Virginia Tech graduate student Shehryar Khan is tackling the growing challenge of patent data through the application of machine learning. With the number of granted patents more than doubling in the past 15 years, experts face increasing difficulty in assessing the novelty of new ideas. Khan’s research aims to develop automated solutions that optimize systems to make research information more organized and accessible, ultimately extracting meaningful insights from complex datasets. “My work focuses on optimizing and building systems that make research information more organized, accessible, and accurate,” Khan stated. He describes his efforts as doing “research in research,” emphasizing the practical applications of his findings.
A surge in patent applications has amplified the demand for innovative methods to evaluate the originality of inventions. Researchers are increasingly leveraging machine learning to address this gap. “We are researching how to push models like ChatGPT to produce new ideas. To attempt this, we aim to ask these models to generate patents for us but avoid ideas that already exist,” Khan explained. This approach seeks to move beyond simple predictive algorithms, which often extrapolate from existing knowledge. Sarah Over, Assistant Director for Research Intelligence at Virginia Tech, underscored the broader implications of Khan’s work, stating, “Virginia Tech needs to keep advancing within fields like machine learning and AI.” She noted that patent analyses complement ongoing initiatives within the University Libraries.
Khan’s research at Virginia Tech’s Institute for Advanced Computing is focused on bringing order to the rapidly expanding volume of patent data, a challenge that has become increasingly critical as the pace of innovation accelerates. His project aims to transcend basic predictive text generation, a common criticism of large language models, by addressing the essential need to determine if a proposed invention genuinely represents a novel contribution. This intricate task is complicated by the vast scale of existing intellectual property.
Virginia Tech was an obvious choice for my interests, especially applied machine learning.
Khan’s work is not merely theoretical; it strives to bridge the gap between machine learning and practical application. As Assistant Director Sarah Over noted, the increase in granted patents over the past 15 years has created significant challenges for experts trying to keep up with rapid innovation. Khan’s research embodies a fusion of intellectual property, research evaluation, and advanced computing. Ultimately, he aims to create systems that can not only organize and access research information but also derive “meaningful insights from very messy data sources,” thereby enhancing the understanding of innovation trends and the impact of research.
This focus on evaluating novelty through sophisticated machine learning models represents a critical step in addressing the overwhelming volume of patent data, which continues to grow exponentially. As Khan’s forthcoming publication details his findings on assessing novelty in patent submissions, the implications for both academic research and practical application in innovation are significant. The intersection of AI and patent evaluation could pave the way for a new era in intellectual property management, with the potential for far-reaching consequences across various industries.
See also
AI Study Reveals Generated Faces Indistinguishable from Real Photos, Erodes Trust in Visual Media
Gen AI Revolutionizes Market Research, Transforming $140B Industry Dynamics
Researchers Unlock Light-Based AI Operations for Significant Energy Efficiency Gains
Tempus AI Reports $334M Earnings Surge, Unveils Lymphoma Research Partnership
Iaroslav Argunov Reveals Big Data Methodology Boosting Construction Profits by Billions
