DNA methylation is a type of epigenetic modification which involves the addition of a methyl group to DNA via DNA methyltransferase (DNMT). In eukaryotic cells, the DNA methylation occurs at the 5’ carbon position of the cytosine residue in the cytosine guanine (CpG) dinucleotide context. DNA methylation is crucial to normal organismal development, which includes cellular differentiation, genomic imprinting, X-chromosome inactivation, suppression of retrovirus transcription, etc. In addition, an accumulating volume of evidence has suggested that aberrant DNA methylation is a major contributor to the onset and progression of cancers. In this paper, we build a computational model to identify those CpG islands that are methylated in colon cancer but unmethylated in normal cells. We develop a highly accurate prediction model for those CpG islands whose methylation differentiation is related to colon cancer and evaluate these models through extensive cross-validation and generalization testing experiments.