A study on elm(election pledge management for local governors model) based on machine learning -focused on on-nara document system

Abstract

The background of this paper is new social trend of more public's interest in the implementation of the pledge of the local governors who were elected by citizens. In these days the election pledge for enhanced local governmental policies became more important. The objective of this paper is to suggest the model of election pledge management for local government heads based on machine learning focused on On-Nara document system. The system is currently used by Korean governmental organizations for document processes. The methods to prove a comparative advantage of the proposed model are the comparison tests between As-Is system and To-Be system based on a few criteria such as time, efficiency and extraction rate. Through this model, local governors could present systematic goals and road map of pledges in order to get closer to citizens and local residents. In other words, this study proposes a model, so called ELM(Election pledge management for Local governors Model), for efficiently extracting necessary data from planned and implemented details of pledge projects that are prepared in the form of unstructured documents. We carried out research to prove empirically our machine learning-based model is more efficient than current semimanual system with some automated processes in order to manage efficiently the pledge project implementation of local governors to get the results. In conclusion, this research proved that the proposed model is more competitive than the existing models. In the 4th industrial revolution era the new approach using machine learning and big data will become more popular. © 2019 Association for Computing Machinery.