Sentiment Analysis of User-Generated Social Media Content
A Case Study and Qualitative Analysis of Simple Cross-Lingual Opinion Mining
We propose a simple, modern, and effective method for building a single topic model with sentimentanalysis capable of covering multiple languages simultanteously, based on a pre-trained state-of-the-art deep neural network for natural language understanding.To demonstrate its feasibility, we apply the model to newspaperarticles and user comments of a specific domain, i.e., organic food productsand related consumption behavior.The themes match across languages.We obtain an high proportion of stable and domain-relevant topics, a meaningful relation between topics and their respective textual textual contents, and an interpretable representation for social media documents.
