Browsing by Subject "Social computing"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Collaborative curation in social production communities.(2012-08) Lam, Shyong (Tony) K.The use of social production communities (SPCs) has become a common approach for building information repositories such as Wikipedia, Yahoo! Answers, and YouTube. In these systems, communities of users collaborate to produce a shared repository of information. We define collaborative curation as the tasks performed by these communities, and the processes, workflows, and policies that guide how users work together. This thesis seeks to study the implications of different curation processes, and the challenges that SPCs face in constructing information repositories. Our goal is to better understand the growth and evolution of SPC information repositories so that we can inform the design of SPCs. The first part of this thesis focuses on collaborative curation practices at a high level to learn about the design space of curation mechanisms and the impact that different mechanisms have on the evolution of SPCs. We begin with an analysis ofWikipedia’s curation practices, studying how Wikipedia’s editors decide which articles merit inclusion in the encyclopedia, and how the encyclopedia has grown over the years. We then conduct a user study using the MovieLens recommender system to compare two typical curation mechanisms – a wiki-like process, and a social voting process – in how they affect the growth of MovieLens’ movie database. In the second part of this thesis, our focus shifts to challenges that SPCs face in collaborative curation. We start by looking at how skews in group composition can influence collaborative curation. SPCs typically rely on the efforts of self-formed and self-organized volunteer groups. Such groups may differ from the larger user community or from the general populace on multiple dimensions, including demographics, attitudes, and experience. We conduct two studies to study these differences in the context of Wikipedia. At the small-scale level, we examine how composition skews in small working groups can affect curation decision quality; at the large-scale level, we explore an apparent gender disparity amongst Wikipedia’s community of editors. We close with an analysis of a type of malicious deviant behavior where users submit false data to an SPC in an attempt to manipulate choices made by fellow users.Item Intelligent tagging systems: machine learning for novel systems: machine learning for novel.(2012-04) Vig, JesseThe Web puts a vast repository of information at users' fingertips, but the size and complexity of this information space can easily overwhelm users. Recommender systems and tagging systems represent two very different approaches to addressing this information overload. Recommender systems use machine learning and statistical models to automatically retrieve the items of most interest to a particular user. Tagging systems leverage the community's collective knowledge to help users explore the information space themselves. While both approaches can be very effective, they each have limitations. Recommender systems require little effort from users, but they leave users with little control over the recommendation process. Tagging systems put control in the hands of the user, but -- because tags are applied by humans -- tagging systems often suffer from issues of tag sparsity. This thesis explores intelligent tagging systems that combine the machine intelligence of recommender systems with the user control and comprehensibility of tagging systems. We first present Tagsplanations, tag-based explanations that help users understand why an item was recommended to them. We then introduce the Tag Genome, a novel data structure that uses machine learning to augment human judgments of the relationships between tags and items. Next we discuss Movie Tuner, a conversational recommender system based on the Tag Genome that enables users to provide multifaceted feedback using tags. For each system, we outline the design space of the problem and discuss our design decisions. We evaluate each system using both offline analyses as well as field studies involving thousands of users from MovieLens, a movie recommender system that also supports tagging of movies. Finally, we draw conclusions for the broader space of related applications.