Google Maps will be used as an external API in order to mark locations on a map for display in the UI and it is an extensive and well-known framework. Tensorflow will be used for Machine Learning as it is open-source, customizable to different types of machine learning algorithms and lets you serve your model with a REST API. Tensorflow also has a lot of support and documentation which makes it easier for to start with it. Tensorflow is also written with Python. Python is easy to write in, efficient and commonly used in ML applications. In relation to Python, SnipsNLU (not shown on stackshare) will also be used in order to easily train NLU models. PredictHQ (not shown on stackshare) will be used for event data and has an easily accessible API. Twitter API will also be used in order to collect social media data and there are many endpoints it currently offers to query tweets in various ways (stackshare doesn't show this utility in the stack yet)