With　the　development　of　modern　property　service　industry,　the　property　perceived　service　quality　(PPSQ)　evaluation　data　is　characterized　by　multiple　evaluation　subjects,　complicated　data　structure　and　large　scale　data.　Since　the　traditional　decision-making　methods　are　difficult　to　solve　the　above　similar　problems,　this　paper　proposes　a　large　group　decision-making　(LGDM)　method　of　generalized　multi-attribute　and　multi-scale　(MAMS)　based　on　the　linear　programming　technique　for　multidimensional　analysis　of　preference　(LINMAP).　In　this　method,　the　large-scale　heterogeneous　data　of　expert　preference　and　user　evaluation　is　fused.　The　decision　matrix　of　generalized　MAMS　is　used　to　process　user　evaluation　information.　The　positive　ideal　solution　(PIS)　and　the　attribute　weights　are　determined　by　the　LINMAP　model.　The　comprehensive　evaluation　values　are　calculated　and　hereby　the　alternatives　are　ranked　order.　According　to　the　relation　between　attribute　weights　and　preset　values,　a　mechanism　for　identifying　invalid　data　is　designed.　This　paper　analyzes　a　set　of　survey　data　of　PPSQ　for　the　four　public　construction　projects　in　the　same　city.　The　analysis　results　show　the　validity　and　rationality　of　the　proposed　method,　and　develop　the　property　service　evaluation　theory.