TV Answers - Using the Wisdom of Crowds to Facilitate Searches with Rich Media Context

Abstract

Television has always been a popular entertainment medium with considerable impact on consumer purchase behaviors. However, unlike mobiles and PCs, it has yet to support an explicit search capability that extends beyond simple content navigation. A core reason for this is lack of usable input interfaces for TV coupled with the difficulty of creating and executing queries based on rich media context. In this paper, we present TV Answers -a system that combines a novel context capture capability (Freeze-Frame) with an intelligent web services mediator (Edge Proxy) to "crowd-source" user-generated queries. We describe the design and implementation of an early prototype of TV Answers and highlight the challenges and opportunities presented by such systems. Our goal is to demonstrate the viability of using "social search" as a complement to existing algorithm-driven search solutions for rich media consumption.

Wisdom of artificial crowds algorithm for solving NP-hard problems

“…The original WoC concept was introduced by James Surowiecki in 2004 (Surowiecki, 2004). It highlights the aggregation of information in groups, resulting in decisions that are often better than could have been made by any single member of the group (Narasimhan et al, 2010;Osorio and Whitney, 2005;Kostakos, 2009).…”

CollecTV intelligence: A 3-screen 'social search' system for TV and video queries

“…A first version of that system is described in more detail in our earlier paper [8]. In this paper, we will simply highlight some key features; the figure below shows a version of our query and response UI for this experience on the set-top box.…”