This paper presents our approach to the Task 5 of Semeval-2019, which aims at detecting hate speech against immigrants and women in Twitter. The task consists of two subtasks, in Spanish and English: (A) detection of hate speech and (B) classification of hateful tweets as aggressive or not, and identification of the target harassed as individual or group. We used linguistically motivated features and several types of n-grams (words, characters, functional words, punctuation symbols, POS, among others). For task A, we trained a Support Vector Machine using a combinatorial framework, whereas for task B we followed a multi-labeled approach using the Random Forest classifier. Our approach achieved the highest F1-score in sub-task A for the Spanish language.

Word embeddings are powerful for many tasks in natural language processing. In this work, we learn word embeddings using weighted graphs from word association norms (WAN) with the node2vec algorithm. Although building WAN is a difficult and time-consuming task, training the vectors from these resources is a fast and efficient process. This allows us to obtain good quality word embeddings from small corpora. We evaluate our word vectors in two ways: intrinsic and extrinsic. The intrinsic evaluation was performed with several word similarity benchmarks, WordSim-353, MC30, MTurk-287, MEN-TR-3k, SimLex-999, MTurk-771 and RG-65, and different similarity measures achieving better results than those obtained with word2vec, GloVe, and fastText, trained on a huge corpus. The extrinsic evaluation was done by measuring the quality of sentence embeddings using transfer tasks: sentiment analysis, paraphrase detection, natural language inference, and semantic textual similarity.

No abstract