Feature selection (FS) can be defined as the problem of finding the minimal number of features from an original set with the minimum information loss. Since FS problems are known as NP-hard problems, it is necessary to investigate a fast and an effective search algorithm to tackle this problem. In this paper, two incremental hill-climbing techniques (QuickReduct and CEBARKCC) are hybridized with the binary ant lion optimizer in a model called HBALO. In the proposed approach, a pool of solutions (ants) is generated randomly and then enhanced by embedding the most informative features in the dataset that are selected by the two filter feature selection models. The resultant population is then used by BALO algorithm to find the best solution. The proposed binary approaches are tested on a set of 18 well-known datasets from UCI repository and compared with the most recent related approaches. The experimental results show the superior performance of the proposed approaches in searching the feature space for optimal feature combinations.