Forecasting keyword activities in social networking sites has been the subject of many studies, as such activities represent, in many cases, a direct estimate of the spread of real-world phenomena, e.g. box-office revenues or flu epidemic. Most of these studies rely on point-wise, regression-like prediction algorithms and focus on few, usually unambiguous, keywords. We study in this paper the impact of keyword activity on three different problems: a) classification of keywords according to the increase of their activity in the near future; b) prediction of the activity value of each keyword in the near future; c) ranking of a set of keywords according to their future activity values. It is the first time, to our knowledge, that such dimensions are evaluated in this framework. Our experiments are conducted on a large dataset built by monitoring Twitter over a year. The different methods tested are evaluated using standard scores as well as a newly defined, application driven quality measure.