The　false　advertising　of　food　and　drag　on　the　Internet　is　mainly　based　on　the　content　of　the　product　website　promotion　pages.　When　people　browse　a　website,　they　get　the　most　parts　of　the　information　from　texts　on　the　web.　In　order　to　help　people　to　distinguish　whether　it　is　false　propaganda　on　this　website,　we　propose　a　solution　for　identifying　false　advertising　of　text　content　on　food　and　drug　websites　by　designing　the　sensitive　word　recognition　model.　This　paper　introduces　in　detail　the　specific　design　and　implementation　of　the　food　webpage　text　sensitive　text　recognition　model,　including　the　system　improvement　of　text　acquisition　and　word　segmentation　algorithm,　feature　extraction　algorithm　and　text　classification　in　the　sensitive　word　list　extraction.　The　detailed　design　and　execution　flow　of　the　voting　decision　determination　result　algorithm　of　the　five　text　classification　algorithms　are　combined　for　filtering.　Finally,　we　conducted　a　series　of　experiments,　and　the　experimental　results　demonstrated　that　the　proposed　filtering　solution　is　effective.