Cross-domain classification using generalized domain acts
نویسندگان
چکیده
Cross-domain classi cation for speech understanding is an interesting research problem because of the need for portable solutions in the design for spoken dialogue systems. In this paper, a twotier classi er is proposed for speech understanding. The rst tier consists of domain independent dialogue acts while the second tier consists of application actions that are domain speci c. A maximum likelihood and a minimum classi cation error formulation are proposed for the rst tier of the classi er, i.e., for dialogue act classi cation. The performance of the classi er is investigated for three application domains. Cross-domain classi cation error is two to four times higher than indomain classi cation error. A 10-15% reduction in cross-domain classi cation error rate is achieved by adding generic domain independent training data for each dialogue act and by mapping words to semantic concepts.
منابع مشابه
Sample-oriented Domain Adaptation for Image Classification
Image processing is a method to perform some operations on an image, in order to get an enhanced image or to extract some useful information from it. The conventional image processing algorithms cannot perform well in scenarios where the training images (source domain) that are used to learn the model have a different distribution with test images (target domain). Also, many real world applicat...
متن کاملDeep Unsupervised Domain Adaptation for Image Classification via Low Rank Representation Learning
Domain adaptation is a powerful technique given a wide amount of labeled data from similar attributes in different domains. In real-world applications, there is a huge number of data but almost more of them are unlabeled. It is effective in image classification where it is expensive and time-consuming to obtain adequate label data. We propose a novel method named DALRRL, which consists of deep ...
متن کاملPhoneme Classification Using Temporal Tracking of Speech Clusters in Spectro-temporal Domain
This article presents a new feature extraction technique based on the temporal tracking of clusters in spectro-temporal features space. In the proposed method, auditory cortical outputs were clustered. The attributes of speech clusters were extracted as secondary features. However, the shape and position of speech clusters change during the time. The clusters temporally tracked and temporal tra...
متن کاملNumerical solution of Fredholm integral-differential equations on unbounded domain
In this study, a new and efficient approach is presented for numerical solution of Fredholm integro-differential equations (FIDEs) of the second kind on unbounded domain with degenerate kernel based on operational matrices with respect to generalized Laguerre polynomials(GLPs). Properties of these polynomials and operational matrices of integration, differentiation are introduced and are ultili...
متن کاملFractional Order Generalized Thermoelastic Functionally Graded Solid with Variable Material Properties
In this work, a new mathematical model of thermoelasticity theory has been considered in the context of a new consideration of heat conduction with fractional order theory. A functionally graded isotropic unbounded medium is considered subjected to a periodically varying heat source in the context of space-time non-local generalization of three-phase-lag thermoelastic model and Green-Naghdi mod...
متن کامل