At　present,　the　informatization　construction　in　the　medical　field　not　only　enables　medical　services　to　break　through　the　constraints　of　time　and　space,　but　also　makes　medical　services　more　efficient　and　scientific.　The　new　generation　of　information　technology　has　become　a　revolutionary　driving　force　for　traditional　medical　care　to　turn　to　higher-level　smart　medical　care.　The　number　of　patients　with　chronic　diseases　in　China　ranks　first　in　the　world,　and　diabetes　and　related　diseases　are　an　important　part　of　it.　In　order　to　help　residents　detect　diabetes　early,　it　is　necessary　to　establish　a　diabetes　risk　monitoring　system　to　detect　high-risk　groups.　This　system　can　monitor　and　warn　of　diabetes,　and　remind　people　at　risk　of　diabetes　to　seek　medical　treatment　as　soon　as　possible.　In　this　paper,　by　comparing　a　variety　of　machine　learning　algorithms,　such　as　logistic　regression,　random　forest,　LightGBM,　XGBoost,　etc.,　to　find　out　the　algorithm　with　better　performance.　Finally,　a　diabetes　risk　prediction　model　was　integrated　by　stacking　method,　and　a　diabetes　monitoring　system　based　on　ensemble　learning　was　designed.　The　system　plays　a　very　good　auxiliary　role　in　the　treatment　of　diabetes.　©　2023　IEEE.