Improving Active Learning With Sharp Data Reduction

dc.creatorSaito P.T.M.
dc.creatorDe Rezende P.J.
dc.creatorFalcao A.X.
dc.creatorSuzuki C.T.N.
dc.creatorGomes J.F.
dc.date2012
dc.date2015-06-25T20:27:29Z
dc.date2015-11-26T15:24:11Z
dc.date2015-06-25T20:27:29Z
dc.date2015-11-26T15:24:11Z
dc.date.accessioned2018-03-28T22:33:04Z
dc.date.available2018-03-28T22:33:04Z
dc.identifier9788086943794
dc.identifier20th International Conference In Central Europe On Computer Graphics, Visualization And Computer Vision, Wscg 2012 - Conference Proceedings. , v. , n. PART 1, p. 27 - 34, 2012.
dc.identifier
dc.identifier
dc.identifierhttp://www.scopus.com/inward/record.url?eid=2-s2.0-84877941420&partnerID=40&md5=751cee646eecf3ec1f3ba3cc6fdfab22
dc.identifierhttp://www.repositorio.unicamp.br/handle/REPOSIP/90749
dc.identifierhttp://repositorio.unicamp.br/jspui/handle/REPOSIP/90749
dc.identifier2-s2.0-84877941420
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/1260629
dc.descriptionStatistical analysis and pattern recognition have become a daunting endeavour in face of the enormous amount of information in datasets that have continually been made available. In view of the infeasibility of complete manual annotation, one seeks active learning methods for data organization, selection and prioritization that could help the user to label the samples. These methods, however, classify and reorganize the entire dataset at each iteration, and as the datasets grow, they become blatantly inefficient from the user's point of view. In this work, we propose an active learning paradigm which considerably reduces the non-annotated dataset into a small set of relevant samples for learning. During active learning, random samples are selected from this small learning set and the user annotates only the misclassified ones. A training set with new labelled samples increases at each iteration and improves the classifier for the next one. When the user is satisfied, the classifier can be used to annotate the rest of the dataset. To illustrate the effectiveness of this paradigm, we developed an instance based on the optimum path forest (OPF) classifier, while relying on clustering and classification for the learning process. By using this method, we were able to iteratively generate classifiers that improve quickly, to require few iterations, and to attain high accuracy while keeping user involvement to a minimum. We also show that the method provides better accuracies on unseen test sets with less user involvement than a baseline approach based on the OPF classifier and random selection of training samples from the entire dataset.
dc.description
dc.descriptionPART 1
dc.description27
dc.description34
dc.descriptionAngluin, D., Queries and concept learning (1988) Machine Learning, 2, pp. 319-342
dc.descriptionCappabianco, F.A.M., Ide, J.S., Falcäo, A.X., Li, C.-S.R., Automatic subcortical tissue segmentation of mr images using optimum-path forest clustering (2011) International Conference on Image Processing (ICIP), pp. 2653-2656
dc.descriptionCheng, Y., Mean shift, mode seeking, and clustering (1995) TPAMI, 17 (8), pp. 790-799
dc.descriptionCohn, D.A., Ghahramani, Z., Jordan, M.I., Active learning with statistical models (1996) JAIR, 4, pp. 129-145
dc.descriptionDa Silva, A.T., Falcäo, A.X., Magalhäes, L.P., A new CBIR approach based on relevance feedback and optimum-path forest classification (2010) Journal of WSCG, pp. 73-80
dc.descriptionDa Silva, A.T., Falcäo, A.X., Magalhäes, L.P., Active learning paradigms for CBIR systems based on optimum-path forest classification (2011) Pattern Recognition, 44, pp. 2971-2978
dc.descriptionDavis, D.T., Hwang, J.N., Attentional focus training by boundary region data selection (1992) Intern. Joint Conference on Neural Networks (IJCNN), 1, pp. 676-681
dc.description(2011) Biometrics Database Distribution, , www.nd.edu/~cvrl/CVRL/Data_Sets.html, Faces The Computer Vision Laboratory, University of Notre Dame
dc.descriptionHolub, A., Perona, P., Burl, M.C., Entropy-based active learning for object recognition (2008) CVPRW, pp. 1-8
dc.descriptionJain, P., Kapoor, A., Active learning for large multi-class problems (2009) IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp. 762-769
dc.descriptionKapoor, A., Grauman, K., Urtasun, R., Darrell, T., Gaussian processes for object categorization (2010) International Journal of Computer Vision (IJCV), 88, pp. 169-188
dc.descriptionLi, X., Wang, L., Sung, E., Multi-label SVM active learning for image classification (2004) International Conference on Image Processing (ICIP), 4, pp. 2207-2210
dc.descriptionPapa, J.P., Falcäo, A.X., De Albuquerque, V.H.C., Tavares, J.M.R.S., Efficient supervised optimum-path forest classification for large datasets (2012) Pattern Recognition, 45, pp. 512-520
dc.descriptionPapa, J.P., Falcäo, A.X., Suzuki, C.T.N., Supervised pattern classification based on optimum-path forest (2009) Intern. Journal of Imaging Systems and Technology (IJIST), 19 (2), pp. 120-131
dc.description(2011) Pen-Based Recognition of Handwritten Digits Dataset, , archive.ics.uci.edu/ml/datasets/Pen- Based+Recognition+of+Handwritten+Digits, Pendigits
dc.descriptionQi, G.-J., Hua, X.-S., Rui, Y., Tang, J., Zhang, H.-J., Two-dimensional multilabel active learning with an efficient online adaptation model for image classification (2009) IEEE Transact. on Pattern Analysis and Machine Intel., 31 (10), pp. 1880-1897
dc.descriptionRocha, L.M., Cappabianco, F.A.M., Falcäo, A.X., Data clustering as an optimum-path forest problem with applications in image analysis (2009) Intern. Journal of Imaging Systems and Technology (IJIST), 19 (2), pp. 50-68
dc.descriptionTong, S., Chang, E., Support vector machine active learning for image retrieval (2001) ICM, pp. 107-118. , ACM
dc.descriptionTong, S., Koller, D., Support vector machine active learning with applications to text classification (2002) Journal of Machine Learning Research (JMLR), 2, pp. 45-66
dc.descriptionYan, R., Yang, J., Hauptmann, A., Automatically labeling video data using multi-class active learning (2003) IEEE Intern. Conference on Computer Vision (ICCV), 1, pp. 516-523
dc.languageen
dc.publisher
dc.relation20th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2012 - Conference Proceedings
dc.rightsfechado
dc.sourceScopus
dc.titleImproving Active Learning With Sharp Data Reduction
dc.typeActas de congresos


Este ítem pertenece a la siguiente institución