With rising population, cities face new challenges. A key challenge for city administrators is to address the overall well-being of its citizens. This includes both physical and emotional health. Towards this objective, cities around the world are heavily investing in green mobility with support for sustainable modes (e.g. public transport, cycling, walking) as an alternative to individual motorized transport using combustion engine. However, very little attention is paid towards identifying the effect of green mobility on the emotional states of citizens. Several studies show a link between an upbeat emotional state and physical signs of good health. Furthermore, as urban centres expand it is imperative to find a balanced combination of physical and emotional health during last mile urban commute. In this paper, we try to find a feasible method for urban emotion detection in the age of last mile green mobility. Our approach relies on Machine Learning (ML) techniques to predict emotions with real-time data.