Greetings! Welcome to Venky Rao's blog on Predictive Analytics, Geospatial Analytics and Visualization. This blog aims to present interesting analysis of geospatial data and to de-mystify predictive analytics for the layman. My blog is featured on: http://www.kdnuggets.com/ - Analytics and Data Mining Resources AND http://www.datasciencecentral.com/ - The Online Resource for Big Data Practitioners.
Using data from USGS (https://earthquake.usgs.gov/earthquakes/feed/v1.0/csv.php), I created a 3D web map of all earthquakes that occurred on 3 May 2017 with a magnitude > 1.0 on the Richter scale. Here's a screenshot of my 3D web map:
Working with a collaborator who published a map service that showed the growth of US cities through time (from the year 1790 through 2000), I created an interactive time enabled web app. Here's a screenshot:
Here's my very first 3D Web Scene that visualizes natural disasters and two cities (Portland and Montreal): http://arcg.is/01jbSq
A 3D Web Scene is Esri-speak for a 3D web map. You can zoom and pan, re-orient, change basemaps, change the daylight settings, explore different views and do lots more. Here's a screen shot of all typhoons represented on the 3D Web Scene:
In this screenshot, typhoons are represented as cylinder symbols, with greater heights representing higher wind speeds and darker colors representing lower barometric pressures.
If you don't want to leave my beautiful blog (I don't blame you), you can check out the embedded version of the 3D Web Scene right here:
Using ArcGIS online and some simple instructions from Dr. Pinde Fu of Esri, I re-created a simple web app for selecting restaurant locations in USA. This web app allows users to choose between two competing locations for opening a full service restaurant based on some interesting analytics capabilities like driving distance in time based on historical traffic patterns, the latest demographic information of the locations including population, disposable income, etc.
Here is a screenshot of the results of analysis done on service area of one of the locations based on a 15-minute drive time distance if usual traffic at 6pm on a Friday is taken into account:
I created a simple survey for collecting Volunteered Geographic Information (VGI) using Survey123 for ArcGIS. I found this this tool intuitive and easy to use. It still has some restrictions (I am sure there are better survey tools out there) but I was able to get up and running very quickly. Of course, I love the mapping option that makes this survey tool unique - it instantly maps the location of the VGI.
I have not yet tried the Analyze and Data capabilities of this tool but in order to give it a serious test, I need lots of people to complete my survey. Here is a link to the survey:
When you click on the link (assuming you are using a mobile device), you will be asked to open the survey in the Survey123 for ArcGIS app. Please download this app from your neighborhood app store (it's free and took me less than 30 seconds to download). Once you download the app, filling out the survey will take you less than 60 seconds.
Being the cricket obsessed fan that I am (my obsession has gotten worse since I moved away from India and Australia to the US where I am starved of watching live cricket - distance does make the heart grow fonder!), I worked on a fun project this weekend. I first started with a story map tour of all the cities that have hosted the finals of the various ICC Cricket World Cup tournaments from inception (1975) until the last edition (2015).
Next, I created a map of all the countries that have ever participated in the World Cup. Here is what the map looks like:
The size of the flag indicates the number of times that a country has been the world champion (min value = 0; max value = 5).
Then I focused on the two World Cups (held in 1987 and 2011) that were hosted by Asian countries. For each of these editions, I created a map that marked the various venues where the cricket matches were played. Here is what the map …
Per Wikipedia, Volunteered Geographic Information (VGI) is the harnessing of tools to create, assemble, and disseminate geographic data provided voluntarily by individuals. VGI is a special case of the larger Web phenomenon known as user-generated content. While there is concern over the authority of the data, research has shown that VGI may provide benefits to the end user above and beyond that of traditional data sources, in part due to its ability to collect and present data not collected or curated by traditional/ professional sources. Additionally, VGI has been shown to provide positive emotional value to users, not only in functionality, but also in satisfaction, social connection, and ethics.
I created a simple (and fun!) app to collect VGI on most wanted Police suspects. I used four famous James Bond villains as my suspects (Dr. No, Goldfinger, Jaws and Oddjob) and created a form for any user to record any sightings of these murderous hoodlums.
I created a web app the mapped recent earthquakes in the world based on data from the US Geological Survey (http://earthquake.usgs.gov/earthquakes/feed/v1.0/summary/2.5_week.csv). To understand the potential impact of these earthquakes, I added a layer that showed the population of the world's major cities (Esri's World's Cities layer). Some countries with large populations (China, India, Indonesia) lie in areas that are prone to earthquakes. Some less populated countries (eg New Zealand) are also in earthquake prone areas. The other observation is that there is significant and regular seismic activity in the Indian and Pacific Oceans especially when compared to the Atlantic Ocean. Here is the web map that I created:
I also created an interactive web app based on this data. To view the web app in all its glory, go here: http://arcg.is/2llj9Qd
I conducted some basic analysis of US population growth in the 50 most populated US cities since 2010 and compared that with the unemployment rate in these cities. Not surprisingly, there were some clear correlations: cities with high unemployment rates saw low population growth and vice versa. Some highlights from the analysis:
The four cities with the most population growth since 2010 are Austin, Denver, New Orleans and CharlotteThe two cities with negative population growth since 2010 are Detroit and Cleveland.
Here is the web map that I created:
I also created an interactive web app based on this data. To view the web app in all its glory, go here: https://tinyurl.com/hzp4flj
What conclusions can you draw from this data?
I have recently started using Esri's Web GIS called ArcGIS Online. As my first attempt at putting together a "story map", I decided to showcase some of Melbourne's main sporting arenas. For those of you that have not been there, Melbourne is not only the world's most beautiful city (IMHO) but is also the world's undisputed sporting capital (IMHO again).
So picking just a few of the many (many!) sporting arenas in Melbourne to showcase in my story map was relatively easy. I didn't need to do any research - they were all top of mind for me.