there　exist　many　problems　in　intrusion　detection　system　such　as　large　number　of　data　volume　and　features,　data　redundancy　and　so　on,　which　seriously　affected　the　efficiency　of　the　assessment.　In　this　paper,　we　propose　an　approach　called　EFSA-CP　to　intrusion　detection　based　on　Cloud　model　and　improved　multi-objective　Particle　Swarm　Optimization.　The　algorithm　evaluates　the　characteristics　of　the　attribute　weights　by　the　Cloud　model　and　generates　the　optimal　feature　subsets　which　achieve　the　best　trade-off　between　detection　rate　and　rate　of　false　alarm　by　MOPSO,　which　solves　the　problem　of　feature　redundancy　and　helps　improve　the　speed　of　the　evaluation.　Experimental　results　show　that　EFSA-CP　can　solve　the　feature　selection　problem　of　intrusion　detection　effectively.　It　can　also　achieve　balanced　detection　performance　on　different　types　of　attacks,　with　better　convergence　at　the　same　time.　©　2011　IEEE.