| D15-1209 | paper proposes a leave-one-out | expectationmaximization algorithm | for unsupervised word alignment |
| P14-1138 | These models are trained using the | expectationmaximization algorithm | ( Dempster et al. , 1977 ) from |
| P15-2086 | representation vector and employ an | expectationmaximization algorithm | to jointly learn distributed |
| C96-2185 | . ( Shin , 1996 ) experimented | expectationmaximization algorithm | with 68.7 % accuracy at phrase |
| P03-1051 | unsupervised technique based on the | expectationmaximization algorithm | and minimum description length |
| P96-1044 | . Given a fixed lexicon , the | expectationmaximization algorithm | ( Dempster et al. , 1977 ) can |
