【摘要】：Purpose: In order to further the understanding of Chinese Web users' image-seeking behavior, this study explores the kinds of images that Chinese Web users seek online and how they express their requests.Design/methodology/approach: We used five pairs of simulated keywords to collect 893 image-seeking questions from Baidu Zhidao. Then, we revised the subject category of questions to analyze popular image needs. In addition, we conducted content analysis and descriptive statistical analysis to identify image-seeking motivations and image features used in the requests in terms of the two theories of image feature classification and image use.Findings: Among the 893 questions, the image searches for entertainment accounted for 47.59%, more than the searches for professional knowledge(37.40%) and personal daily activities(15.01%). With regard to motivation, over 60% of the questions were identified as used for learning, which is well over the proportion of questions used for illustrating. Thus, these questions requested images as sources of data rather than sources of objects. Non-visual features(47.58%) were used most frequently in question descriptions, slightly higher than semantic features(45.96%). Users who lacked domain knowledge tended to use general words rather than specific words to describe their requests. However, not many users used syntactic features when seeking images. Nevertheless, most of the users had a fairly clear idea about what the target image should look like.Research limitations: We studied only one question and answer(QA) community using five pairs of simulated keywords. Practical implications: The findings should be helpful in strengthening the functionality of QA systems, promoting the theories of image feature classification, and shedding light on information literacy training.Originality/value: This study is one of the first research efforts that discusses Chinese Web users' daily image searches and querying behavior in natural language in a QA community, which should help to further the understanding of the principles of image-seeking behavior among Chinese Web users.