The Austin Chapter of the IEEE Computer Society was created to provide a forum for technology professionals to share knowledge and network.
Located in beautiful Austin, Texas where we enjoy rolling hills, 9 area lakes, plenty of live music and of course, a technology hotbed. Austin was third after San Jose and Sunnyvale in 2006 in the number of patents issued.
Want to stay up-to-date with what's happening in your industry?
Keep in the know and network!
TOPIC: A Graph-Theoretic Markovian Algorithm in Web Analytics
PRESENTERS: Dr. Ram Kosuru, HP and Dr. Choudur K. Lakshminarayan, HP
DATE: October 21, 2015
TIME: 6:30 pm – 8:30 pm
LOCATION: The Advisory Board Company, 12357-C Riata Trace Parkway, Bldg 7, Suite 100, Austin, Texas, United States 78727
FREE Food & Drinks, Limited Seats
RSVP is required: https://meetings.vtools.ieee.org/m/36356
Much effort in Big Data Analytics is dedicated to discerning human intent in order to improve revenue, profit, satisfaction and other key performance indices (KPI). One example is modeling clickstreams to predict behaviors of web browsers. More particularly, we model browser intent (buy/no buy) as a function of sequence of clicks at the session-level. We use the labeled data for the two classes. Often the training set for the class “buy” is much smaller than that of the class “no buy.” In this exposition we present techniques to augment the training set of the class buy to improve the classification of unknown data (new sequences). We apply Markov chains of various orders (complexity) to distinguish buyers from non-buyers. The techniques are also used in multi-class classification and are thus extensible.
Dr. Ram Kosuru holds a Ph.D. in Engineering (Computer Science Major) from University of Wisconsin at Milwaukee. He has been working in HP for the last 17 years; and briefly worked in Hitachi Computers (North America). IDr. Kosuru has 18 patents/patent applications in database systems, data analysis with US patent office and one patent application with India patent office. His areas of interests include Database architectures (systems and design), machine learning algorithms, data mining, algorithm design.
Choudur K. Lakshminarayan is a Principal research Scientist in HP. He works in the areas of mathematical statistics, applied mathematics, and machine learning with applications in data mining, data compression, and data analysis of structured and unstructured data in real-time settings. He holds a Ph.D. in Mathematical Sciences and lives in Austin, Texas.