PatentSBERTa: A Deep NLP based Hybrid Model for Patent Distance and Classification using Augmented SBERT

TL;DR

This paper introduces PatentSBERTa, a hybrid deep NLP model using SBERT embeddings for patent similarity measurement and classification, achieving high accuracy and F1 scores in multi-label patent classification tasks.

Contribution

The study presents a novel hybrid framework combining SBERT-based embeddings with a simple KNN classifier for efficient patent similarity assessment and classification, outperforming existing methods.

Findings

Achieved 54% accuracy and >66% F1 score in multi-label patent classification.

Validated that p2p similarity effectively captures technological features.

Demonstrated the framework's usefulness for semantic patent search and technology intelligence.

Abstract

This study provides an efficient approach for using text data to calculate patent-to-patent (p2p) technological similarity, and presents a hybrid framework for leveraging the resulting p2p similarity for applications such as semantic search and automated patent classification. We create embeddings using Sentence-BERT (SBERT) based on patent claims. We leverage SBERTs efficiency in creating embedding distance measures to map p2p similarity in large sets of patent data. We deploy our framework for classification with a simple Nearest Neighbors (KNN) model that predicts Cooperative Patent Classification (CPC) of a patent based on the class assignment of the K patents with the highest p2p similarity. We thereby validate that the p2p similarity captures their technological features in terms of CPC overlap, and at the same demonstrate the usefulness of this approach for automatic patent classification based on text data. Furthermore, the presented classification framework is simple and the results easy to interpret and evaluate by end-users. In the out-of-sample model validation, we are able to perform a multi-label prediction of all assigned CPC classes on the subclass (663) level on 1,492,294 patents with an accuracy of 54% and F1 score > 66%, which suggests that our model outperforms the current state-of-the-art in text-based multi-label and multi-class patent classification. We furthermore discuss the applicability of the presented framework for semantic IP search, patent landscaping, and technology intelligence. We finally point towards a future research agenda for leveraging multi-source patent embeddings, their appropriateness across applications, as well as to improve and validate patent embeddings by creating domain-expert curated Semantic Textual Similarity (STS) benchmark datasets.