A unique pattern detection tool for behaviour analysis
From Noldus Information Technology
Theme is a tool for the detection and analysis of patterns in time-based data. It finds relationships that human observers typically overlook and commonly used statistical methods cannot find.
Theme uses a unique pattern detection algorithm that was especially designed for behavioral research. It gives insight into the structure of behavior in time, and gives measures for its complexity and organization.
Theme is based on an algorithm that tests the distribution of each pair of behavior types in time. If there is a significant relationship they are called a T-pattern.
The relationships between the T-patterns themselves are then tested in the same way. The T-patterns are extended with more elements and other T-patterns, so that higher order T-patterns emerge. Related patterns are combined and incomplete patterns are eliminated.
This procedure is repeated until the patterns can no longer be extended. The result of the calculation is a group of patterns called a T-set.
Theme is used for testing differences between groups or treatments, and for structural analysis. It has been successfully applied in a broad range of fields, including psychology, zoology, neuroscience, and sports science.
Successful case studies (which you may see on the Noldus website) include:
- Discovering behavioral patterns in the use of a universal remote control
- Soccer analysis
- Reveal the organization of autistic behavior
- Hormone-behavior interaction
- Feed pecking in chicks
- Analyzing web logs