some questions about PatentMatch dataset

[qiushixiong]

hi,
Thank you for providing such good work！
Understanding that you have created a dataset named PatentMatch, which form 250,000 patents granted by the World Intellectual Property Organisation (WIPO) from 2010 to 2022 collected from Google's public dataset, I would like to know how these 250,000 patents were obtained and if they are available for direct download? Is there a detailed instruction on how to use it after accessing the url provided in the paper?
Your answers and help are very important for my work, thank you!