Semi-Supervised Learning Literature Survey

Semi-Supervised Learning Literature Survey. Zhu, X. (. Technical Report University of Wisconsin-Madison Department of Computer Sciences, 2005. Accepted: 2012-03-15T17:19:12Z

Paper abstract bibtex

We review some of the literature on semi-supervised learning in this paper. Traditional classifiers need labeled data (feature / label pairs) to train. Labeled instances however are often difficult, expensive, or time consuming to obtain, as they require the efforts of experienced human annotators. Meanwhile unlabeled data may be relatively easy to collect, but there has been few ways to use them. Semi-supervised learning addresses this problem by using large amount of unlabeled data, together with the labeled data, to build better classifiers. Because semi-supervised learning requires less human effort and gives higher accuracy, it is of great interest both in theory and in practice.

@techreport{zhu_semi-supervised_2005,
	type = {Technical {Report}},
	title = {Semi-{Supervised} {Learning} {Literature} {Survey}},
	url = {https://minds.wisconsin.edu/handle/1793/60444},
	abstract = {We review some of the literature on semi-supervised learning in this paper. Traditional classifiers need labeled data (feature / label pairs) to train. Labeled instances however are often difficult, expensive, or time consuming to obtain, as they require the efforts of experienced human annotators. Meanwhile unlabeled data may be relatively easy to collect, but there has been few ways to use them. Semi-supervised learning addresses this problem by using large amount of unlabeled data, together with the labeled data, to build better classifiers. Because semi-supervised learning requires less human effort and gives higher 
accuracy, it is of great interest both in theory and in practice.},
	language = {en},
	urldate = {2022-03-20},
	institution = {University of Wisconsin-Madison Department of Computer Sciences},
	author = {Zhu, Xiaojin (Jerry)},
	year = {2005},
	note = {Accepted: 2012-03-15T17:19:12Z},
}

Downloads: 0

{"_id":"3r5f6MKcehNxtSiwz","bibbaseid":"zhu-semisupervisedlearningliteraturesurvey-2005","downloads":0,"creationDate":"2017-09-14T16:34:37.294Z","title":"Semi-Supervised Learning Literature Survey","author_short":["Zhu, X. (."],"year":2005,"bibtype":"techreport","biburl":"https://bibbase.org/zotero/mh_lenguyen","bibdata":{"bibtype":"techreport","type":"Technical Report","title":"Semi-Supervised Learning Literature Survey","url":"https://minds.wisconsin.edu/handle/1793/60444","abstract":"We review some of the literature on semi-supervised learning in this paper. Traditional classifiers need labeled data (feature / label pairs) to train. Labeled instances however are often difficult, expensive, or time consuming to obtain, as they require the efforts of experienced human annotators. Meanwhile unlabeled data may be relatively easy to collect, but there has been few ways to use them. Semi-supervised learning addresses this problem by using large amount of unlabeled data, together with the labeled data, to build better classifiers. Because semi-supervised learning requires less human effort and gives higher accuracy, it is of great interest both in theory and in practice.","language":"en","urldate":"2022-03-20","institution":"University of Wisconsin-Madison Department of Computer Sciences","author":[{"propositions":[],"lastnames":["Zhu"],"firstnames":["Xiaojin","(Jerry)"],"suffixes":[]}],"year":"2005","note":"Accepted: 2012-03-15T17:19:12Z","bibtex":"@techreport{zhu_semi-supervised_2005,\n\ttype = {Technical {Report}},\n\ttitle = {Semi-{Supervised} {Learning} {Literature} {Survey}},\n\turl = {https://minds.wisconsin.edu/handle/1793/60444},\n\tabstract = {We review some of the literature on semi-supervised learning in this paper. Traditional classifiers need labeled data (feature / label pairs) to train. Labeled instances however are often difficult, expensive, or time consuming to obtain, as they require the efforts of experienced human annotators. Meanwhile unlabeled data may be relatively easy to collect, but there has been few ways to use them. Semi-supervised learning addresses this problem by using large amount of unlabeled data, together with the labeled data, to build better classifiers. Because semi-supervised learning requires less human effort and gives higher \naccuracy, it is of great interest both in theory and in practice.},\n\tlanguage = {en},\n\turldate = {2022-03-20},\n\tinstitution = {University of Wisconsin-Madison Department of Computer Sciences},\n\tauthor = {Zhu, Xiaojin (Jerry)},\n\tyear = {2005},\n\tnote = {Accepted: 2012-03-15T17:19:12Z},\n}\n\n","author_short":["Zhu, X. (."],"key":"zhu_semi-supervised_2005","id":"zhu_semi-supervised_2005","bibbaseid":"zhu-semisupervisedlearningliteraturesurvey-2005","role":"author","urls":{"Paper":"https://minds.wisconsin.edu/handle/1793/60444"},"metadata":{"authorlinks":{}}},"search_terms":["semi","supervised","learning","literature","survey","zhu"],"keywords":[],"authorIDs":[],"dataSources":["iCsmKnycRmHPxmhBd","iwKepCrWBps7ojhDx"]}