A definition of information fusion (IF) as a field of research can benefit researcherswithin the field, who may use such a definition when motivating their own work and evaluatingthe contributions of others. Moreover, it can enable researchers and practitioners outside thefield to more easily relate their own work to the field and more easily understand the scope of IFtechniques and methods. Based on strengths and weaknesses of existing definitions, a definitionis proposed that is argued to effectively fulfill the requirements that can be put on a definitionof IF as a field of research. Although the proposed definition aims to be precise, it does not fullycapture the richness and versatility of the IF field. To address that limitation, we highlight sometopics to explore the scope of IF, covering the systems perspective of IF and its relation to ma-chine learning, optimization, robot behavior, opinion aggregation, and databases.