Patents Phrase to Phrase Semantic Matching Dataset
This provides a resource for researchers and practitioners working on semantic matching in patents and scientific publications, but it is incremental as it adapts existing benchmark concepts to a specific domain.
The authors tackled the lack of benchmark datasets for semantic textual similarity in technical domains by creating a new human-rated dataset of nearly 50,000 phrase pairs with CPC class contexts, and they described baseline models for it.
There are many general purpose benchmark datasets for Semantic Textual Similarity but none of them are focused on technical concepts found in patents and scientific publications. This work aims to fill this gap by presenting a new human rated contextual phrase to phrase matching dataset. The entire dataset contains close to $50,000$ rated phrase pairs, each with a CPC (Cooperative Patent Classification) class as a context. This paper describes the dataset and some baseline models.