Signal Boosting for Translingual Topic Tracking Document Expansion and n best Translation

The University of Maryland participated in the TDT topic tracking task This chapter describes the system architec ture including source dependent normalization and then fo cuses on the cross language case in which English training stories were used to nd Mandarin stories on the same topic Processes that may introduce noise including errorful trans lation and transcription are described and ve techniques for minimizing the impact of a reduced signal to noise ratio are identi ed Three techniques focus on signal boosting augmenting story representations with topically related ter minology through document expansion exploiting knowl edge of alternative translations using balanced n best term translation and enriching the bilingual term list to improve translation coverage The remaining two techniques focus on noise reduction removing common stopwords before translation and using corpus statistics to guide translation selection Two of the signal boosting strategies yielded sub stantial gains using techniques that can be ported to other languages fairly easily while outperforming state of the art general purpose machine translation By contrast neither of the noise reduction strategies produced signi cant improve ments The chapter concludes with a brief discussion of fu ture research directions suggested by these results