About this subject
Wikidata Entity Descriptor Word Sense Induction
This page shows how accurately we can identify different meanings of words used in Wikidata, a collaborative knowledge base. We measure the performance of our word sense induction systems, which aim to distinguish between the various senses a word can have. For example, 'bank' can refer to a financial institution or the side of a river, and our systems try to tell these apart.
These numbers reflect the progress in Natural Language Processing (NLP) specifically applied to Wikidata's structured data. By tracking word sense induction accuracy and the number of senses identified, we gain insights into the evolving complexity of language within this vast dataset. This is important for improving how computers understand and process information, but it's rarely published due to the technical NLP expertise required.