Transfer Learning has emerged as one of the main image classification techniques for reusing architectures and weights trained on big datasets so as to improve small and specific classification tasks. In Natural Language Processing, a similar effect is obtained by reusing and transferring a language model. In particular, the Universal Language Fine-Tuning (ULMFiT) algorithm has proven to have an impressive performance on several English text classification tasks. In this paper, we aim at improving current state-of-the-art algorithms for Spanish Sentiment Analysis of short texts. In order to do so, we have adapted a ULMFiT algorithm to this setting. Experimental results on benchmark datasets show the potential of our approach. © Springer Nature Switzerland AG 2020.

In the last years, innovative techniques like Transfer Learning have impacted strongly in Natural Language Processing, increasing massively the state-of-the-art in several challenging tasks. In particular, the Universal Language Model Fine-Tuning (ULMFiT) algorithm has proven to have an impressive performance on several English text classification tasks. In this paper, we aim at developing an algorithm for Spanish Sentiment Analysis of short texts that is comparable to the state-of-the-art. In order to do so, we have adapted the ULMFiT algorithm to this setting. Experimental results on benchmark datasets (InterTASS 2017 and InterTASS 2018) show how this simple transfer learning approach performs well when compared to fancy deep learning techniques. © Springer Nature Switzerland AG 2019.