As a fraud practitioner using data mining techniques to detect fraud, anomalies, outliers or other indicators of potential problems I use a combination of data mining and data matching techniques.
The volumes of data in a client assignment can vary from 15 million records of company directors, 60,000 employees, accounts payables data of suppliers 900,000 and invoice transaction 11,million. I'm not a great fan of predictive technologies as the disparate data sets don't seem to fit with the techniques, but I'm open to alternative methodologies. I've recently tested a single fraud profile using "Receiver Operating Characteristic" to evaluate the sensitivity and specificity of the profile. The results fell within the ROC space. This was a single indicator ans so I need to look at multivariate data visualisation, but this is a new area for me so any help or suggestions would be gratefully received.
In creating a fraud profile which reviews the standing data about an organisation and is indicative of potential problems I have over 10 different variables. Visualizing this using a radar chart is 2 dimensional and I can see that a vertical dimension of financial impact would create a 3 dimension visualization. adding some sort of colour graduation would effectively create a heat map. I've looked at Datawatch Designer and Tableau but they don't seem to give me the functionality that I'm looking for. Again, any suggestions?
Take a peruse through the d3.js gallery https://github.com/mbostock/d3/wiki/Gallery you probably won't find a solution, but it will give you an idea of the "visualization zoo" that can be combined. I doubt you will be able to do it all in a single viz.
I would consider Krzywinski’s hive (network) plot in conjunction with a scatterplot matrix and a heat map.
hope this helps,
Colormapping of Multivariate data might be tricky and complicated sometimes. There are some good papers by K.R.Gabriel on how to interpret and display such data.
1983 Multivariate Graphics, Encyclopedia of Statistical Science, chapter 5.
1981 Biplot Display of Multivariate matrices for inspection and diagnosis. interpreting Multivariate Data (v. Barnett, Ed) 167-173 Wiley chapter 5
There is also a book on "Multivariate Observations" by G.A.F. Seber, Wiley publications. The chapter 5 on "Dimension Reduction and Ordination" might be of interest to you.
Hope this helps.
A new member since 4/28/2014