DATAMINING APPROACH TO HUMAN CANCER DIAGNOSIS: A REVIEW


Cancer is an intriguing disease in which certain set of cells continue to grow unnecessarily and may later invade other parts of the body. Microarray data are mostly employed in cancer researches where early diagnosis of cancer disease is very significant in determining the nature of treatment and its survivability. Data mining classification techniques are apt to solve this problem where unknown sample of cells can be categorized into diseased or normal cells. Microarray data are usually characterized with high dimensionality features. The main objective of dimensionality reduction is to augment the diagnosis accuracy of classification algorithms by removing redundant or irrelevant features and minimize computation cost. In this work, we try to explore the state-of-arts of research being carried out using the data mining approaches to augment the cancer diagnosis. This survey briefly discusses the processes involved in cancer diagnosis and elaborate on hybrid features selection used in cancer diagnosis and analysis.