Using social connection information to improve opinion mining: Identifying negative sentiment about HPV vaccines on Twitter

摘要: The manner in which people preferentially interact with others like themselves suggests that information about social connections may be useful the surveillance of opinions for public health purposes. We examined if connection from tweets human papillomavirus (HPV) vaccines could used to train classifiers identify anti-vaccine opinions. From 42,533 posted between October 2013 and March 2014, 2,098 were sampled at random two investigators independently identified Machine learning methods using first three months data, including content (8,261 text fragments) (10,758 relationships). Connection-based performed similarly content-based on training more consistently than test data subsequent months. most accurate classifier achieved an accuracy 88.6% set, only features. Information how are connected, rather what they write, improving Twitter.