Because　the　network　consisting　of　computers　worldwide　has　its　individual　network　nodes　are　also　unusually　large,　traditional　manual　testing　methods　can　not　be　applied　to　computer　network　monitoring,　and　fault　may　cause　slight　paralysis　of　large-scale　networks,　for　this　problem,　in　this　paper,　chaotic　neural　network　artificial　intelligence　monitoring　methods　for　large-scale　networks　can　cause　paralysis　of　the　failure　to　monitor　the　accuracy　of　the　algorithm　is　significantly　improved　compared　with　other　algorithms,　and　the　algorithm　converges　faster,　less　calculation,　suitable　for　online　testing.　©　2016　SERSC.