Finding high-quality content in social media

Abstract

The quality of user-generated content varies drastically from excellent to abuse and spam. As the availability of such content increases, the task of identifying high-quality content in sites based on user contributions-social media sitesbecomes increasingly important. Social media in general exhibit a rich variety of information sources: in addition to the content itself, there is a wide array of non-content information available, such as links between items and explicit quality ratings from members of the community. In this paper we investigate methods for exploiting such community feedback to automatically identify high quality content. As a test case, we focus on Yahoo! Answers, a large community question/answering portal that is particularly rich in the amount and types of content and social interactions available in it. We introduce a general classification framework for combining the evidence from different sources of information, that can be tuned automatically for a given social media type and quality definition. In particular, for the community question/answering domain, we show that our system is able to separate high-quality items from the rest with an accuracy close to that of humans.

A probabilistic method for emerging topic tracking in Microblog stream

“…It may just be a bubble of someone's daily life, such as "Good Morning". According to a study of tweets [21], nearly 36% of the rated tweets are worth reading, 25% are not, and 39% are middling. On the other hand, besides textual content, URLs sometimes can provide useful information for judging feed quality.…”

Quality Prediction in Collaborative Platforms: A Generic Approach by Heterogeneous Graphs

“…Campbell et al [5] and more recently Jurczyk et al [12] make use of the HITS algorithm over a users-interaction graph to show a positive correlation between authority and quality. Recent analysis on Stack Overflow [15] and Quora [2,18] underlines the cyclic relation between content quality and producers authority. If this mutual reinforcement principle has been extensively exploited for simple graphs considering a few types of nodes and relations, it seems that no formulation has been proposed for more complex graphs and in particular for heterogeneous graphs.…”

QoS-Based Task Scheduling in Crowdsourcing Environments

“…With their notion of HITs that can be created using a Web servicebased interface they are closely related to our aim of mediating the capabilities of crowds to service-oriented business environments. Despite the fact that AMT offers HITs on various topics [13], the major challenges are to find on request skilled workers that are able to provide high quality results for a particular topic (e.g., see [1]), to avoid spamming, and to recognize low-performers. To the best of our knowledge, these problems are still not faced by AMT.…”