Abstract：Collaborative matrix factorization mechanism for group recommendation in big data-based library systems Purpose Academic groups are designed specifically for researchersA group recommendation procedure is essential to support scholarsresearch-based social activitiesHowevergroup recommendation methods are rarely applied in online libraries and they often suffer from scalability problem in big data contextThe purpose of this paper is to facilitate academic group activities in big data-based library systems by recommending satisfying articles for academic groupsDesignmethodologyapproach The authors propose a collaborative matrix factorizationCoMFmechanism and implement paralleled CoMF under Hadoop frameworkIts rationale is collaboratively decomposing researcher-article interaction matrix and group-article interaction matrixFurthermorethree extended models of CoMF are proposedFindings Empirical studies on CiteULike data set demonstrate that CoMF and three variants outperform baseline algorithms in terms of accuracy and robustnessThe scalability evaluation of paralleled CoMF shows its potential value in scholarly big data environmentResearch limitationsimplications The proposed methods fill the gap of group-article recommendation in online libraries domainThe proposed methods have enriched the group recommendation methods by considering the interaction effects between groups and membersThe proposed methods are the first attempt to implement group recommendation methods in big data contextsPractical implications The proposed methods can improve group activity effectiveness and information shareability in academic groupswhich are beneficial to membership retention and enhance the service quality of online library systemsFurthermorethe proposed methods are applicable to big data contexts and make library system services more efficientSocial implications The proposed methods have potential value to improve scientific collaboration and research innovationOriginalityvalue The proposed CoMF method is a novel group recommendation method based on the collaboratively decomposition of researcher-article matrix and group-article matrixThe process indirectly reflects the interaction between groups and memberswhich accords with actual library environments and provides an interpretable recommendation resultSearch and indexHelp academic groups find satisfactory articles in big data-based library systemsIn additionbecause these methods also involve the interaction between researchers and articlesthey can also be related toconsulting servicesandcollection classificationThis article mainly proposes a method based on the synergy matrix decompositionCOMFto recommend articles to academic groupsThis involves the category ofdata explorationbecause COMF technology and its variants are used to extract information and generate recommendations from a large number of scholars-articles interactive data