Grasping and Visualizing Patent Claim Structure

What is it about?

Though patent claims are the most important section in a patent, it is hard to examine them. This paper shows a machine learning-based method to extract key features from patent claim structures and an interface to visualize patent structures, compare patents, and search similar patents.

Featured Image

Photo by madeleine ragsdale on Unsplash

Photo by madeleine ragsdale on Unsplash

Why is it important?

There are two viewpoints: claim analysis cost and readability improvement. We used machine learning-based information extraction techniques with distant supervision labeling instead of using only syntactic parsing to reduce the cost of extracting important terms from patent claims and organizing them. After key terms and phrases in claims are annotated, we could implement a patent claim viewer interface with improved readability. When the original claim is being read, highlighted text is shown to improve the readability of the text. If the aim is to grasp the claim structure, a simplified view showing only extracted terms and phrases is shown.