Automated vibration-based structural health monitoring (SHM) strategies have been recently proven promising in the presence of aging and material deterioration threatening the structural safety of civil structures. Within such a framework, ensuring high quality and informative data is a critical aspect, highly dependent on the deployment of the sensors in the network and on their capability to provide damage-sensitive features to be exploited. This paper presents a novel data-driven approach to the optimal sensor placement, devised to identify sensor locations that maximize the information effectiveness for SHM purposes. The optimization of the sensor network is addressed by means of a deep neural network (DNN) equipped with an attention mechanism, a state-of-the-art technique in natural language processing useful to focus on a limited number of important components in the information stream. The trained attention mechanism eventually allows to quantify the relevance of each sensor in terms of the so-called attention scores, and therefore enables to identify the most useful input channels to solve the relevant downstream SHM task. With reference to the damage localization task, framed here as a classification problem handling a set of predefined damage scenarios, the DNN is trained to locate damage on labeled data, that have been formerly simulated to emulate the effects of damage under different operational conditions. The capabilities of the proposed method are demonstrated by referring to an eight-story shear building, characterized by damage states of unknown severity and possibly located at any story.