Data Visualization of Google Play store apps using web scrapped data for analyzing the android market
Android operating system is one of the most popular operating system in the world. As a result, mobile app distribution platform such as Google play store is getting flooded due to daily increase in the number of new applications. This has led to immense pressure on the app developers due to increase in the competition all over the globe. To survive this competition, it is important for the developers to study the information about apps concerning their users, important app features and business focused attributes. User information contains details related to ratings and reviews accorded to the app by the users who have already used it. This can provide quantitative as well as qualitative data about the users’ perception regarding the app. Significant app features can contribute in app success and its popularity. And business information concerns number of downloads of the app. The contribution of this thesis is threefold (1) Computation of overall app rating by averaging ordinal star ratings and sentiments scores extracted by performing sentiment analysis on user reviews. This technique helps us extract meaningful content present in the user reviews. (2) To determine important app features affecting app’s success. This is performed using Exploratory data analysis, a visualization technique used to highlight the vital aspects of the analyzed data. (3) Final step is to predict number of downloads of the app using various prediction models such as Linear Regression, K-Nearest Neighbor, Decision Tree, Random Forest, Naïve Bayes, and Artificial Neural network (ANN). This prediction of number of downloads can facilitate developers to allocate resources more efficiently.