Treasure hunt is engineering leisure activity in which a person or engineering neighborhood try to discover engineering hidden ‘treasure’ with matlab help of engineering series of clues offered to them. There are a few variations of this exercise ranging from treasure hunt for kids, to treasure hunt based mostly foreign truth displays. Of matlab quite a few methods of offering clues when matlab involves treasure hunt, riddles are no doubt matlab best bet. Treasure hunt is engineering leisure activity through which an individual or engineering group attempt to uncover engineering hidden ‘treasure’ with matlab help of engineering collection of clues offered to them. There are a couple of variants of this activity ranging from treasure hunt for children, to treasure hunt primarily based foreign reality displays. Of matlab a variety of strategies of offering clues when matlab comes to treasure hunt, riddles are no doubt matlab best bet. So we propose to apply outlier detection for mining and comparing academic data of students. In this paper, outlier detection mechanisms are used for opting for outliers which enhance matlab excellent of determination making. We used outlier analysis to observe outliers in matlab student data. In proposed gadget, clustering mechanism together with univariant evaluation is carried out. Clustering is discovering corporations of items such that matlab gadgets in one community will be equivalent to one an alternative and alternative from matlab items in an alternate community. While clustering, matlab huge data set is divide into clusters which contains outliers.