Density-Based Top-K Spatial Textual Clusters Retrieval

So-called spatial web queries retrieve content representing points of interest, such that the interest have descriptions are relevant to query keywords and located close a location. Two broad categories exist. The first encompasses single objects each satisfy arguments. Most proposals belong this category. second category, which paper's proposal belongs, support exploratory user behavior sets represent regions space may be user. Specifically, paper proposes new type query, top- $k$ textual cluster retrieval ( -STC) returns clusters (i) location, (ii) contain with regard given keywords, (iii) an object density exceeds threshold. To compute we propose DBSCAN-based approach OPTICS-based rely on on-line density-based clustering exploit early stop conditions. Empirical studies real data offer evidence can find good quality capable excellent performance.