Improving Feature Representation Based on a Neural Network for Author Profiling in Social Media Texts

Table 15

Obtained results (accuracy, %) for age and gender classification using SVM classifier on the PAN author profiling 2016 English training corpus under 10-fold cross-validation with vectors extracted only from the training data.

Feature set	Age		Gender
	SVM-NP	SVM-WP	SVM-NP	SVM-WP
D2V (1-gram)	41.89	43.68	75.05	73.27
D2V (1 + 2-grams)	43.06	44.84	76.75	78.67
D2V (1 + 2 + 3-grams)	42.83	47.71	75.77	78.43