Papers › AWARE: Aspect-Based Sentiment Analysis Dataset of Apps Reviews for Requirements Elicitation
AWARE: Aspect-Based Sentiment Analysis Dataset of Apps Reviews for Requirements Elicitation
Nouf Alturaief, Hamoud Aljamaan, Malak Baslyman
The smartphone apps market is growing rapidly which challenges apps owners to continue improving their products and to compete in the market. The analysis of users feedback is a key enabler for improvements as stakeholders can utilize it to gain a broad understanding of the successes and failures of their products as well as those of competitors. That leads to generating evidence-based requirements and enhancing the requirements elicitation activities. Aspect-Based Sentiment Analysis (ABSA) is a branch of Sentiment Analysis that identifies aspects and assigns a sentiment to each aspect. Having the aspect information adds a more accurate understanding of opinions and addresses the limited use of the overall sentiment. However, the ABSA task has not yet been investigated in the context of smartphone apps reviews and requirements elicitation. In this paper, we introduce AWARE as a benchmark dataset of 11323 apps reviews that are annotated with aspect terms, categories, and sentiment. Reviews were collected from three domains: productivity, social networking, and games. We derived the aspect categories for each domain using content analysis and validated them with domain experts in terms of importance, comprehensiveness, overlapping, and granularity level. We crowdsourced the annotations of aspect categories and sentiment polarities and performed quality control procedures. The aspect terms were annotated using a partially automated Natural Language Processing (NLP) approach and validated by annotators, which resulted in 98% correct aspect terms. Lastly, we built machine learning baselines for three tasks, namely (i) aspect term extraction using a POS tagger, (ii) aspect category classification, and (iii) aspect sentiment classification, using both Support Vector Machine (SVM) and Multi-layer Perceptron (MLP) classifiers.
Code
No code repository is listed for this paper in the archive or in Syntology's graph.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Datasets
Introduced by this paper, per the archive.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Aspect Category Detection | AWARE | Baseline | F1-score | 0.32 | #1 of 1 | Archive leaderboard | report |
| Aspect Category Polarity | AWARE | Baseline | Accuracy (%) | 67 | #1 of 1 | Archive leaderboard | report |
| Term Extraction | AWARE | Baseline | F1-Score | 0.82 | #1 of 1 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections