Access　control　is　an　important　means　of　ensuring　information　security,　and　trust-based　access　control　has　been　considered　as　an　effective　mechanism　for　access　control　in　open　networks.　To　make　access　control　mechanisms　more　effective,　the　assessment　of　trust　has　become　a　key　issue.　Among　the　different　components　of　trust,　recommendation　trust　plays　a　key　role　in　environments　and　applications　in　which　there　are　not　enough　direct　interactions　between　entities,　making　direct　trust　less　reliable.　In　this　paper,　we　propose　a　method　for　the　assessment　of　recommendation　trust　in　which　we　try　to　solve　the　problem　of　weight　assignment　to　decision　attributes　by　using　relevant　theories　in　expert　investigation,　fuzzy　analysis　and　gray　correlation　analysis.　We　also　perform　some　experiment　to　show　that　the　results　of　the　proposed　assessment　method　are　in　line　with　the　actual　ones　to　demonstrate　the　effectiveness　of　the　proposed　method　as　well　as　its　advantages　over　some　other　comparable　methods.