Recent years have seen a large influx of applications in the field of Body Sensor Networks (BSN). BSN, and in general wearable computers with sensors, can give researchers, users or clinicians access to tremendously valuable information extracted from data that were collected in users' natural environment. With this information, one can monitor the progression of a disease, identify its early onset or simply assess user's wellness. One major obstacle is managing repositories that store large amounts of BSN data. To address this issue, we propose a data mining approach for large BSN data repositories. We represent sensor readings with motion transcripts that maintain structural properties of the signal. To further take advantage of the signal's structure, we define a data mining technique using n-grams. We reduce overwhelmingly large number of n-grams via information gain (IG) feature selection. We report the effectiveness of our approach in terms of the speed of mining while maintaining an acceptable accuracy in terms of precision and recall. We demonstrate that the system can achieve average 99% precision with an average 100% recall on our pilot data with the help of only one transition for each movement.