Factored language models and generalized parallel backoff Export

@misc{citeulike:570947, abstract = {We introduce factored language models (FLMs) and generalized parallel backoff (GPB). An FLM represents words as bundles of features (e.g., morphological classes, stems, data-driven clusters, etc.), and induces a probability model covering sequences of bundles rather than just words. GPB extends standard backoff to general conditional probability tables where variables might be heterogeneous types, where no obvious natural (temporal) backoff order exists, and where multiple dynamic...}, author = {Bilmes, J. and Kirchhoff, K.}, citeulike-article-id = {570947}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.7777}, keywords = {flms}, posted-at = {2006-09-15 11:24:42}, priority = {0}, title = {Factored language models and generalized parallel backoff}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.7777}, year = {2003} }

TY - GEN ID - citeulike:570947 L3 - citeulike-article-id:570947 N2 - We introduce factored language models (FLMs) and generalized parallel backoff (GPB). An FLM represents words as bundles of features (e.g., morphological classes, stems, data-driven clusters, etc.), and induces a probability model covering sequences of bundles rather than just words. GPB extends standard backoff to general conditional probability tables where variables might be heterogeneous types, where no obvious natural (temporal) backoff order exists, and where multiple dynamic... TI - Factored language models and generalized parallel backoff KW - flms AU - Bilmes, J AU - Kirchhoff, K PY - 2003/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.7777 ER -