Internet Lurkers: Large-Scale Measurements and Modeling
Dec 04, 2012
from 10:00 AM to 11:00 AM
|Where||4750 Boelter Hall|
|Contact Name||Prof. Mihaela van der Schaar|
|Add event to calendar||
K.P (Suba) Subbalakshmi
Department of Electrical and Computer Engineering
Stevens Institute of Technology
Active user engagement on the Internet is often used to track topical public sentiment, ranging from business and politics to healthcare. These engagements include tweets, social media posts, blogs, comments, url-sharing, etc., to name a few. However, little is known about the activities of a passive consumer, or a “lurker”. Lurkers consume topical information but refrain from participation actively, and hence leave very few measurable traces. Our large scale Internet measurements indicate that as much as 99% of online users are, in fact, lurkers. Despite the major role they play in Internet information consumption, our understanding of topical lurking is still in its infancy.
In this talk we will discuss our large-scale, measurement based study to establish, characterize, and track the activities of Internet lurkers. Using a variety of collected datasets on Internet activity patterns (e.g. YouTube, twitter) and case studies on trending topics, we characterize lurking entirely by measuring publicly available information on the Internet. Stochastic models to understand how lurking affects information spread and the information theoretic capacity of this type of passive information consumption channel will be discussed. We will conclude by offering insights into the activities and role of a lurker, including: (i) lurking is the predominant activity on the Internet, (ii) lurking depends upon topical genre and age of information, and (iii) lurking leads to a heavy tailed propagation of information. To the best of our knowledge, this is the ﬁrst large-scale experiment to measure and model lurking on the Internet.
K.P. (Suba) Subbalakshmi is an Associate Professor in the Department of Electrical and Computer Engineering (ECE) at Stevens Institute of Technology and Co-Director of the Information Networks and Security (iFINITY) research laboratory. She is a Co-Founder of Dynamic Spectrum, LLC---a startup offering cloud-enabled cognitive radio technologies for various markets including consumer communications, public safety, and the DoD; and Jaasuz.com that provides a suite of advanced text forensics technologies. Her research interests include: Cognitive radio networks security – attacks, counter measures and systems level impact; Internet media security — text forensics, image/video steganography and steganalysis; Social media and Internet data mining.
She is the Vice-Chair North America region of IEEE Technical Committee on Cognitive Networks. Two of her papers are featured in the IEEE COMSOC Best Reading Topics on Cognitive Radio, 2012 as well as one in Wireless Spectrum Research and Development (WSRD), Senior Steering Group (SSG) Suggested Readings, 2011. She has given several keynote addresses, plenary talks, tutorials and served as panelist on DSA security at several international conferences. Her work on psycholinguistic text analytics has been highlighted in several news media outlets including: New Scientist, ACM Tech News etc. Her work is funded by NSF, various DoD agencies, NIJ etc.