Design and implementation of Data Analysis Algorithms

One of our company’s strengths is in the design and implementation of data analysis algorithms to clean and process data acquired in the context of a behavioral task. Such a task might be playing a serious game or using devices designed for rehabilitation therapies.  The data analysis algorithms we can implement go beyond simple ‘frequency-band filters’, and extend to single-sensor and multi-sensor statistical filtering and classification methods.  In addition, our expertise extends to the analysis of data from biometric sensors and the analysis of behavioral data such as latencies to goal completion.  A summary of types of data are given below.

Data from Input Devices:

• Joysticks
• Button pads
• Movement sensors
• Eye-trackers
• Cameras
• Temperature devices

Data from Biometric Sensors:

• Electroencephalography
• Electromyography
• Skin Conductance

Data relating to behavior and user characteristics:

• Latency to goal completion
• Weight
• Height
• Hours of sleep per day

These types of data can be processed independently and/or together, and can be used to:

• Obtain accurate measurements of the feature/signal of interest

• Identify new and important data characteristics via statistical data mining methods

• Identify personas or sub-populations that each have somewhat unique characteristics in the data via statistical sub-grouping procedures