"What Parts of Your Apps Are Loved by Users?"
1. Does the paper propose a new opinion mining approach?
Yes
2. Which opinion mining techniques are used (list all of them, clearly stating their name/reference)?
SURMiner (Software User Review Miner): new tool proposed in the paper - Max Entropy classifier: review classification - pattern-based parser: to determine aspect-opinion pairs from the sentence patterns, implemented as a sequence of cascading finite state machines - Stanford CoreNLP
3. Which opinion mining approaches in the paper are publicly available? Write down their name and links. If no approach is publicly available, leave it blank or None.
No
4. What is the main goal of the whole study?
To propose a framework which can summarize users' sentiments and opinions toward corresponding software aspects
5. What the researchers want to achieve by applying the technique(s) (e.g., calculate the sentiment polarity of app reviews)?
SURMiner: classifies app reviews into five categories and extracts aspects in sentences which include evaluation of aspect using a pattern-based parser
6. Which dataset(s) the technique is applied on?
recent user reviews of 17 Android apps
7. Is/Are the dataset(s) publicly available online? If yes, please indicate their name and links.
No
8. Is the application context (dataset or application domain) different from that for which the technique was originally designed?
Yes for the Sentiment Analysis part.
9. Is the performance (precision, recall, run-time, etc.) of the technique verified? If yes, how did they verify it and what are the results?
Yes.
1) Review Classification, Aspect-Opinion Extraction, Sentiment Analysis: selected 2000 reviews from each dataset, manually labeled and compared
Results: average F1-scores of 0.75, 0.85 and 0.80
2) compared with state-of-the-art tools: ReviewSpotlight (K. Yatani, M. Novati, A. Trusty, and K. N. Truong. Review spotlight: a user interface for summarizing user-generated reviews using
10. Does the paper replicate the results of previous work? If yes, leave a summary of the findings (confirm/partially confirms/contradicts).
No
11. What success metrics are used?
F1-score and developers' agreement on usefulness
12. Write down any other comments/notes here.
-