Data Mining is a process of building models or equations that describe the salient but often subtle characteristics a collection of data. The process often starts by finding natural segmentations in data where logical groupings appear. Those groups can then be classified into segments that have particular meaning to a business. Based on these classifications and segmentations, models can be built which have some ability to make predictions on new data entering a system.

There are many different approaches for developing models from data. Most rely on advanced statistical methods like linear and non-linear regression techniques. While these techniques are important, they do not fill all possible needs for models of data and do not always identify all the salient features of the data. Other non-statistical methods include decision trees and neural networks, a data modeling technique developed from computational modeling of human brains.

Depending on the problem to be solved, NuTech Solutions can implement traditional methods of data mining as well as more advanced, non-traditional methods that are particularly well-suited to finding non-linear relationships in data. These approaches allow us to create models to make better predictions from incoming data. With these modeling techniques we are not only able to make predictions, but are also able to explain what factors were considered in making those predictions.