Some mad machine learning tings in the most hipster language known to man: Visual Basic.

As seen on the CodeProject.

Currently Implements:

• Logistic Regression

Will Probably Implement Sometime in the Distant Distant Future

• Linear Regression
• ANNs/DNNs
• SVMs and other exciting Kernel things
• Multivariate stuff and reguarlisation
• K-Means/Modes
• Boltzmann Machines
• DBNs

Machine Learning. What languages come to mind? R? Python? Matlab? Bet you didn't think Visual Basic.

There are probably reasons as to why few have even bothered building open-source machine learning applications in VB (along the lines of uselessness, crappy performance and a general lack of VB people out there). I have, however, in an intoxicated, sleep-deprived and overly-confident state, set out on this gargantuan (and probably pointless) exercise to teach a friend (who we shall call Hiroshige) something ever-so-slightly beyond the scope of the Computer Science GCSE syllabus.

I have neither any qualifications in Computer Science nor am I familiar with VB. Feel free to fix any horrible errors I've made in my incredibly unstable mental state.

NB: If you want to use your own data, make sure you've saved it as a .CSV file (you can do this in Excel) and make sure that

• The output values are in the last column (green box)
• You have a headings row (red box)
• Boolean values are represented as 1s or 0s (green box)

This demo takes in training examples and trains a binary (yes/no) classifier on them. This can be vaguely useful in some situations - for example, if you're selling glow-in-the-dark towels, you can use a logistic classifier to work out which of your customers would be interested in your new range of Disney-themed bathroom essentials based upon their previous purchases.

The demo looks a bit like this:

Either enter a number or type a path to a .csv file to begin training a classifier. You'll then be allowed to give it inputs to make predictions on.