Kernel Density Estimation Approach for a Heatmap-Enhanced Curriculum Vitae Generator

The increasingly competitive job market requires job seekers, particularly fresh graduates, to have a Curriculum Vitae (CV) that is not only visually appealing but also compliant with Applicant Tracking System standards. At the same time, job seekers often lack insight into the spatial distribution of job opportunities, so career decisions are not sufficiently supported by location-based information. This study develops a web-based CV generator system integrated with a job vacancy heatmap visualization using the Kernel Density Estimation (KDE) method. The KDE process is carried out on a two-dimensional evaluation grid constructed from the geographic coordinates of job vacancies and computed using a Gaussian kernel in the Web Mercator coordinate system. The quality of the heatmap is evaluated through sensitivity analysis and configuration assessment of KDE by varying the bandwidth and grid resolution parameters, then comparing the distribution of density classes as well as the stability of the resulting hotspots across several configurations. The experimental results show that a medium bandwidth combined with a medium grid resolution produces a balanced density pattern, stable hotspots in major metropolitan areas in Indonesia, and a visualization that is easy for users to interpret. The resulting system not only helps users compose ATS-friendly CVs, but also provides spatial insights into regions with high job opportunity concentrations, thereby supporting more strategic career decision-making.