still more summaries

38. Summary #1: Inferring Gender from the Content of Tweets: A Region Specific Example

More and more people are interested in employing social media sites like Twitter to collect real-time data about the attitudes and viewpoints of population in a region. Nevertheless, people’s important points of view and reactions greatly vary based on their gender and ethnicity. On the other hand, the information about people’s gender and ethnicity is not always available on these social media sites. Therefore, automatic methods are needed to deduce these implicit attributes. Fink et al. (2012) presented an automatic method to estimate Twitter users’ gender from their tweet content. A Supervised machine learning technique, in which features of training a classifier were extracted from tweet content, was applied on Twitter users in Nigeria, Africa. As a result, tweet content could be a good indicator to predict people’s gender because 80% of estimations are correct. Moreover, it was found that gentlemen and women have some differences in topic and emotion. Finally, the proposed method was argued to provide us an opportunity to know who is using social media. (FL)

Fink, C. et al. (2012). Inferring Gender from the Content of Tweets: A Region Specific Example. In Proceedings of the Sixth International AAAI Conference on Weblogs and Social Media (Dublin, Ireland, June 4-7, 2012). Available from: http://www.aaai.org/ocs/index.php/ICWSM/ICWSM12/paper/viewFile/4644%26lt%3B/5031 (Accessed date: Sept. 22, 2014)

37. Summary #2: Understanding the Demographics of Twitter Users

In every instance, Twitter records millions of people’s ideas and feelings around the world in the form of tweets that are less than 140 characters. However, who is using Twitter? Does the Twitter population represent the entire population? These questions still can not be answered, because we do not understand the Twitter population well, although Twitter was considered to have a huge potential. In this paper, Misolve et al. (2011) for the first time tried to analyze Twitter user data that represents greater than 1% of the U.S.A population to answer these questions. Techniques were developed to tell the differences between the Twitter population and the U.S.A population in gender, geography and ethnicity/race. Moreover, it was found that the Twitter population is a highly uneven sample of the U.S.A population. (FL)

Mislove, A. et al. (2011). Understanding the Demographics of Twitter Users. In Proceedings of the Fifth International AAAI Conference on Weblogs and Social Media (Barcelona, Spain, July 17-21, 2011). Available from: http://www.aaai.org/ocs/index.php/ICWSM/ICWSM11/paper/viewFile/2816/3234 (Accessed date: Sept. 22, 2014)