Drighten Research staff have experience performing data mining on petabytes of data in real time. Along with experience with large datasets, our staff also has studied data from a wide variety of disciplines. Contact us about improving the use of your data sets by utilizing Drighten Research's extensive data mining capabilities.
Through data visualization information can be communicate intuitively through graphical means. By communicating essential features we can convey ideas hidden within complex data sets effectively.
Classification algorithms can be used on numerous problems such as recognition of chemical signatures, facial recognition, identification of malware, etc. Through classification one or more variables allows identification of similarly attributed items in a dataset. Our staff is currently comparing novel quantum phase space classification methods with more standard neural network and bayesian methods.
Pattern recognition looks for patterns (including anomalous patterns) associated with activities of interest. Such patterns may go unrecognized by human observation, but with appropriate tools even small signals can be identified in an ocean of noise. One of our staff members discovered unknown extinction events utilizing these methods.
Our staff is experienced in applying agent based models and other techniques to perform predictive modeling. Current applications range from prediction of hacker attack patterns to automation of google ad word selection.