A Lightweight Vision-Language Fusion Framework for Predicting App Ratings from User Interfaces and Metadata

TL;DR

This paper introduces a lightweight multimodal framework combining visual UI features and semantic data to accurately predict app ratings, improving over models that use only one data type.

Contribution

It presents a novel fusion of visual and textual features using MobileNetV3, DistilBERT, and a gated fusion module for app rating prediction, optimized for edge deployment.

Findings

Achieved low MAE of 0.1060 in rating prediction

Demonstrated high correlation coefficient of 0.9251

Validated effectiveness through extensive ablation studies

Abstract

App ratings are among the most significant indicators of the quality, usability, and overall user satisfaction of mobile applications. However, existing app rating prediction models are largely limited to textual data or user interface (UI) features, overlooking the importance of jointly leveraging UI and semantic information. To address these limitations, this study proposes a lightweight vision--language framework that integrates both mobile UI and semantic information for app rating prediction. The framework combines MobileNetV3 to extract visual features from UI layouts and DistilBERT to extract textual features. These multimodal features are fused through a gated fusion module with Swish activations, followed by a multilayer perceptron (MLP) regression head. The proposed model is evaluated using mean absolute error (MAE), root mean square error (RMSE), mean squared error (MSE), coefficient of determination (R2), and Pearson correlation. After training for 20 epochs, the model achieves an MAE of 0.1060, an RMSE of 0.1433, an MSE of 0.0205, an R2 of 0.8529, and a Pearson correlation of 0.9251. Extensive ablation studies further demonstrate the effectiveness of different combinations of visual and textual encoders. Overall, the proposed lightweight framework provides valuable insights for developers and end users, supports sustainable app development, and enables efficient deployment on edge devices.