Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 1708.04968

Fault in your stars: An Analysis of Android App Reviews

Rahul Aralikatte·Giriprasad Sridhara·Neelamadhav Gantayat·Senthil Mani
View original source ↗Read source document ↗Export metadata ↓

Abstract

Mobile app distribution platforms such as Google Play Store allow users to share their feedback about downloaded apps in the form of a review comment and a corresponding star rating. Typically, the star rating ranges from one to five stars, with one star denoting a high sense of dissatisfaction with the app and five stars denoting a high sense of satisfaction. Unfortunately, due to a variety of reasons, often the star rating provided by a user is inconsistent with the opinion expressed in the review. For example, consider the following review for the Facebook App on Android; "Awesome App". One would reasonably expect the rating for this review to be five stars, but the actual rating is one star! Such inconsistent ratings can lead to a deflated (or inflated) overall average rating of an app which can affect user downloads, as typically users look at the average star ratings while making a decision on downloading an app. Also, the app developers receive a biased feedback about the application that does not represent ground reality. This is especially significant for small apps with a few thousand downloads as even a small number of mismatched reviews can bring down the average rating drastically. In this paper, we conducted a study on this review-rating mismatch problem. We manually examined 8600 reviews from 10 popular Android apps and found that 20% of the ratings in our dataset were inconsistent with the review. Further, we developed three systems; two of which were based on traditional machine learning and one on deep learning to automatically identify reviews whose rating did not match with the opinion expressed in the review. Our deep learning system performed the best and had an accuracy of 92% in identifying the correct star rating to be associated with a given review.

Explore related subjects

cs.LGcs.CL

Keep this discovery

BibTeXRIS

Rahul Aralikatte, Giriprasad Sridhara, Neelamadhav Gantayat, Senthil Mani. 2017-08-16. Fault in your stars: An Analysis of Android App Reviews. https://doi.org/10.1145/3152494.3152500

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

AUC Maximization from Biased Positive-unlabeled Data with Confidence

Maximizing the area under the receiver operating characteristic curve (AUC) is a standard approach to imbalanced binary classification. Although positive and negative data are required for maximizing the AUC, negative data are often difficult to collect in some real-world applications due to privacy concerns or the need for specialized expertise to annotate them. Thus, AUC maximization from positive and unlabeled (PU) data has been attracting attention. Existing methods assume that labeled positive data are unbiased samples from the true positive distribution. However, this ideal assumption is often violated in practice. In this paper, we propose a method to maximize the AUC from biased PU data. To address the bias, our key idea is to exploit {\it confidence}, i.e., the probability that an instance is positive, associated with the small number of labeled positive data. We derive an estimator of the AUC risk using biased PU data with confidence, enabling AUC maximization under such bias. We further show that the rewritten AUC risk induces a Bayes-optimal AUC ranking even when the available confidence is any strictly increasing transformation of the true posterior probability. We experimentally show the effectiveness of our method on eight real-world datasets.

cs.LG↗
arXiv2026

Measuring the Value of World-Model Updates: A Counterfactual Utility Protocol for Continual Adaptation

Continual world models must decide whether new data justify changing the model. Fixed replay schedules and prediction-error triggers specify when to update, but neither reveals the value of an individual update: one deployment run cannot show how the same model would have performed at that moment had it held its parameters. We introduce the fork ledger, which branches a deployment stream at pre-registered decision points into matched update and hold continuations under common random numbers. It evaluates both continuations on the same episodes and records $\Delta R = R_{\mathrm{update}} - R_{\mathrm{hold}}$. Always applying one fixed update mechanism lowers return on all three simulated control tasks: CartPole ($-144.0$; checkpoint-bootstrap $95\%$ CI $[-185.4,-116.1]$, against a converged return near $650$), Walker ($-82.8$; $[-101.1,-61.7]$) and Cheetah ($-18.6$; $[-29.0,-6.6]$). Divergence is an outcome of applying the update, so the estimand counts every attempted fork; restricted to the $693$ of $720$ that did not collapse, CartPole and Walker are unchanged in sign ($-113.4$ and $-82.1$) and Cheetah becomes unresolved ($-3.9$; $[-17.5,+13.0]$). The task is the unit of inference: each contributes $240$ attempted forks over five pretrained checkpoints crossed with two drift directions. The ledger makes counterfactual utility observable for a fixed mechanism, allowing triggers to be judged by the updates they select rather than by surprise detection alone.

cs.LG↗
arXiv2026

When More Is Not Better: Component Anti-Synergy in a P300 Speller

P300 brain-computer interface (BCI) spellers can provide hands-free communication for people with severe motor impairments. Modern pipelines combine multiple individually promising components, often assuming that 'more-is-better'. We tested this assumption using a four-component full-factorial experiment varying the inclusion of Euclidean Alignment (EA), xDAWN spatial filtering, subject calibration, and language model priors on a public P300 dataset. Performance was evaluated using accuracy, repetitions, and information transfer rate (ITR) with mixed-effects models. Results show that the value of components is conditional rather than additive. Calibration was the strongest singular contributor, while EA compensated for its absence in zero-calibration settings. Adding independently useful components could also reduce performance, revealing component anti-synergy. Contrary to conventional wisdom, LM support was not universally beneficial: its effect depends strongly on the strength of the underlying EEG pipeline, while results from a larger LM showed a similar pattern. Together, these findings challenge maximal 'all-on' pipeline design and highlight the value of selecting spatial and language-support components according to the quality of available EEG evidence.

cs.LG↗