1, information
publication:www2016
Author:julian McAuley
2. What
Learn the information in product reviews, ask questions about items, and automatically give answers: Sort by degree of relevance
3. Dataset
Amazon Product Reviews, descriptions, questions and answers
4. How:
input:1) binary Question:review label:yes,no,eg:is This bike a medium? My daughter is 5 ' 8 | This was a great bike for a tall person. " (yes,. 711)
2) Product query, Top-answer,review rank. "How many hours does it keep hot and cold?" | It doesn ' t, I returned the one I purchased| "Does keep the coffee very hot for several hours."
OUTPUT:1) Review is the probability of a label and LABEL;2) review is the probability of best answer
Method:Bilinear Models,moqa
5, Evaluation:[email protected]
See PPT for details.
Addressing Complex and subjective product-related Queries with Customer reviews-www2016-20160505