search

cbadenes / patent-classification

Patent IPC class prediction and analysis
Apache License 2.0
0 stars 0 forks source link

Next Steps #6

Closed cbadenes closed 3 years ago

cbadenes commented 3 years ago

actions designed to improve the performance of the model

cbadenes commented 3 years ago

Use of Python-based NLP tools:

cbadenes commented 3 years ago

Add Named Entities Recognition (NER) tasks during text pre-processing since patents often use domain-specific names

cbadenes commented 3 years ago

Incorporate external knowledge (e.g. DBpedia, Wikidata, ..) by identifying entities (entity linking) to extend the original text with associated information that can better characterize the topics: