Identifying Ideological Perspectives of Web Videos using Patterns Emerging from Folksonomies

We are developing a classfier that can automatically identify a web video's ideological perspective on a political or social issue (e.g., pro-life or pro-choice on the abortion issue). The problem has received little attention, possibly due to inherent difficulties in content-based approaches. We propose to develop such a classfier based on the pattern of tags emerging from folksonomies. The experimental results are positive and encouraging.