Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Combines information retrieval and natural language processing Answer questions (posed in natural language)
Retrieve answers
No documents
No passages
More Data
Other QA groups:
Higher Accuracy
Web More likely to find answers in simple relation to the question. Less likely to deal with NLP systems difficulties.
1. ______ killed Abraham Lincoln. 2. Abraham Lincoln was killed by ______ etc.
Rewrite Query
Rewrite Query
High precision query Abraham Lincoln was born on more likely to be correct
Rewrite Query
W1 IS W2 W3 . Wn W1 W2 IS W3.Wn etc.
Final rewrite: ANDing of non-stop words Louvre Museum AND located Stop Words ( in, the, etc..) Important indicators of likely answers
N-Gram Mining
N-Grams
1-,2-,3-grams are extracted from the summaries Scored based on the weight of the query rewrite that retrieved them Scores summed Final score based on rewrite rules and number of unique summaries in which it occurred
N-Gram Filtering/Reweighting
Query assigned one of seven question types (who, what, how many, etc..) System determines what filters to apply
Boost score of a potential answer Remove strings from the candidate list
N-Gram Tiling
0.450 0.186
0.187
128
0.256
0.355
0.383
227
243
0.454
0.486
Combined QA results