Slaughter Analytics – Part 2


In the first part of Slaughter Analytics, I showcased how I performed self-service ETL to pull in and model data from a livestock slaughter statistics webpage. In this post I shall focus on performing some simple analytics through visualizations.

Excel 2013 gives us two more add-ins for visualization, interactive analytics and storytelling: Power View and Power Map. The video below showcases the demo that I used to perform the analysis of the extracted and modeled data.

Slaughter Analytics – Visualization

 

The first part of the demo uses Power View. Here, I add two sets of data to the visualization area. Data by default is added as a table, which I subsequently change to a graph of my choice. The first graph shows the trend of slaughter throughout the years for the three categories of animals. The second graph shows the total slaughter count from each district. As a data analyst, I wanted to see what sort of an impact, the opposition against the slaughter of cattle on religious grounds has had (We know that we’ve had a lot of those in the recent past). So I chose the two districts whose namesakes are sacred cities: Anuradhapura and Kandy:

Slaughter Analytics - Anuradhapura Slaughter Analytics - Kandy

Anuradhapura

Kandy

I click on the graph line for Anuradhapura on the second graph and the data on the first graph changes to show data for Anuradhapura. I can do the same thing for Kandy as well.

Looking at Anuradhapura, I can see that the trend in cattle killings has not changed more or less since 2006 – the slaughter continues. Kandy has a different story, a great dip is experienced from 2007 with a low in 2011, before taking off once again in 2012. Since there is no more data, we cannot be too sure what the status is right now in 2014. Similarly, I do more analysis on aspects that interest me, such as what animals are slaughtered in Colombo district, and an interesting comparison between the home districts of Hambantota and Polonnaruwa, which at current are two very “important” places – Note that the total number of slaughters in both these districts are almost the same, and they both have declining trends in cattle slaughter – something that makes it difficult for you (if you want to make a certain decision based on this) in January 2015, if you know what I mean – No computer gilmaat intended.

Slaughter Analytics - Hambantota Slaughter Analytics - Polonnaruwa

Hambantota

Polonnaruwa

Next, Power Map. Power Map allows to visualize data on maps of the world, and from recent times on custom maps as well. In my demo, I choose the District field along with slaughter counts which intuitively maps to districts on the Sri Lanka map without the need for geo-coordinates – Ratnapura does not get mapped though (I would probably have to report a bug for that). As a data analyst I want to see how many more cattle are slaughtered over other animals in each district. I first choose contrasting colors to represent each type of animal on the map, and then I choose an option called Shade based on category value compared to other values in the category. I use this option so that I can see how many more of one type of animal is slaughtered more than others. For example, In the Puttalam district more cattle are slaughtered over other animals than in the Trincomalee district. Whereas in the Galle district more cattle are slaughtered over other animals, but less than how it is done in Trincomalee. How do we know this? The shading over Puttalam is darker red than Trincomalee, which in turn is darker red than Galle. Colombo however, slaughters goats and sheep more than other animals.

I hope you now have an understanding of what self-service BI is in its basic manifestation. There’s more to it than what meets the eye. To learn more about Power BI, I recommend this video series. You can also try out the Power View interactions of my demo.

UPDATE [26-Dec-2014]:

Just noticed that provisional data for the year 2013 is updated on the source web page. Bad news for those who will be taking that all-important decision in January 2015: you will still be in a dilemma to choose between the two:

Slaughter Analytics - Hambantota New Slaughter Analytics - Polonnaruwa New

Hambantota (with 2013 data)

Polonnaruwa (with 2013 data)

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s