A Patent Analysis Model that Combines Document Clustering and Time Series Analysis

[ 31 May 2024 | vol. 17 | no. 1 | pp. 35-44 ]

About Authors:

Ravi Yadahalli
-VTU, Belgium

Abstract:

Patent analysis (PA) is a systematic approach to evaluating patent documents, which includes examining aspects such as abstracts, claims, and the quantity of patents issued over a specific period. This analysis is crucial in informing research and development (R&D) policy because it enables stakeholders to forecast future trends and advancements in technology based on historical data and emerging patterns. As a result, numerous companies actively engage in PA as a strategic effort to enhance their competitive edge in the market. In this paper, we propose an innovative PA model specifically designed for technology forecasting (TF). Our approach aims to integrate advanced clustering methodologies with predictive analytics to provide a more comprehensive understanding of technological growth areas. By leveraging the extensive database of patent documents obtained from the United States Patent and Trademark Office (USPTO), we will conduct a series of experiments. These experiments are intended to evaluate the effectiveness and accuracy of our proposed model in predicting future technological developments. Through our analysis, we seek to demonstrate how our model can be instrumental in guiding strategic R&D decisions and fostering innovation within various industries.

Keywords:

Patent Analysis, Document Clustering, Technology Forecasting, Time Series Analysis

 

About this Article: