We've already started to implement some visualization ideas on the Google Playstore Dataset in previous posts on this blog, however we never included a broader view of the dataset necessary to understand what we are implementing. This post aims to do that showing the variables available in this dataset and its content.
Firstly, we may say that this dataset contains 9,660 datapoints, each one corresponding
to an App downloaded from Google Play Store. This data was scraped from Google Play Store in August 2018 and it's worth to mention than the apps showed in the list depends on the geographic location of the user scraping the data (more information about the dataset in https://bit.ly/2PSvQRh). The fields or dimensions available for each observation
are 13, and a brief description of them is included in Table 1 and Table 2.
Many question arise regarding to the content available in this dataset, but I aimed my interest to the main question in a broader sense that we could ask with this dataset if we are in the Apps Busines: What features should my app have?. In other words, if I'm an app developer I want to know what features have the most installed and better rated apps. So this is the question we'are looking to answer through visualizations.
In addition, I would like to mention that the exploration of the dataset to implement our visualitazions idea has also resulted in discovering some issues regarding our dataset. For example the variable size contains numeric data for the most of the datapoints, however an important percentage of them (16%) contains the string 'Varies with device', which is a non-informative value for this field. This kind of issues could end up decreasing the number of effective datapoints in our dataset, however this information will be considered when we present our final visualization project.
