There are thousands of Mobile applications on Google Play store and Apple app store. Whenever a mobile user performs a search for a particular type of app, they are presented with a series of similar apps mostly in the descending order of the ratings and number of reviews. Whereas there are various other factors that can categorize the apps as most suitable for the user to increase user satisfaction. The various features that can uplift the quality of the app can be number of installs, content rating, last updated, Current version of the app and so on. For this purpose, real-world Google Play store apps dataset and Mobile App Store dataset from Kaggle is used in this paper to identify the importance of these factors. For identification of important variables, Random Forest, Linear Regression Model, K-Nearest Neighbors and Support Vector Regression are used. The performance of the model is evaluated using standard performance evaluation techniques – Mean Squared Error, Mean Absolute Error and Root Mean Squared Error. The results show that some factors have higher significance and influence the app ratings. The project also focuses on determining the profitable App Profiles for the App Store and Google Play store markets. The purpose of this project is to define the kind of applications that are likely to attract higher number of users. Keywords: Apple app store, Google Play, Random Forest Classifier, Logistic Regression, K-Nearest Neighbor
Zenodo: https://zenodo.org/record/4739437#.YJLubpNKhQI
Github.io : https://mishkink.github.io/viz/