” Quasar” or” Quasi-stellar object” astronomical star-like object having a large ultraviolet flux of radiation accompanied by generally broad emission lines and absorption lines in some cases found at large redshift. Some of the quasars (10 %) are radio loud. The luminosity increases with redshift up to z = 2 after which it slows down and there is a decline toward higher redshifts. The data set we are working on is extracted from Veron Cetti Catalogue of AGN and Quasar (13th Edition). Consisting of Parameters like Color indexes redshift, absolute magnitude, and magnitude. The dimension of this data set is 168940 x 13. The objective of this work is to partition the quasar based on their spectral properties (their absolute Magnitude, redshift, and Color Indexes of luminosity) and classify them with respect to the obtained clusters. To achieve our objectives, we have considered the K-means partitioning method where the optimal number of Clusters is determined by Three methods which are” The Dunn Index”,” Elbow plot”,” Silhouette Method”, Evaluating the Three Mentioned Algorithm we came to the conclusion of considering 2 optimal clusters to Carry forward our analysis. The k-means clustering algorithm, considering two optimal clusters, yields two distinctive clusters of sizes ”44468”, ”124472”. Based on the obtained clusters we applied classification techniques,” Xgboost”,” Linear Discriminant Analysis” for identifying the percentage misclassification. The Xgboost and LDA evaluates a misclassification of around 26% Respectively. So, it is safe to infer that the miss classification rate is around 26 percent for our partitioning.