Extraction Model
Extraction Model:Answer sentence selection
Due to increasing complexity in question answering, deep learning has become a popular trend in solving difficult problems.
Approaches
Word Count: counts the number of non-stopwords in the question that also occur in the answer sentence.
Weighted Word Count: re-weights the counts by the IDF values of the question words.
LCLR :makes use of rich lexical semantic features, including word/lemma matching, WordNet and vector-space lexical semantic models.
Convolutional Neural Network
Recurrent Neural Network
Comments
Post a Comment