Automatic Classification of Non-Functional Requirements from Augmented App User Reviews
The DOI matches its Crossref record; 3 of the 3 references it cites were found in a registry or an index.
Abstract
Context: The leading App distribution platforms, Apple App Store, Google Play, and Windows Phone Store, have over 4 million Apps. Research shows that user reviews contain abundant useful information which may help developers to improve their Apps. Extracting and considering Non-Functional Requirements (NFRs), which describe a set of quality attributes wanted for an App and are hidden in user reviews, can help developers to deliver a product which meets users' expectations. Objective: Developers need to be aware of the NFRs from massive user reviews during software maintenance and evolution. Automatic user reviews classification based on an NFR standard provides a feasible way to achieve this goal. Method: In this paper, user reviews were automatically classified into four types of NFRs (reliability, usability, portability, and performance), Functional Requirements (FRs), and Others. We combined four classification techniques BoW, TF-IDF, CHI2, and AUR-BoW (proposed in this work) with three machine learning algorithms Naive Bayes, J48, and Bagging to classify user reviews. We conducted experiments to compare the F-measures of the classification results through all the combinations of the techniques and algorithms. Results: We found that the combination of AUR-BoW with Bagging achieves the best result (a precision of 71.4%, a recall of 72.3%, and an F-measure of 71.8%) among all the combinations. Conclusion: Our finding shows that augmented user reviews can lead to better classification results, and the machine learning algorithm Bagging is more suitable for NFRs classification from user reviews than Naïve Bayes and J48.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Efficient Estimation of Word Representations in Vector Space
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1301.3781 · Cited by 26,568 · cited 1×
“A recent study shows that word2vec provides the state-of-the-art performance for measuring words similarity [11].”
BackgroundUnderstanding bag-of-words model: a statistical framework
Zhang, Y., Jin, R., & Zhou, Z.-H. (2010) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-010-0001-0 · Cited by 1,384 · cited 1×
“The Bag-of-Words (BoW) model [13] is widely used to represent textual documents in information retrieval systems and is one of the most popular representation techniques for object classification.”
MethodA Survey of App Store Analysis for Software Engineering
Martin, W., Sarro, F., Jia, Y., Zhang, Y., & Harman, M. (2017) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2016.2630689 · Cited by 459 · cited 1×
“In recent years, App Store analysis has become a popular topic in software engineering [28], and users can easily submit their feedback about the Apps they used in App platforms.”
Background
Cited by
Automatic Identification of Decisions from the Hibernate Developer Mailing List
Li, X., Liang, P., & Li, Z. (2020) · Proceedings of the Evaluation and Assessment in Software Engineering · doi:10.1145/3383219.3383225 · Cited by 21
“For instance, SVM performs best when identifying assumptions from mailing lists [8], while NB gets the best results when classifying user reviews into three types (i.e., NFRs, FRs, and Others) in [18].”
MethodSelf-citation
Metadata
- Authors
- Lu, M., & Liang, P.
- Venue
- Proceedings of the 21st International Conference on Evaluation and Assessment in Software Engineering
- Publisher
- ACM
- Published
- 2017-06-15
- Pages
- 344-353
- OpenAlex
- W2622918375
- Cited by
- 177
- References
- 3
- FWCI
- 9.32
- Type
- conference paper
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.