This plot shows the monthly sum of tweets that are sake related. Several hashtags were used to pull data from Twitter. The data have been processed to remove a large number of false hits (e.g. muslim users using #sake to indicate 'for the sake of Allah'). The data are from all of Twitter and have not been processed based on language.

This map projects the number of #saké tweeters across the world. Pop ups indicate the country and the total number of tweeters using the hashtag.

This map shows the locations of users who tweeted about #sake, the rice wine beverage. Note that this doesn't take into account the number of tweets, just the user accounts using the hashtag. Each point is partly transparent, so a high density of users adds up to a dark field on the map.

Google Trends was accessed July 2nd 2019 to obtain the data for this graphic. "sake near me" was the applied search term. Using "sake" alone produces a more dispersed signal and it is also a more ambiguous search term---is the person interested in how sake is made or interested in purchasing sake? Louisiana and Florida appear to have a large interest in sake purchases, over the past five years, with South Carolina and North Carolina close by. Some caveats: 1) This plot is an aggregate of five years and doesn't say what is happening right now, 2) Many states have too few related searches and are zeroed out, 3) I'm assuming that when each person searched for sake nearby they intended to purchase sake.

This is a rudimentary analysis of homicide victims in Baltimore. I do not claim to know how to solve this problem. My document simply highlights, as many others have, that there is a problem.

For final project of Coursera data science course.

First submission for Data Science Capstone Project

Hopkins Data Science assignment