Proceedings of the International Conference on Internet Computing and Big Data, Amerika Birleşik Devletleri, 01 Temmuz 2013, cilt.1, no.1, ss.127-133
Sentimental analysis on web-mined data has an increasing impact on most of the studies. Sentimental influence of any content on the web is one of the most curios questions by the content creators and publishers. Also the impact of the web- mined data is completely another issue than the sentiments. For example categorizing a blog post into positive or negative sentiment is a parameter and measuring the like and dislike numbers of the blog post is completely another issue. In this study, the impact and sentimental of the comments collected from five different web sites in Turkish with more than 300,000 comments in total. The web sites are from newspa- pers; movie reviews, e-marketing web site and a literature web site. All the comments are mixed into a single file. The comments have a like or dislike number, which are used as ground proof of the impact of the comment. The ground proof of the sentiment of the blog is the smileys in the post text. Also the sentiments are spread to the rest of the blogs without smileys by using the bag of words. The correlation is imple- mented by using the support vector machine classifier and success rates are 50.7% similarity between the sentiment and the impact of the comment.