Task Engine: A User-centric Service

Gordon Sun,Tencent Technology, China.

Abstract

When a user starts a search he or she usually has a task in mind. The normal web search engine is meant to retrieve relevant information for the user, but not for the fulfillment of the user's ultimate task. In this talk, I will propose a new product concept – “task engine”, with its conceptual model and functional characteristics.  An effective task engine can act as user's personal online assistant for various tasks and generate greater user loyalty and business revenue in the Internet industry. I start the talk with an analysis of query logs from the SOSO commercial search engine, categorize query sessions into various types of user task. I will describe the research work and product development at Tencent for several tasks, our successes and the technical challenges we had. Although a task engine could be partly implemented using the current web search engine infrastructure, two major limitations of the latter need to be addressed for a more appropriate implementation: “single query box input” and “one-shot search process”. Our experience showed that in addition to search engine, a task engine should also involve decision engine, recommendation engine and various data mining engine.

Bio

Dr. Gordon Sun (孙国政)Chief Scientist, Tencent Technology, China.

He is leading the next generation search technology program at Tencent, serving as the General Manager of Research department of Search Technology. Dr. Sun has been working on algorithmic search technology since 1998 when he joined Inktomi (the leading US search engine company during 1990s) as the senior scientist and architect. He joined Yahoo early 2004 where he was leading the Global Search Relevance team as the director of research until 2009 when he joined Tencent and moved to China. Gordon received his Ph.D. in theoretical physics from University of Iowa, 1984. He worked as the Research faculty in University of Maryland for 9 years before he moved to Silicon Valley, 1993. He has broad work experiences, knowledge and publications in neural networks, pattern recognition, machine learning, data mining, information retrieval, speech recognition, hand-writing recognition and non-linear dynamics.