5 ms·
...And just distributional knowledge makes it easy to train new models on domain-specific corpora, or new languages. Is it possible to do the same with ConceptN
by glup 10y ago
...And just distributional knowledge makes it easy to train new models on domain-specific corpora, or new languages. Is it possible to do the same with ConceptNet?
I generally find that expert-derived ontologies suffer from bad coverage of low frequency items, rigidly discrete relationships, and are usually limited to a single language. That said, they're vastly better than nothing for a lot of tasks (same goes for WordNet).
- rspeer 10y agoYou can retrain your distributional knowledge and keep your lexical knowledge. Moving to a new domain shouldn't mean you have to forget everything about what words mean and hope you manage to learn it again. The whole idea of Numberbatch is that a combination of distributional and lexical knowledge is much better than either one alone. BTW, ConceptNet is only partially expert-derived (much of it is crowd-sourced), aims not to be rigid like WordNet is, and is in a whole lot of languages. "Retraining" ConceptNet itself is a bit of a chore, but you can do it. That is, you can get the source [1], add or remove sources of data, and rebuild it. Meanwhile, if you wanted to retrain word2vec's Google News skip-gram vectors, you would have to get a machine learning job at Google. [1] https://github.com/commonsense/conceptnet5 https://github.com/commonsense/conceptnet5