Feature Selection for Prediction of User-Perceived Streaming Media Quality
Amy Csizmar Dalal and Jamie Olson
International Symposium on Performance Evaluation of Computer and Telecommunication Systems (SPECTS 2007)
San Diego, California (USA), July 16-18, 2007
SPECTS_Summary
This paper considers the selection of features, measurements collected from an instrumented media player application, that most accurately predict the user-perceived quality of a media stream. The features are utilized by a nearest-neighbor stream quality prediction algorithm using a distance metric of dynamic time warping. We explore three ways of selecting features from this data: manually, by observing how application-layer measurements change with changing network congestion conditions; correlation-based; and a mathematically-based technique using principal component analysis (PCA). We compare the prediction algorithm's accuracy obtained using the features selected by each method, using a performance evaluation metric we term hit rate. Our results show that each method selects one feature set that, when used by our predictor, yields very high hit rates (typically 70-90%), and that each of these feature sets includes one particular feature in common: retransmitted packets. We also show that the correlation-based and PCA-based methods of selecting features do not consistently select acceptable feature sets for our stream quality predictor, in terms of the hit rates generated by the predictor.