It's fascinating (and boring at times) to know how much research is done just to build a road, tunnel, or bridge. That includes sample testing, design, and quality assurance to ensure the dirt below the foundation is structurally stable enough to build on and to make recommendations on which materials to use. Failure to do so will result in major catastrophes and lower the quality of life in  particular environments. My husband loves to tell people that "engineers saves more lives than doctors."

Taking everything I've soaked in about engineering I thought about how we as researchers may apply it to the research industry. Obviously the more structure we have in place the more efficient we will be in gathering, segmenting, and analyzing data as well. With the growth of the internet and social media marketing many companies are wondering how to better understand what clients are saying when they are leaving comments on the company blog or speaking to them via Twitter or Facebook. "How can I compare my open-ended comment answers from a survey with what Facebook fans are saying?"

In the old days text analysis will be out-sourced to a research firm with a coding department where actual people would thumb through paper surveys that were sent in the mail and mark/sort them according to the categories set forth by the client. These days new text analytics technologies are popping up everywhere to fill the void and cut costs while allowing analysts more control over the process. Trend watchers say text analytics technologies is one of the top 15 trends to watch in the next 3 years.

If you are interested in learning more about this growing trend and want to know more about how to bring "structure" to your unstructured data I urge you to join us for tomorrow's webinar:

Tuesday 1/18/10 at 9:00 AM PST

Unlock the Power of Text Analysis with DiscoverText


Dr. Stu Shulman, President and CEO of Texifter, LLC, will present the following:

* How to augment your research with powerful, time-saving text tools
* Import thousands of documents from various sources and formats
* Discover information using powerful software and human techniques
* How easy it is to get started!

We will then show you an example using existing data from SurveyAnalytics and how easy it is to send your text files to DiscoverText, set-up codes, and reporting capabilities.

The session with conclude with a Question and Answer Session.
All slides and recording of webinar will be available on blog.surveyanalytics.com within 24 hrs of the presentation.

Space is limited so register today: https://www2.gotomeeting.com/register/967147715

About The Presenter:

Dr. Stuart Shulman, CEO and founder of Texifter, LLC is a spin-out company based on text analysis research He is also the founder of QDAP-Pitt and current Director of QDAP-UMass. Over the last 10 years, Dr. Shulman has conducted research primarily in the domain of electronic rulemaking, with a specific focus on the development of human language tools for reviewing large numbers of public comments. Over the past three years, he has been sole inventor on five distinct software applications that support tasks ranging from text classification, blog capture, and manuscript review management, to the development of better annotation for improving optical character recognition. New research focuses on the detection of threats and novel ideas communicated in public comments, blogs and other media. Texifter is a vehicle for the commercialization of new tools developed in the QDAP labs.
iframe {max-width:100%;} .embed{ width: 100%; }