The military service system of early Joseon Korea was nominally founded on the principle of “conscription in proportion to household registration (戶口比例徵兵).” Yet the degree to which this principle was implemented equitably across regions has rarely been verified through quantitative and systematic analysis. This study reexamines the equity of military obligations in the early Joseon period by analyzing the Sejong Sillok Jiriji (Geographical Treatise in the Annals of King Sejong, 1454), which records data on households (戶), population (口), and soldiers (軍丁) for 334 administrative units nationwide. The research employs a set of digital humanities methodologies, including text mining, historical statistics, GIS visualization, and exploratory AI-assisted interpretation.

A distinctive feature of this project is that it does not rely on a single digital humanities method. Instead, it integrates multiple techniques in a problem-driven manner to address the specific research question of conscription equity. In the preprocessing stage, XML-encoded primary sources were parsed and structured through text mining methods, allowing household, population, and soldier data to be systematically extracted. This provided a basis for quantitative evaluation that moves beyond traditional manual data handling in historical statistics.

In the analysis stage, statistical approaches were employed to measure correlations among households, population, and soldiers, while inequality indices such as the Gini and Theil coefficients were applied to assess the fairness of military burdens across regions. By adopting these digitalized historical statistical techniques, the study was able to evaluate the degree to which the official principle of proportional conscription was upheld, as well as to identify regional variations and deviations from the norm.

To further examine spatial patterns, GIS was utilized to visualize the distribution of military obligations across localities. This spatial humanities approach enabled “distant viewing” of inequality, allowing macro-level patterns—such as clusters of heavier or lighter military burdens—to be discerned at a glance. GIS visualization not only provided insights into regional disparities but also demonstrated how digital humanities tools can extend traditional historical inquiries into new visual and analytical dimensions.

In addition, this study explored the potential of AI-assisted interpretation in a preliminary fashion. AI has shown remarkable capabilities in image analysis, and we tested its application by tasking it to interpret GIS visualizations generated in this project. By re-analyzing spatial outputs through AI, we were able to explore new perspectives on the same data and confirm the possibility of leveraging AI to augment human interpretation. While human historians remain central in contextual analysis, this experiment suggests that AI can serve as a complementary lens for examining patterns in visualized historical data.

In conclusion, this study confirms that while early Joseon’s military service system generally adhered to the principle of proportional conscription, significant regional inequalities nonetheless existed. Methodologically, the contribution of this research lies in its integrative approach: combining text mining, digitalized historical statistics, GIS visualization, and exploratory AI-assisted interpretation in order to address a concrete historical question. Rather than showcasing a single tool, the project demonstrates how digital humanities methods can be mobilized collectively in a problem-driven way to serve the core inquiries of historical scholarship.