We introduce a computational model that elicits influential subjects from consumer perceptions in social media. We devise a novel Genetic Algorithm to improve clustering of data in semantically coherent groups. Our model produces insights for two fundamental brand equity dimensions: brand awareness and brand meaning. We empirically apply the model to guide the analysis of over 280.000 tweets related to the Uber transportation network. Our experimental results reveal negative and positive consumer perceptions for specific subjects related to Uber.