| S13-2017 | model , in combination with latent | vector weighting | . The system computes the similarity |
| S13-2017 | with a technique called latent | vector weighting | . The system computes the similarity |
| J09-3004 | improving the quality offeature | vector weighting | in distributional word similarity |
| S13-2017 | that our model based on latent | vector weighting | performs quite a bit better than |
| S13-2017 | Methodology Our method uses latent | vector weighting | ( Van de Cruys et al. , 2011 |
| S13-2017 | and our model that uses latent | vector weighting | . The results indicate that our |
| W14-1502 | ( i.e. be - fore , or after , | vector weighting | ) . Table 7 shows the average |
| D11-1094 | density as well . <title> Latent | Vector Weighting | for Word Meaning in Context </title> |
| P04-1080 | tested CGDterm using various word | vector weighting | methods when deriving context |
| P14-1023 | that motivates traditional count | vector weighting | measures such as PMI ) . This |
| S13-2007 | pus , in combination with Latent | Vector Weighting | ( Van de Cruys et al. , 2011 |
| S13-2017 | and Compositional using Latent | Vector Weighting | </title> Tim Van_de_Cruys Stergos |
| E14-1025 | word , wi and context word , cj . | Vector Weighting | We used the tTest and PPMI weighting |
