Ranking application software retrieval results using application attributes and context

Nowadays, various search engines provide an easy way to search an application that meets the user requirements. However, there are an increasing number of similar feature applications or a large number of applications that are intentionally implanted popular keywords in their information. Thus, search results may have the tendency not to satisfy the user needs. In this paper, we present a new method to rank application search results focusing on two criteria: application attributes and application context. Using application attributes and its context, we propose three ranking scores; rating score, content relevant score and context score. From our experiment, applying all three scores for application retrieval, significantly gives a better search quality results than the baseline and the selected combined scores using r-precision and normalized discounted cumulative gain score metric.