scispace - formally typeset
Search or ask a question
Author

J. Stephen Downie

Bio: J. Stephen Downie is an academic researcher from University of Illinois at Urbana–Champaign. The author has contributed to research in topics: Music information retrieval & Digital library. The author has an hindex of 30, co-authored 164 publications receiving 4135 citations. Previous affiliations of J. Stephen Downie include University of Western Ontario & National Center for Supercomputing Applications.


Papers
More filters
Journal ArticleDOI
TL;DR: Les approches utilisees pour the recherche d'information musicale (RIM) sont multidisciplinaires : bibliotheconomie and science de l'information, musicologie, theorie musicale, ingenierie du son, informatique, droit et commerce...
Abstract: Les approches utilisees pour la recherche d'information musicale (RIM) sont multidisciplinaires : bibliotheconomie et science de l'information, musicologie, theorie musicale, ingenierie du son, informatique, droit et commerce... L'article vise a identifier et a expliquer la problematique de de la RIM alors qu'elle devient une discipline a part entiere, les influences historiques, l'etat-de-l'art de la recherche et les solutions potentielles. L'information musicale est multifacette - ton, temporalite, harmonie, timbre, edition, texte et bibliographie - , l'acces a chacune de ces facettes constituant un defi pour la recherche et le developpement. Mais la RIM represente egalement un defi multirepresentationnel, multiculturel, multiexperience et multidisciplinaire. Les systemes de RIM deploient differents degres d'exhaustivite representationnelle. Ils relevent generalement de deux types : les systemes analytiques ou de production, et les systemes de localisation. En conclusion, l'A. mentionne quelques ateliers et symposiums recents ainsi que les principaux projets de recherche concernant la RIM.

372 citations

Journal ArticleDOI
TL;DR: The background, structure, challenges, and contributions of MIREX are looked at and it is indicated that there are groups of systems that perform equally well within various MIR tasks.
Abstract: The Music Information Retrieval Evaluation eXchange (MIREX) is the community-based framework for the formal evaluation of Music Information Retrieval (MIR) systems and algorithms. By looking at the background, structure, challenges, and contributions of MIREX this paper provides some insights into the world of MIR research. Because MIREX tasks are defined by the community they reflect the interests, techniques, and research paradigms of the community as a whole. Both MIREX and MIR have a strong bias toward audio-based approaches as most MIR researchers have strengths in signal processing. Spectral-based approaches to MIR tasks have led to advancements in the MIR field but they now appear to be reaching their limits of effectiveness. This limitation is called the “glass ceiling” problem and the MIREX results data support its existence. The post-hoc analyses of MIREX results data indicate that there are groups of systems that perform equally well within various MIR tasks. There are many challenges facing MIREX and MIR research most of which have their root causes in the intellectual property issues surrounding music. The current inability of researchers to test their approaches against the MIREX test collections outside the annual MIREX cycle is hindering the rapid development of improved MIR systems.

310 citations

Proceedings Article
01 Jan 2004
TL;DR: A multigroup survey in an attempt to acquire information that can help eradicate false assumptions in designing MIR systems, and two major themes have been uncovered thus far that could have a significant influence on the future development of successful MIR/MDL systems.
Abstract: User studies focusing upon real-life music information needs, uses and seeking behaviours are still very scarce in the music information retrieval (MIR) and music digital library (MDL) fields. We are conducting a multigroup survey in an attempt to acquire information that can help eradicate false assumptions in designing MIR systems. Our goal is to provide an empirical basis for MIR/MDL system development. In this paper, we present our preliminary findings and analyses based on the 427 user responses we have received to date. Two major themes have been uncovered thus far that could have a significant influence the future development of successful MIR/MDL systems. First, people display “public information-seeking” behaviours by making use of collective knowledge and/or opinions of others about music such as reviews, ratings, recommendations, etc. in their music information-seeking. Second, respondents expressed needs for contextual metadata in addition to traditional bibliographic metadata.

197 citations

Proceedings Article
01 Jan 2009
TL;DR: This work introduces a new method for evaluating audio tagging algorithms on a large scale by collecting set-level judgments from players of a human computation game called TagATune.
Abstract: Search by keyword is an extremely popular method for retrieving music. To support this, novel algorithms that automatically tag music are being developed. The conventional way to evaluate audio tagging algorithms is to compute measures of agreement between the output and the ground truth set. In this work, we introduce a new method for evaluating audio tagging algorithms on a large scale by collecting set-level judgments from players of a human computation game called TagATune. We present the design and preliminary results of an experiment comparing five algorithms using this new evaluation metric, and contrast the results with those obtained by applying several conventional agreement-based evaluation metrics.

