PhD, Information and Library Science, North Carolina
Office HoursEfron is on sabbatical during Academic Year 2015-2016.
Areas of Research
Information retrieval in emerging domains such as social media and large collections of digitized books; temporal (diachronic) issues in information retrieval; human interactions with information search and retrieval systems.
Other Professional AppointmentsDepartment Affiliate, Computer Science
Miles Efron is an associate professor at GSLIS who also holds a courtesy appointment in the Department of Computer Science. His research focuses on information retrieval, particularly in emerging domains such as social media and massive repositories of digitized books. He is currently working on information filtering problems, with special emphasis on applying unsupervised and semi-supervised statistical learning to filtering-related tasks. In 2003, He received his PhD from the University of North Carolina in 2003, where he held a post-doctoral appointment for a year following his doctoral work. He earned an MSIS from the School of Information and Library Science at UNC in 2000 and a BA from Occidental College (summa cum laude) in 1994.
Microblogging services like Twitter are becoming an important part of how many people manage information in their day to day activities. As microblog traffic increases (Twitter currently sees about 50 million tweets per day) information management and organization will become keen problems in this area. The project will define the core problems in microblog search and propose solutions to these challenges in the form of both theoretical models and prototype search systems.
In order for older texts to be searchable, contemporary English needs to be translated into language from various historical timeframes. The project will develop software that will let people enter a query in contemporary English, and search over English texts throughout history—from Medieval times to the present day. The project will mostly involve training statistical models that assign probabilities of the translation to a word or phrase in a target English language.
Time affects information retrieval in many ways. Collections of documents change as new items are indexed. The content of documents themselves may change. Users submit queries at particular moments in time. And perhaps most importantly, people’s assessment of a document’s relevance to a query is often time-dependent. For example, searchers of news archives might seek information on a past event where relevant documents cluster in a window of time. Users of social media services such as...
This project aims to improve search engine effectiveness by using knowledge base (KB) entries to inform query expansion. While the intersection of KBs and information retrieval (IR) is a growing research area, this project proposes a novel approach to KB-based query modeling. In particular, this project proposes to let the structure that KB authors impose within individual KB entries guide the final query model. For instance, authors of Wikipedia pages divide individual entries into sections...