Speaker to address machine learning techniques for text, imagery, video processing

10:11 a.m., March 17, 2016--Cybersecurity expert James Nolan will deliver the third and final lecture in a series hosted by the University of Delaware Cybersecurity Initiative on Wednesday, April 6. 

The lecture, “Machine Learning Techniques and Applications for Text, Imagery, and Video Processing,” will take place in Mitchell Hall from 2:30-3:30 p.m., with a reception to follow in the lobby of DuPont Hall.

Events Stories

June 5: Blue Hen 5K

University of Delaware graduates planning to attend Alumni Weekend are encouraged to register for the annual Blue Hen 5K walk and run, which will be held Sunday morning, June 5.

June 6-9: Food and culture series

The 20th annual June Lecture Series at the Osher Lifelong Learning Institute at UD in Wilmington will be held June 6-9. Titled 'June a la Carte,' this year's program focuses on great political documents, feminism, world politics and a Newark cuisine sampler.

• June 6-9: Food and culture series
• Through June 30: 'Saving White Clay Creek'
• July 22: WWE Heatwave Tour
• Through Sept. 16: Women of color in academia

Lecture abstract

Today’s technological environment provides an unprecedented number of sensors to collect data. From data collected by satellites and surveillance cameras, to social media, to data collected on iPhones, people are experiencing a data explosion. 

In this data-rich, information-poor situation, people’s ability to develop and deploy new sensors to capture data far outstrips their ability to process and make sense out of it.  Much of this information is captured as raw text, imagery, and video, and extracting the important information from this sea of data remains a difficult challenge.

In this talk, Nolan will discuss machine learning techniques used to solve this problem — for example, statistical topic modeling, entity disambiguation, semantic role labeling, relationship extraction, and deep learning — and their real-world solutions and applications.

About the speaker

James Nolan is the vice president for analytic technologies at Decisive Analytics Corporation (DAC) in Arlington, Virginia. He leads a world-class team of mathematicians, computer scientists and engineers in the development of patented text, video, audio and data fusion capabilities that are deployed worldwide in military, intelligence and commercial systems.

Nolan received the 2007 Tibbetts Award for transitioning statistical techniques for intelligence analysis to the Department of Defense.

Nolan holds master’s and doctoral degrees from George Mason University, where he currently serves as an adjunct professor of computer science.

Article by Diane Kukich