“…At present, in spatial data mining, there are mainly three kinds methods of mining spatial association rules written in [3], such as, layer covered based on clustering written in [3], mining method based on spatial transaction written in [2, 4, 5 and 6] and mining method based on non-spatial transaction written in [3]. We use the first two methods to extract frequent neighboring class set written in [4, 5 and 6], but AMFNCS written in [4] and TDA written in [5] are not able to efficient extract frequent neighboring class set with constraint class, and MFNCSWCC written in [6] via iterative search is unsuitable for extracting long frequent neighboring class set with constraint class. Hence, this paper proposes an algorithm of frequent neighboring class set mining with constraint class based on breadth first search, denoted by FNCSMBFS, which is suitable for mining long frequent neighboring class set with constraint class in large spatial database.…”