The skyrocketing growth of mobile apps and mobile devices has significantly fueled the competition among the developers. App developers endeavoured the store capabilities as an opportunity to analyse data in order to provide improvement recommendations for the evolution of any given app. Previous research shows that app developers mostly rely on in-domain (i.e., same domain or same app) data to improve their apps. However, relying on in-domain data causes low diversity and lacks novelty in recommended features. In this work, we present an approach to automatically classify, group and rank popular features from cross-domain apps. We follow 3 steps- (1) identify cross-domain apps that are relevant to the target app in terms of feature relevance and determine non-similar feature co-existence among relevant apps, (2) filter and group cross-domain features that are complementary relevant to target app using semantic feature relevance technique, (3) rank and prioritize popular features from cross-domain for adoption based on the distribution of domains, apps, popularity, features in the relevant feature group. We run extensive experiments on 100 target apps from 10 categories and 15200 cross-domain apps from 31 categories. We found encouraging results from the experiments, especially, the semantic feature grouping technique outperforms two other baseline techniques. The empirical evaluation validates the efficacy of our approach, thereby providing personalised feature recommendations to developers and enhance user serendipity.