Active Learning from Data Streams. Zhu, X., Zhang, P., Lin, X., & Shi, Y. In Seventh IEEE International Conference on Data Mining (ICDM 2007), pages 757–762, October, 2007. ISSN: 2374-8486doi abstract bibtex In this paper, we address a new research problem on active learning from data streams where data volumes grow continuously and labeling all data is considered expensive and impractical. The objective is to label a small portion of stream data from which a model is derived to predict newly arrived instances as accurate as possible. In order to tackle the challenges raised by data streams' dynamic nature, we propose a classifier ensembling based active learning framework which selectively labels instances from data streams to build an accurate classifier. A minimal variance principle is introduced to guide instance labeling from data streams. In addition, a weight updating rule is derived to ensure that our instance labeling process can adaptively adjust to dynamic drifting concepts in the data. Experimental results on synthetic and real-world data demonstrate the performances of the proposed efforts in comparison with other simple approaches.
@inproceedings{zhu_active_2007,
title = {Active {Learning} from {Data} {Streams}},
doi = {10.1109/ICDM.2007.101},
abstract = {In this paper, we address a new research problem on active learning from data streams where data volumes grow continuously and labeling all data is considered expensive and impractical. The objective is to label a small portion of stream data from which a model is derived to predict newly arrived instances as accurate as possible. In order to tackle the challenges raised by data streams' dynamic nature, we propose a classifier ensembling based active learning framework which selectively labels instances from data streams to build an accurate classifier. A minimal variance principle is introduced to guide instance labeling from data streams. In addition, a weight updating rule is derived to ensure that our instance labeling process can adaptively adjust to dynamic drifting concepts in the data. Experimental results on synthetic and real-world data demonstrate the performances of the proposed efforts in comparison with other simple approaches.},
booktitle = {Seventh {IEEE} {International} {Conference} on {Data} {Mining} ({ICDM} 2007)},
author = {Zhu, Xingquan and Zhang, Peng and Lin, Xiaodong and Shi, Yong},
month = oct,
year = {2007},
note = {ISSN: 2374-8486},
keywords = {Accuracy, Association rules, Computer science, Data engineering, Data mining, Decision making, Labeling, Predictive models, USA Councils, Uncertainty},
pages = {757--762},
}
Downloads: 0
{"_id":"K7rtvnnFhbh32qQGz","bibbaseid":"zhu-zhang-lin-shi-activelearningfromdatastreams-2007","author_short":["Zhu, X.","Zhang, P.","Lin, X.","Shi, Y."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Active Learning from Data Streams","doi":"10.1109/ICDM.2007.101","abstract":"In this paper, we address a new research problem on active learning from data streams where data volumes grow continuously and labeling all data is considered expensive and impractical. The objective is to label a small portion of stream data from which a model is derived to predict newly arrived instances as accurate as possible. In order to tackle the challenges raised by data streams' dynamic nature, we propose a classifier ensembling based active learning framework which selectively labels instances from data streams to build an accurate classifier. A minimal variance principle is introduced to guide instance labeling from data streams. In addition, a weight updating rule is derived to ensure that our instance labeling process can adaptively adjust to dynamic drifting concepts in the data. Experimental results on synthetic and real-world data demonstrate the performances of the proposed efforts in comparison with other simple approaches.","booktitle":"Seventh IEEE International Conference on Data Mining (ICDM 2007)","author":[{"propositions":[],"lastnames":["Zhu"],"firstnames":["Xingquan"],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["Peng"],"suffixes":[]},{"propositions":[],"lastnames":["Lin"],"firstnames":["Xiaodong"],"suffixes":[]},{"propositions":[],"lastnames":["Shi"],"firstnames":["Yong"],"suffixes":[]}],"month":"October","year":"2007","note":"ISSN: 2374-8486","keywords":"Accuracy, Association rules, Computer science, Data engineering, Data mining, Decision making, Labeling, Predictive models, USA Councils, Uncertainty","pages":"757–762","bibtex":"@inproceedings{zhu_active_2007,\n\ttitle = {Active {Learning} from {Data} {Streams}},\n\tdoi = {10.1109/ICDM.2007.101},\n\tabstract = {In this paper, we address a new research problem on active learning from data streams where data volumes grow continuously and labeling all data is considered expensive and impractical. The objective is to label a small portion of stream data from which a model is derived to predict newly arrived instances as accurate as possible. In order to tackle the challenges raised by data streams' dynamic nature, we propose a classifier ensembling based active learning framework which selectively labels instances from data streams to build an accurate classifier. A minimal variance principle is introduced to guide instance labeling from data streams. In addition, a weight updating rule is derived to ensure that our instance labeling process can adaptively adjust to dynamic drifting concepts in the data. Experimental results on synthetic and real-world data demonstrate the performances of the proposed efforts in comparison with other simple approaches.},\n\tbooktitle = {Seventh {IEEE} {International} {Conference} on {Data} {Mining} ({ICDM} 2007)},\n\tauthor = {Zhu, Xingquan and Zhang, Peng and Lin, Xiaodong and Shi, Yong},\n\tmonth = oct,\n\tyear = {2007},\n\tnote = {ISSN: 2374-8486},\n\tkeywords = {Accuracy, Association rules, Computer science, Data engineering, Data mining, Decision making, Labeling, Predictive models, USA Councils, Uncertainty},\n\tpages = {757--762},\n}\n\n\n\n","author_short":["Zhu, X.","Zhang, P.","Lin, X.","Shi, Y."],"key":"zhu_active_2007","id":"zhu_active_2007","bibbaseid":"zhu-zhang-lin-shi-activelearningfromdatastreams-2007","role":"author","urls":{},"keyword":["Accuracy","Association rules","Computer science","Data engineering","Data mining","Decision making","Labeling","Predictive models","USA Councils","Uncertainty"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/mh_lenguyen","dataSources":["iwKepCrWBps7ojhDx"],"keywords":["accuracy","association rules","computer science","data engineering","data mining","decision making","labeling","predictive models","usa councils","uncertainty"],"search_terms":["active","learning","data","streams","zhu","zhang","lin","shi"],"title":"Active Learning from Data Streams","year":2007}