Name
Title | ||
|---|---|---|
Unsupervised concept annotation using latent Dirichlet allocation and segmental methods |
Abstract | ||
|---|---|---|
Training efficient statistical approaches for natural language understanding generally requires data with segmental semantic annotations. Unfortunately, building such resources is costly. In this paper, we propose an approach that produces annotations in an unsupervised way. The first step is an implementation of latent Dirichlet allocation that produces a set of topics with probabilities for each topic to be associated with a word in a sentence. This knowledge is then used as a bootstrap to infer a segmentation of a word sentence into topics using either integer linear optimisation or stochastic word alignment models (IBM models) to produce the final semantic annotation. The relation between automatically-derived topics and task-dependent concepts is evaluated on a spoken dialogue task with an available reference annotation. |
Year | Venue | Keywords |
|---|---|---|
2011 | ULNLP@EMNLP | unsupervised concept annotation,ibm model,integer linear optimisation,dialogue task,stochastic word alignment model,segmental method,final semantic annotation,latent dirichlet allocation,automatically-derived topic,available reference annotation,word sentence,efficient statistical approach,segmental semantic annotation |
Field | DocType | Volume |
Integer,Latent Dirichlet allocation,IBM,Annotation,Segmentation,Computer science,Pachinko allocation,Natural language understanding,Natural language processing,Artificial intelligence,Sentence,Machine learning | Conference | W11-22 |
Citations | PageRank | References |
3 | 0.39 | 13 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Nathalie Camelin | 1 | 39 | 14.29 |
| Boris Detienne | 2 | 53 | 5.78 |
| Stéphane Huet | 3 | 40 | 9.15 |
| Dominique Quadri | 4 | 35 | 6.18 |
| Fabrice Lefèvre | 5 | 185 | 26.62 |