An improved boosting algorithm and its application to automated text categorization

摘要

We describe AdaBoost.MH , an improved boosting al-gorithm, and its application to text categorization. Boostingis a method for supervised learning which has successfullybeen applied to many different domains, and that has provenone of the best performers in text categorization exercisesso far. Boosting is based on the idea of relying on the collec-tive judgment of a committee of classifiers that are trainedsequentially. In training the i-th classifier special emphasisis placed on the correct categorization of the training docu-ments which have proven harder for the previously trainedclassifiers. AdaBoost.MHKR is based on the idea to build,at every iteration of the learning phase, not a single classi-fier but a sub-committee of the K classifiers which, at thatiteration, look the most promising. We report the resultsof systematic experimentation of this method performed onthe standard Reuters-21578 benchmark. These experimentshave shown that AdaBoost.MHKR is both more efficient totrain and more effective than the original AdaBoost.MHRalgorithm.