The type of data, and how it is collected affects the way it can be aggregated to form
justifications. If users could thoughtfully provide accurate scores on a consistent
scale for each item, or numerical descriptions of themselves with their preferences
expressed to a degree of certainty, an RS could quite comfortably make some relevant recommendations. Of course, the aesthetic preference is usually to limit the
explicit information required from the user and hence enhance the interactive experience. We will briefly consider the different types of data that systems are able to
obtain and how this might affect the suitability of certain aggregation functions.