MR Profiler
MR Profiler performs author profiling on anonymous texts and predicts the age and gender of the author of the text. MR Profiler uses MapReduce and thus has a very fast processing speed. We have trained it on PAN 2013 author profiling data. There are three classes across age: 10s, 20s and 30s and two classes across gender: male and female. You can try out our author profiling project here.
We also participated in the PAN’14 Author Profiling task and obtained third position. Here, the classification task had five classes for age: 18-24, 25-34, 35-49, 50-64, 65-plus and for gender there are two classes: male and female. To try a demo of our PAN’14 system go here. You can also read the notebook paper here.