Unsupervised detection of music boundaries by time series structure features
Publication Type:
Conference PaperSource:
AAAI Conf. on Artificial Intelligence, AAAI Press, Toronto, Canada, p.1613-1619 (2012)URL:
http://www.aaai.org/ocs/index.php/AAAI/AAAI12/paper/view/4907Keywords:
Time Series Structure; FeaturesAbstract:
Locating boundaries between coherent and/or repetitive segments of a time series is a challenging problem pervading many scientific domains. In this paper we propose an unsupervised method for boundary detection, combining three basic principles: novelty, homogeneity, and repetition. In particular, the method uses what we call structure features, a representation encapsulating both local and global properties of a time series. We demonstrate the usefulness of our approach in detecting music structure boundaries, a task that has received much attention in recent years and for which exist several benchmark datasets and publicly available annotations. We find our method to significantly outperform the best accuracies published so far. Importantly, our boundary approach is generic, thus being applicable to a wide range of time series beyond the music and audio domains.
PDFFile:
An Interaction-Based Approach to Semantic Alignment
Publication Type:
Journal ArticleSource:
Journal of Web Semantics, Volume 12-13, p.131-147 (2012)Keywords:
semantic alignment; agent interaction context; interaction model; communication product; alignment protocol; matching criteriaAbstract:
We tackle the problem of semantic heterogeneity in the context of agent communication and argue that solutions based solely on ontologies and ontology matching do not capture adequately the richness of semantics as it arises in dynamic and open multiagent systems. Current solutions to the semantic heterogeneity problem in distributed systems usually do not address the contextual nuances of the interaction underlying an agent communication. The meaning an agent attaches to its utterances is, in our view, very relative to the particular dialogue in which it may be engaged, and that the interaction model specifying its dialogical structure and its unfolding should not be left out of the semantic alignment mechanism. In this article we provide the formal foundation of a novel, interaction-based approach to semantic alignment, drawing from a mathematical construct inspired from category theory that we call the communication product. In addition, we describe a simple alignment protocol which, combined with a probabilistic matching mechanism, endows an agent with the capacity of bootstrapping —by repeated successful interaction— the basic semantic relationship between its local vocabulary and that of another agent. We have also implemented the alignment technique based on this approach and prove its viability by means of an abstract experimentation and a thorough statistical analysis.
A formal model for Situated Semantic Alignment
Publication Type:
Conference PaperSource:
6th International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS'07, IFAAMAS, Honolulu, Hawaii, USA, p.1270-1277 (2007)Keywords:
semantic alignment; distributed logic; channel refinementAbstract:
Ontology matching is currently a key technology to achieve the semantic alignment of ontological entities used by knowledge-based applications, and therefore to enable their interoperability in distributed environments such as multiagent systems. Most ontology matching mechanisms, however, assume matching prior integration and rely on semantics that has been coded a priori in concept hierarchies or external sources. In this paper, we present a formal model for a semantic alignment procedure that incrementally aligns differing conceptualisations of two or more agents relative to their respective perception of the environment or domain they are acting in. It hence makes the situation in which the alignment occurs explicit in the model. We resort to Channel Theory to carry out the formalisation.
