In　view　of　the　many　shortcomings　of　the　current　industry　recommendation　system,　this　article　has　carried　out　meaningful　and　valuable　in-depth　research　and　exploration　on　the　collaborative　filtering　recommendation　algorithm　[1]　[2]　adopted　by　the　personalized　news　recommendation　system,　hoping　to　help　users　accurately　from　the　huge　amount　of　news　ocean　Quickly　find　interesting　news　and　get　a　good　user　experience.　The　main　tasks　of　this　paper　are:　1)　Conduct　an　in-depth　study　and　review　of　related　technologies　for　personalized　news　recommendation.　2)　Propose　an　improved　collaborative　filtering　recommendation　algorithm　based　on　user　item　hybrid　model.　3)　Design　and　implement　a　personalized　news　recommendation　system　[3]　[4]　based　on　a　collaborative　filtering　recommendation　algorithm　based　on　an　improved　user　item　hybrid　model　[5].　The　experimental　data　proves　that　this　system　has　a　good　personalized　recommendation　function,　and　the　personalized　news　recommendation　system　based　on　this　algorithm　is　more　effective.　The　recommendation　effect　of　the　traditional　personalized　recommendation　system　has　been　significantly　improved.