Mobile app repositories have been largely used in scientific research as large-scale, highly adaptive crowdsourced information systems. These software platforms can potentially nourish multiple software and requirements engineering tasks based on user reviews and other natural language documents, including feedback analysis, recommender systems and topic modelling. Consequently, researchers often endeavour to overcome domain-specific challenges, including integration of heterogeneous data sources, large-scale data collection and adaptation of a publicly available data set for a given research scenario. In this paper, we present MApp-KG, a combination of software resources and data artefacts in the field of mobile app repositories to support extended knowledge generation tasks. Our contribution aims to provide a framework for automatically constructing a knowledge graph modelling a domain-specific catalogue of mobile apps. Complementarily, we distribute MApp-KG in a public triplestore and as a static data snapshot, which may be promptly employed for future research and reproduction of our findings.

翻译：移动应用仓库已被广泛用作大规模、高度自适应的众包信息系统，在科学研究中发挥着重要作用。这些软件平台能够基于用户评论及其他自然语言文档（包括反馈分析、推荐系统和主题建模）为多项软件与需求工程任务提供支撑。因此，研究者常常需要克服领域特定的挑战，包括异构数据源整合、大规模数据采集以及针对特定研究场景调整公开可用的数据集。本文提出了MApp-KG系统——一种融合移动应用仓库领域软件资源与数据工件的组合方案，旨在支持扩展的知识生成任务。我们的贡献在于提供了一套自动化构建领域特定移动应用目录知识图谱的框架。作为补充，我们还以公共三元组存储和静态数据快照的形式分发MApp-KG，便于未来研究及成果复现。