The Differential Evolution (DE) has stood as a cornerstone of Evolutionary Computation (EC), inspiring numerous approaches. Despite its foundational role, the selection stage of DE has received little attention, with only 2% of documented modifications in the literature on this aspect. Recent research has underscored the potential for significant algorithmic improvement through thoughtful modifications to this critical stage, particularly in accelerating the exploitation phase. This study introduces a novel EC strategy rooted in DE principles. To enhance algorithmic exploration, a systematic decision-making process regarding function evaluations is employed to select between two of the most prevalent mutations in the field. Similarly, a new selection operator is introduced to augment the exploitation phase by comparing each individual with its respective 25% neighborhood population. The proposed algorithm, Differential Evolution with Selection by Neighborhood (DESN), undergoes comprehensive evaluation against eight classical and recent approaches, leveraging the CEC-2017 set of benchmark functions.