180 citations

Proceedings Article
01 Dec 2009
TL;DR: Findings show patterns at odds with findings in previous studies: audio features do not always outperform lyrics features, and combining lyrics and audio features can improve performance in many mood categories, but not all of them.
Abstract: This research examines the role lyric text can play in improving audio music mood classification. A new method is proposed to build a large ground truth set of 5,585 songs and 18 mood categories based on social tags so as to reflect a realistic, user-centered perspective. A relatively complete set of lyric features and representation models were investigated. The best performing lyric feature set was also compared to a leading audio-based system. In combining lyric and audio sources, hybrid feature sets built with three different feature selection methods were also examined. The results show patterns at odds with findings in previous studies: audio features do not always outperform lyrics features, and combining lyrics and audio features can improve performance in many mood categories, but not all of them.

171 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: This survey reviews 100+ recent articles on content-based multimedia information retrieval and discusses their role in current research directions which include browsing and search paradigms, user studies, affective computing, learning, semantic queries, new features and media types, high performance indexing, and evaluation techniques.
Abstract: Extending beyond the boundaries of science, art, and culture, content-based multimedia information retrieval provides new paradigms and methods for searching through the myriad variety of media all over the world. This survey reviews 100p recent articles on content-based multimedia information retrieval and discusses their role in current research directions which include browsing and search paradigms, user studies, affective computing, learning, semantic queries, new features and media types, high performance indexing, and evaluation techniques. Based on the current state of the art, we discuss the major challenges for the future.

1,652 citations

Book
19 Apr 2012
TL;DR: In this paper, the authors introduce concepts relevant to Information Behavior Models, Paradigms, and Theories in the study of Information Behavior Methods for Studying Information Behavior Research Results and Reflections.
Abstract: Abbreviated Contents Figures and Tables Preface Introduction and Examples Concepts Relevant to Information Behavior Models, Paradigms, and Theories in the Study of Information Behavior Methods for Studying Information Behavior Research Results and Reflections Appendix: Glossary Appendix: Questions for Discussion and Application References Index

1,347 citations

Book
01 Jan 1972
TL;DR: Invisible colleges diffusion of knowledge in scientific communities is also a way as one of the collective books that gives many advantages as discussed by the authors The advantages are not only for you, but for the other peoples with those meaningful benefits.
Abstract: No wonder you activities are, reading will be always needed. It is not only to fulfil the duties that you need to finish in deadline time. Reading will encourage your mind and thoughts. Of course, reading will greatly develop your experiences about everything. Reading invisible colleges diffusion of knowledge in scientific communities is also a way as one of the collective books that gives many advantages. The advantages are not only for you, but for the other peoples with those meaningful benefits.

1,262 citations

Book
14 Apr 2006
TL;DR: A theory of expectation is used to explain how music evokes various emotions for readers interested in cognitive science and evolutionary psychology as well as music as mentioned in this paper, which can be found in the book "Sweet Anticipation".
Abstract: A theory of expectations is used to explain how music evokes various emotions for readers interested in cognitive science and evolutionary psychology as well as music. The psychological theory of expectation that David Huron proposes in "Sweet Anticipation" grew out of experimental efforts to understand how music evokes emotions. These efforts evolved into a general theory of expectation that will prove informative to readers interested in cognitive science and evolutionary psychology as well as those interested in music. The book describes a set of psychological mechanisms and illustrates how these mechanisms work in the case of music. All examples of notated music can be heard on the Web. Huron proposes that emotions evoked by expectation involve five functionally distinct response systems: reactive responses (which engage defensive reflexes); tension responses (where uncertainty leads to stress); predictive responses (which reward accurate prediction); imaginative responses (which facilitate deferred gratification); and appraisal responses (which occur after conscious thought is engaged). For real-world events, these five response systems typically produce a complex mixture of feelings. The book identifies some of the aesthetic possibilities afforded by expectation, and shows how common musical devices (such as syncopation, cadence, meter, tonality, and climax) exploit the psychological opportunities. The theory also provides new insights into the physiological psychology of awe, laughter, and "spine-tingling chills." Huron traces the psychology of expectations from the patterns of the physical/cultural world through imperfectly learned heuristics used to predict that world to the phenomenal qualia experienced by those who apprehend the world.

1,158 citations