This is the second part of this amazing video series on affinity analysis in R! This is one of my all time favorite data science processes and I love the new insights that I can usually find in doing an affinity analysis. The new trends, customer behaviors and more I can find from data mining the customer purchasing data are usually mind blowing. This is where you can find new, developing customer subsets and demographic makeups - maybe you have a store that has a growing middle eastern or Indian customer group or breakout? Maybe you have a campaign that on the surface looked like it didn't achieve much of anything, but underneath you have identified a new group of young mothers that are shopping for school lunch ideas? It could be anything and it could mean everything to your Marketing or Category departments. By correctly identifying new trends early on you can easily help your company, charity, etc. become more profitable as it can better target and categorize their offerings. Affinity analysis and data mining the customer data will show you these things and help you to open all sorts of new ideas. This is a great data science tutorial for beginners! In this video I will show you 2 different approaches with using the Apriori algorithm (the current de facto method used in data science for association rule mining (aka affinity analysis). We will create a bar chart as you can see in the YouTube video thumbnail and in the video. This will help users to see the occurrence and count of the affinity items. Then we will do another which places all of the affinity item data in a chart for easy analysis. We will use the top 300 affinity relationships in the video, but this is easily tweaked or modified to be whatever you want, but you just can't exceed the number of rules (line items for affinity relationships based on your data). In the video I will cover all of this and show you the complete code in R (RStudio).
Again, affinity analysis is one of my favorite data science processes in r because it almost always has great, quick insights I can use to add to pretty much any analysis and take it over the top. It is also just great data science for beginners that want to be able to get great insights and make a difference right away! Requestors for data science projects at my company (top 200 international Forbes company) are always looking for ways to get a positive spin on any and every campaign, etc. Not all campaigns look positive on the surface, but with an affinity analysis I can pretty much always get a positive spin on new customer developments, trends, etc. that will interest any audience. Affinity analysis is clearly a very important tool in any data scientist toolbox and here I am basically giving you the entire code and explanation so that you can start doing these yourself and impress your coworkers, boss, get that elusive data science job, become a data scientist, lead data analyst, etc...
I hope you found this data science tutorial on affinity analysis and using the apriori algorithm both interesting and helpful.
Please take a moment to subscribe, like and share!
Thanks and have a great day!
God bless!
0 Comments