With the proliferation of geo-positioning techniques, there has been increasing popularity of online location-based services. Specifically, reverse top-k spatial preference queries provide such services to retrieve the users that deem a given database object as one of their top-k results. The attributes of the query object are given by the spatial distance from users' preference. However in real world, users not only consider the non-spatial attributes about the objects, but also hope to find the spatial objects based on the qualities of features in their spatial neighborhood. While reverse top-k spatial preference queries have significant amount of real-life applications such as market analysis, for example, to predict the popularity of a facility in a region, they face a great challenge to compute the score of the spatial attributes online. This paper presents a processing framework and some optimal techniques including pruning and user preference grouping methods. Theoretical analysis and experimental evaluation demonstrate the efficiency of the proposed algorithms and the improvement on running time and I/O.