The journal provides an international forum for the publication of theory, algorithms, analysis and experiments across the broad area of information retrieval. Topics of interest include search, indexing, analysis, and evaluation for applications such as the web, social and streaming media, recommender systems, and text archives. This includes research in areas such as human factors, artificial intelligence, and domain-specific challenges in information retrieval.
The ideal article may be theoretical, experimental, analytical or applied. A theoretical article will report a significant conceptual advance in the design of algorithms or other processes for some information retrieval task. It will establish the validity or potential validity of the proposed ideas in terms of their relation to already accepted ideas and/or in terms of some modest prototype experiment or simulation. An experimental article will report on a test of one or more theoretical ideas in a laboratory or natural setting. Experimental articles will be reviewed for both scientific and statistical merit and will be expected to discuss the limitations and generality of the reported results. Analytical articles will report on the results of analysis of data collected in settings such as user studies, surveys and log analysis. These articles will be reviewed based on statistical rigor, the novelty and generalizability of the findings, and the reproducibility of the methods. An application article will report the successful application of some already established technique to a significant real-world problem involving information retrieval.
The Information Retrieval Journal overlaps with a variety of technical and behavioral fields. Articles on such technical issues as compression and optimization, and on issues of human behavior and cognition are appropriate insofar as they bear specifically on the issues of methods, tasks or media as outlined above. Variations from these prototypes, such as critical reviews of existing work and significant tutorials will be considered if they make a clear contribution to the field. Preference will be given to articles which unify concepts across several traditional disciplinary boundaries, with specific application to problems of information retrieval.