摘要：The purpose of this study is to explore the information retrieval process in scientific data and to better understand the concepts and internal relationships of metadata and relevance criteria. Qualitative and quantitative analyses were performed using interview and eye movement data from 36 subjects. The results show that users paid attention to 45 types of metadata and used nine relevance criteria to judge the relevance of scientific data. There was a complex relationship between the metadata and criteria, mainly manifesting as one stimulus–multiple responses and multiple stimuli–one response. Metadata associated with the relevance criterion of topicality is the most complex, which includes common metadata and subject-related metadata. Metadata associated with the other relevance criteria (such as quality and authority) has no obvious professional characteristics. What’s more, because of the essential difference between scientific data and documents, users use different criteria. When retrieving data, users pay more attention to the availability of data and whether they can be further analyzed and processed. This study clarifies the concepts of metadata and relevance criteria as well as their roles in relevance judgments. In addition, this study deepens the understanding of the scientific data relevance judgments and their cognitive process and provides a theoretical basis for improving scientific data-sharing platforms.
