100 likes | 288 Views
Automatic Product Profiling via NLP. Jiho Han Ronny ( Dowon ) Ko. Automatic Product Profiling via NLP. Objective: automatically generate the summary of review extracting the strength/weakness of the product Use NLP techniques to predict ratings Similar to sentimental analysis
E N D
Automatic Product Profiling via NLP Jiho Han Ronny (Dowon) Ko
Automatic Product Profiling via NLP • Objective: automatically generate the summary of review extracting the strength/weakness of the product • Use NLP techniques to predict ratings • Similar to sentimental analysis • Key Insight: Imposing market structure assumption • Different type of information extraction • Amazon review text
Overview • Opinion = (orientation, polarity) Review Texts Orientation Profile Rating m ∞ k
Steps (cont’d) • Parsing – through Stanford NLP syntax parser • Initializing orientation and polarity • Selecting polarity words through decision tree (Max-Ent) • Orientation using N-gram (uni + bi) • Use wordnet when testing • Extract market profiling and pricing kernel • Update word polarity • Repeat until no more improvement
Initializing Decision Tree • Extract the words that have significant effect on rating (in terms of maximizing entropy)
Resul • Initial word polarity
Result (cont’d) • Change in polarity • Performance