This task needs to download glove.6B.300d.txt at [GloVe: Global Vectors for Word Representation] (https://nlp.stanford.edu/projects/glove/).
This task is now only use English corpus, follow-up will be based on Chinese corpus for emotional analysis.