StatSoft Webinar (Feb 19): Reduce Journal Research from Weeks to Hours with STATISTICA Text Miner

Can a recommender service be created to automatically identify a few relevant articles from many thousands ? This webinar shows a successful case study with STATISTICA Data Miner and Text Miner.

StatSoft February Webinar: StatSoft"Reduce Journal Research from Weeks to Hours
with STATISTICA Text Miner: A Success Story"

Free Registration =

Researchers find it difficult to keep up with the ever-increasing number of journal articles in their fields of research. For instance, a PubMed search for "Breast Cancer Genetics" returns over 45,000 hits alone. Refining search strings only goes so far to reduce this number, so experts still must sift through many abstracts - a cumbersome and laborious process - perhaps only eventually to find that very few are relevant.

Can a recommender service be created to identify relevant articles and, if so, could such a service be automated? The answer to this question is yes, with tools such as STATISTICA Data Miner and Text Miner.

Furthermore, since XML is becoming a standard means of storing and sharing information, this presentation may have additional cross-appeal, as importing XML with a STATISTICA Visual Basic (SVB) macro is discussed.

Join us Tuesday, Feb. 19, as Toby Barrus, Sr. Quality Engineer at Myriad Genetics, presents a case study for a PubMed recommender service using STATISTICA Data Miner and Text Miner. He based this successful implementation on an example from the textbook, "Practical Text Mining," by Dr. Gary Miner, et al. Toby will cover the following topics:

  • Brief overview of PubMed search results and the "Send to" feature to export data in XML format
  • Import XML data into STATISTICA using a SVB macro
  • Text Miner to convert free-form text into usable format for data mining
  • Data Miner to create and compare predictive models
  • Rapid Deployment of best model to new search results

Presenter Bio: Toby Barrus is a Senior Quality Engineer for Myriad Genetics, where he is primarily responsible for validation of new molecular diagnostic tests. He is also heavily involved in continuous improvement activities where he uses techniques such as Lean, Six Sigma, and Data Mining. He obtained his Six Sigma Master Black Belt Certification from Arizona State University in July of 2011. He also has earned a Graduate Certificate in Data Mining from the University of Louisville and a MS in Physics from The Ohio State University.