Then follow the results (Section 5), and Section 6 concludes the paper. For whom we already know that they are an individual person rather than, say, a husband and wife couple or a board of editors for an official Twitterfeed. the identification of author traits like gender, age and geographical background.
2009) managed to increase the gender recognition quality to 89.2%, using sentence length, 35 non-dictionary words, and 52 slang words.
They used lexical features, and present a very good breakdown of various word types.
When using all user tweets, they reached an accuracy of 88.0%.
We then experimented with several author profiling techniques, namely Support Vector Regression (as provided by LIBSVM; (Chang and Lin 2011)), Linguistic Profiling (LP; (van Halteren 2004)), and Ti MBL (Daelemans et al.
2004), with and without preprocessing the input vectors with Principal Component Analysis (PCA; (Pearson 1901); (Hotelling 1933)).