the　recommendation　system　is　one　of　the　core　tasks　of　data　mining.　It　divided　into　the　recommendation　system　for　explicit　feedback　and　the　recommendation　system　for　implicit　feedback.　In　recent　years,　many　researchers　have　combined　deep　learning　with　recommendation　system　of　explicit　feedback　and　achieved　excellent　results.　But　it　is　very　rare　for　implicit　feedback.　In　this　paper,　we　apply　deep　learning　to　the　recommendation　system　for　implicit　feedback,　and　propose　a　new　model　that　is　combined　with　Neural　Collaborative　Filtering　and　Variable　automatic-encoder.　We　use　the　MovieLens　dataset　to　evaluate　our　proposed　model.　Experimental　results　show　that　the　proposed　model　effectively　improves　the　accuracy　and　quality　of　the　recommended　results,　the　Precision　is　0.715　and　the　NDCG　is　0.436　without　manual　parameters.