Data Visualisation Experiments

Earthquake Start of a project to bring in information from a website with an Application Programming Interface (API). We are going to use a modified version of "Project: Fetching Current Weather Data" from "Automate the boring stuff with Python" by Al Sweigart What is going on below? We import libraries to dela with json, reuest from the server and the pandas library. In [1]: import json , requests import pandas as pd from pandas import json_normalize In this section we creating a string made up of the URL. Requesting the information from the site with the URL we created and pass back the information. Data comes from the US Geological survey  https://www.usgs.gov/about/about-us/who-we-are and one of their earthquake feeds. Then print out what was returned. In [2]: url = 'https://earthquake.usgs.gov/earthquakes/feed/v1.0/summary/all_hour.geojson' response = requests . get ( url ) response . raise_for_status () Now we need load the data which i

In a previous post ( http://bit.ly/TAGS2GEPHI  ) I look at moving data from a TAGS ( https://tags.hawksey.info/ ) spreadsheet collecting tweets into a network visualisation tool (GEPHI) to plot a network of interactions based on how replied to whom. We left it at a graph like the one below with the thick the line the stronger/more replies there were. Still going to use the same data but look at some of the other options to visualise these relationship. Tools like Gephi offer many ways to layout the plots The one below is the ForceAtlas2 layout is good for showing groupings for this problem: One I like because it is attractive and I feel shows the number of nodes and their interactions is the Fructerman Reingold (shown below) One of the features of Gephi is you can add plug-ins; including new layouts. Two I am enjoying playing with is the Circular layout Also the Dual circular layout - where you set two circles based on some criterian in this case number of interactions - the outer ring

All views are the authors and do not reflect the views of any organisations the author is associated with. Twitter: @scottturneruon

Recorded with https://screencast-o-matic.com Visualising the behaviour on Twitter for a particular hashtag in this case #lthechat All views are the authors and do not reflect the views of any organisations the author is associated with. Twitter: @scottturneruon

Recorded with https://screencast-o-matic.com Video looks at the using NodeXL https://nodexl.com/ to look at tweets. The professional version was used but a free version is available. Uploaded the graph produced to the NodeXL Graph Gallery and so the hidden gems other than the graph that is produced. The links for downloading settings, etc used in a particular graph are at the bottom of the page linked out from the gallery. All views are the authors and do not reflect the views of any organisations the author is associated with. Twitter: @scottturneruon