Unveiling Black-boxes: Explainable Deep Learning Models for Patent Classification

TL;DR

This work tackles the problem of explaining deep learning-based patent classification, where multi-label outputs and technical jargon make models hard to interpret. It introduces a two-part framework that trains Bi-LSTM, CNN, and CNN-BiLSTM classifiers on semantically enriched patent text using FastText embeddings, and uses Layer-wise Relevance Propagation to produce word-level explanations for each prediction. Experimental results on two large patent datasets show competitive accuracy across models, with Bi-LSTM often delivering the strongest performance, while the LRP explanations highlight domain-relevant terms that align with predicted classes. The approach promises greater transparency and potential adoption in AI-enabled patent management, and points to future work on domain-specific embeddings and transformer-based methods alongside user-centric evaluation.

Abstract

Recent technological advancements have led to a large number of patents in a diverse range of domains, making it challenging for human experts to analyze and manage. State-of-the-art methods for multi-label patent classification rely on deep neural networks (DNNs), which are complex and often considered black-boxes due to their opaque decision-making processes. In this paper, we propose a novel deep explainable patent classification framework by introducing layer-wise relevance propagation (LRP) to provide human-understandable explanations for predictions. We train several DNN models, including Bi-LSTM, CNN, and CNN-BiLSTM, and propagate the predictions backward from the output layer up to the input layer of the model to identify the relevance of words for individual predictions. Considering the relevance score, we then generate explanations by visualizing relevant words for the predicted patent class. Experimental results on two datasets comprising two-million patent texts demonstrate high performance in terms of various evaluation measures. The explanations generated for each prediction highlight important relevant words that align with the predicted class, making the prediction more understandable. Explainable systems have the potential to facilitate the adoption of complex AI-enabled methods for patent classification in real-world applications.

Figures (6)

• Figure 1: A conceptual overview diagram of our explainable patent classification framework.
• Figure 2: The distribution of the patents for different class on AI-growth-Lab data
• Figure 3: The distribution of the patents for different class on BigPatent data
• Figure 4: An example explanation for a patent classified as Chemistry patents highlighting relevant words. The higher the intensity of the color, the better the relevancy of the words contributing to the prediction.
• Figure 5: An example explanation for a patent classified as Chemistry patents highlighting relevant words. The higher the intensity of the color, the better the relevancy of the words contributing to the prediction.