A novelty detection approach to classification Export

@inproceedings{citeulike:3366638, abstract = {Novelty Detection techniques are concept-learning methods that proceed by recognizingpositive instances of a concept rather than differentiating between its positive and negative instances. Novelty Detection approaches consequently require very few, if any, negative training instances. This paper presents a particular Novelty Detection approach to classification that uses a Redundancy Compression and NonRedundancy Differentiation technique based on the (Gluck \& Myers 1993) model of the hippocampus, a part of the brain critically involved in learning and memory. In particular, this approach consists of training an autoencoder to reconstruct positive input instances at the output layer and then using this autoencoder to recognize novel instances. Classification is possible, after training, because positive instances are expected to be reconstructed accurately while negative instances are not. The purpose of this paper is to compare Hippo, the system that implements this technique, to C4.5 and feedforward neural network classification on several applications.}, author = {Japkowicz, Nathalie and Myers, Catherine and Gluck, Mark}, booktitle = {In Proceedings of the Fourteenth Joint Conference on Artificial Intelligence}, citeulike-article-id = {3366638}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.40.3663}, keywords = {backpropagation, detection, network, neural, novelty, unsupervised}, pages = {518--523}, posted-at = {2008-10-02 15:23:26}, priority = {2}, title = {A novelty detection approach to classification}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.40.3663}, year = {1995} }

TY - CONF ID - citeulike:3366638 L3 - citeulike-article-id:3366638 N2 - Novelty Detection techniques are concept-learning methods that proceed by recognizingpositive instances of a concept rather than differentiating between its positive and negative instances. Novelty Detection approaches consequently require very few, if any, negative training instances. This paper presents a particular Novelty Detection approach to classification that uses a Redundancy Compression and NonRedundancy Differentiation technique based on the (Gluck & Myers 1993) model of the hippocampus, a part of the brain critically involved in learning and memory. In particular, this approach consists of training an autoencoder to reconstruct positive input instances at the output layer and then using this autoencoder to recognize novel instances. Classification is possible, after training, because positive instances are expected to be reconstructed accurately while negative instances are not. The purpose of this paper is to compare Hippo, the system that implements this technique, to C4.5 and feedforward neural network classification on several applications. EP - 523 TI - A novelty detection approach to classification T2 - In Proceedings of the Fourteenth Joint Conference on Artificial Intelligence SP - 518 KW - backpropagation KW - detection KW - network KW - neural KW - novelty KW - unsupervised AU - Japkowicz, Nathalie AU - Myers, Catherine AU - Gluck, Mark PY - 1995/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.40.3663 ER -