<![CDATA[Cadastral surveys are vital for assessing individual properties and generating national statistics. In South Korea, rapid territorial changes have exposed personal-level issues, while current systems limited public engagement capabilities have constrained policymaking. Although efforts have been made to bridge this gap, they have been hindered by the lack of an effective medium. This study introduces a novel framework, PRiSm (Policy Recommendation Systems in Cadastral Survey), which leverages National Public Opinion Big Data. We collected and analyzed two key data sources: 1) public opinion data from 2018 to 2023, which correlates strongly with cadastral resurveys across South Korea, and 2) content from 54 major news media outlets over the same period. The first data set represents bottom-up opinions at the individual level, while the second reflects top-down perspectives on national issues. The PRiSm system, developed in this study, utilizes Natural Language Processing (NLP) and advanced Machine Learning (ML) techniques, including Word2Vec and a Genetic Algorithm for hyperparameter optimization, to process over a thousand inquiries and news articles. Our results highlight how different groups engage in discussions shaped by their interests and concerns, revealing key sensitivities and recommending terms invaluable for stakeholders and policymakers. We anticipate that PRiSm will offer meaningful insights for the public and decision-makers. Additionally, with more advanced ML and/or Deep Learning algorithms, there is significant potential for further advancements in NLP within the PRiSm framework.]]>
Copyrights © 2024
