Abstract: The explosive growth of mobile apps in recent years makes it much more difficult for users to find out interesting apps. For this reason, online app markets, e.g., the Google Play market, have employed recommender systems. Such systems construct recommending networks of mobile apps so that they alleviate the challenge of app discovery. However, research efforts on the recommender systems are mainly focusing on the improvement of recommending accuracy. Little attention has been paid to measure and optimize the navigating effects of the recommending networks. To be specified, rare works in the literature have focused on advancing the efficiency of helping users explore more apps while discovering them with fewer jumps. This study therefore initially addresses and formulates such a problem. It further proposes to reconstruct the recommending networks after they have been formed by the recommender systems. Since mobile apps in the online markets have constituted complex networks, this study designs reconstructing schemes leveraging the complex network metrics and methods. Particularly, based on specific complex network measurements, e.g., the number of SCCs (strongly connected components), the APL (Average path length) and the node centrality, this study proposes two reconstructing schemes. After all, real-data evaluations have verified the effectiveness of the schemes proposed by this study.