Diabetes　is　one　of　the　most　common　diseases　worldwide.　High　level　of　blood　sugar　in　diabetes　patients　can　harm　a　large　number　of　the　body＇s　system.　Early　prediction　of　diabetes　can　prevent　or　delay　the　disease.　Various　machine　learning　methods　are　used　to　predict　diabetes　in　the　past.　Researchers　usually　aim　for　higher　accuracy,　the　models　used　become　more　and　more　complex　and　their　decision-making　process　is　extremely　difficult　understood　by　users.　However,　explainability　of　a　model　is　also　critical　to　prediction　task　in　medicine.　A　model　which　can　enable　users　to　easily　understand　its　decision-making　logic　while　maintaining　good　accuracy　is　more　likely　to　be　trusted　by　users.　Therefore,　we　propose　a　special　multivariate　polynomial　model　to　predict　diabetes.　This　model　has　ability　to　show　the　relationship　between　each　medical　factor　and　diabetes　with　some　polynomial　curves,　and　the　product　of　these　curves　and　a　specific　constant　is　the　decision-making　process　of　the　model.　The　experiment　results　show　that　our　model　also　has　a　good　accuracy　compare　with　some　other　methods.　©　2022　ACM.