The dimensionality reduction problem can be handled by employing feature selection techniques. Feature selection (FS) is a key preprocessing technique which selects more informative and useful features from data based on some criterion. Researchers have suggested many feature selection algorithms that differ in their selection criteria. Furthermore, feature selection is intrinsically a multi-objective problem with several conflicting objectives like size, redundancy and error rate. Thus, Multi-Objective Genetic Algorithms (MOGAs) are an obvious choice for solving the feature selection problem. Moreover, Micro-GA is used with a very small size population in cooperated with different type of elitisms. This work presents a Multi-objective Micro-CHC based Algorithm (MO-Micro-CHC) for feature selection. The proposed hybrid algorithm exploits the peculiarities of CHC (Cross generational elitist selection, Heterogeneous recombination, and Cataclysmic mutation), NSGA-II (Non- Dominated Sorting Genetic Algorithm-II) the most popular MOGA and Micro-GA to the advantage of arriving at better Pareto optimal solutions. The algorithm has been validated and compared to several other similar approaches on many datasets available from UCI data repository. The comparison endorses the superiority of the suggested approach.