Feature Adaptive Developmental Mechanisms for Mobile Apps Recommendations System using the Nearest Centroid Classification Algorithm
G. Twinkle Geojini, M. Ramnath, C. Yesubai Rubavathi · 2023
Reviews and ratings play an important part in modern technology since they provide insights that may be used to enhance the functionality and performance of apps. Some customers give app reviews that don't provide developers with useful feedback for improving the app. Developers can't fix bugs or release new versions of their software if they don't catch them as soon as possible. With this method, we extract the most helpful user comments from existing app evaluations and compare them to the apps' overall star ratings. We present a Nearest Centroid Classification (NCC) approach based on supervised machine learning with the goal of spotting new app problems via review analysis. To successfully illustrate resilience to the extreme values that are beyond the range, the suggested framework contains mean vectors of k to evaluate both distance closeness and geographical extent of k-neighbors in each class. After a long time of waiting, developers may at last quickly correct bugs and fine-tune the app's features to enhance the user experience.