In the last decades, the remarkable advance in microarray technology opened ahuge opportunities in genomic researches and especially in cancer researches tomove from clinical decisions and standard medicine toward personalise medicine.The analysis of gene expression level may reveal a lot of informations about thecancer type, it’s outcomes also allow the possibility to predict about the besttherapy in order to improve the survival rate.Gene expression microarrays is a new breakthrough technology developed inthe late 1990s 1 that can measure the gene expression level of thousands ofgenes correspond to different samples or experiments simultaneously 2. In whichmany solution schemes for cancer classification and therapy process on molecular and cellular level may be concluded from the analysis and the comparisonof the generated data through different experiments3. Microarray technologyhave tow variant in the market 3,(1)cDNA microarrays-On Spotted array- and(2)oligonucleotide microarrays-On GeneChip-. cDNA microarrays are cheaperand more flexible as custom-made arrays it was developed by Stanford University. While oligonucleotide arrays (developped by Affymetrix) are more automated, stable, and easier to be compared between different experiments34. The data produced by microarrays technology represent the result of thousandsof genes for few experiments where this matrix can be used to evaluate the variation of gene through samples or the interaction of genes in different samples.Since DNA micro-array technology allows to analyse the gene data quickly andat one time in order to get the expression pattern of a huge amount of genessimultaneously5, gene expression data are unique in their nature due to threereasons:(1)their high dimensionality(more than thousands of genes),(2)the publicly available data are very small just hundred or fewer of samples,(3) a bigpartial of the genes are irrelevant in cancer classification and analysis, wherethe problem is to find the difference between cancerous gene expression tissuesand non-cancerous tissues. So, In order to handle those kind of data researchersproposed that feature selection and/or dimensionality reduction is a relevantprocess in order to take advantage of the data and to converge toward accurateclassifiers. Several machine learning methods have been used in caner classification, yet recently deep learning start to be investigated as well in this processdue to its ability to work on raw and high dimensional data.The paper cites the different machine learning and deep learning recent achievements in gene expression data analysis and cancer classification, also discussand compare the deep learning latest research in cancer classification. As wellWe present a comparison between four classical machine learning classifiers: support vector machine, naive bayes, k-nearest neighbours, shallow neural networksand a proposed feed forward multi layer perceptron to test the impact of deeplearning on the cancer classification problems. Our comparison was tested onseven publicly available Cancer datasets of the omnibus library.