Approaches for Prioritizing Feature Improvements Extracted from App Reviews

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)?

To extract the emotions in a given sentence, the authors consider words conveying either sadness, anger, or fear, which they consider relevant for their analysis. This is done by looking at words in emotion lists obtained from LIWC dictionary (https://liwc.wpengine.com/), and from word lists used in emotion mining in Twitter (Wang, W., Chen, L., Thirunarayan, K. and Sheth, A. P. Harnessing Twitter "Big Data" for Automatic Emotion Identification. In Proc. of PASSAT, 2012, 587-592.)

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.

LIWC dictionary: https://liwc.wpengine.com/

4. What is the main goal of the whole study?

Defining automatic approaches for app reviews' analysis to help developers prioritizing feature development for app improvements.

5. What the researchers want to achieve by applying the technique(s) (e.g., calculate the sentiment polarity of app reviews)?

Identify the emotions conveyed by a review, with special focus on three negative emotions (namely, anger, sadness, and fear).

6. Which dataset(s) the technique is applied on?

4,442 reviews for the MyTracks app from the Google Play Store using Google API. Such reviews were preprocessed to obtain sentences from the raw documents. This pre-processing also included filtering of sentences that were not considered relevant for the study (i.e., they were part of reviews with low rates). This resulted in 1,271 sentences (out of a total of 8,623 sentences from 4,442 reviews), which were used for further analyses.

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, LIWC is a general-purpose lexicon/tool.

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?

NA

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?

The emotion extraction method is not evaluated. The authors only evaluate the three approaches they propose for feature prioritization (emotions are among the features).

12. Write down any other comments/notes here.

The proposed approach for emotion mining simply consists in the task of looking up a dictionary to decide if words in a text convey any of the relevant emotions. I am not sure if this is enough for answering yes to question 2.