Novelty Detection for Multi-Label Stream Classification. Costa Júnior, J. D., Faria, E. R., Silva, J. A., Gama, J., & Cerri, R. In 2019 8th Brazilian Conference on Intelligent Systems (BRACIS), pages 144–149, October, 2019. ISSN: 2643-6264doi abstract bibtex In Multi-Label Stream Classification (MLSC) examples arriving in a stream can be simultaneously classified into multiple classes. This is a very challenging task, especially considering that new classes can emerge during the stream (Concept Evolution), and known classes can change over time (Concept Drift). In real situations, these characteristics come together with a scenario with Infinitely Delayed Labels, where we can never access the true class labels of the examples to update classifiers. In order to overcome these issues, this paper proposes a new method called MultI-label learNing Algorithm for Data Streams with Binary Relevance transformation (MINAS-BR). Our proposal uses a new Novelty Detection (ND) procedure to detect concept evolution and concept drift, being updated in an unsupervised fashion. We also propose a new methodology to evaluate MLSC methods in scenarios with Infinitely Delayed Labels. Experiments over synthetic data sets attested the potential of MINAS-BR, which was able to adapt to different concept drift and concept evolution scenarios, obtaining superior or competitive performances in comparison to literature baselines.
@inproceedings{costa_junior_novelty_2019,
title = {Novelty {Detection} for {Multi}-{Label} {Stream} {Classification}},
doi = {10.1109/BRACIS.2019.00034},
abstract = {In Multi-Label Stream Classification (MLSC) examples arriving in a stream can be simultaneously classified into multiple classes. This is a very challenging task, especially considering that new classes can emerge during the stream (Concept Evolution), and known classes can change over time (Concept Drift). In real situations, these characteristics come together with a scenario with Infinitely Delayed Labels, where we can never access the true class labels of the examples to update classifiers. In order to overcome these issues, this paper proposes a new method called MultI-label learNing Algorithm for Data Streams with Binary Relevance transformation (MINAS-BR). Our proposal uses a new Novelty Detection (ND) procedure to detect concept evolution and concept drift, being updated in an unsupervised fashion. We also propose a new methodology to evaluate MLSC methods in scenarios with Infinitely Delayed Labels. Experiments over synthetic data sets attested the potential of MINAS-BR, which was able to adapt to different concept drift and concept evolution scenarios, obtaining superior or competitive performances in comparison to literature baselines.},
booktitle = {2019 8th {Brazilian} {Conference} on {Intelligent} {Systems} ({BRACIS})},
author = {Costa Júnior, Joel D. and Faria, Elaine R. and Silva, Jonathan A. and Gama, João and Cerri, Ricardo},
month = oct,
year = {2019},
note = {ISSN: 2643-6264},
keywords = {Adaptation models, Computational complexity, Computational modeling, Concept Evolution, Data models, Infinitely Delayed Labels, Multi-label Stream Classification, Novelty Detection, Task analysis, Training, Training data},
pages = {144--149},
}
Downloads: 0
{"_id":"wRkXR8bDxN9oriudr","bibbaseid":"costajnior-faria-silva-gama-cerri-noveltydetectionformultilabelstreamclassification-2019","author_short":["Costa Júnior, J. D.","Faria, E. R.","Silva, J. A.","Gama, J.","Cerri, R."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Novelty Detection for Multi-Label Stream Classification","doi":"10.1109/BRACIS.2019.00034","abstract":"In Multi-Label Stream Classification (MLSC) examples arriving in a stream can be simultaneously classified into multiple classes. This is a very challenging task, especially considering that new classes can emerge during the stream (Concept Evolution), and known classes can change over time (Concept Drift). In real situations, these characteristics come together with a scenario with Infinitely Delayed Labels, where we can never access the true class labels of the examples to update classifiers. In order to overcome these issues, this paper proposes a new method called MultI-label learNing Algorithm for Data Streams with Binary Relevance transformation (MINAS-BR). Our proposal uses a new Novelty Detection (ND) procedure to detect concept evolution and concept drift, being updated in an unsupervised fashion. We also propose a new methodology to evaluate MLSC methods in scenarios with Infinitely Delayed Labels. Experiments over synthetic data sets attested the potential of MINAS-BR, which was able to adapt to different concept drift and concept evolution scenarios, obtaining superior or competitive performances in comparison to literature baselines.","booktitle":"2019 8th Brazilian Conference on Intelligent Systems (BRACIS)","author":[{"propositions":[],"lastnames":["Costa","Júnior"],"firstnames":["Joel","D."],"suffixes":[]},{"propositions":[],"lastnames":["Faria"],"firstnames":["Elaine","R."],"suffixes":[]},{"propositions":[],"lastnames":["Silva"],"firstnames":["Jonathan","A."],"suffixes":[]},{"propositions":[],"lastnames":["Gama"],"firstnames":["João"],"suffixes":[]},{"propositions":[],"lastnames":["Cerri"],"firstnames":["Ricardo"],"suffixes":[]}],"month":"October","year":"2019","note":"ISSN: 2643-6264","keywords":"Adaptation models, Computational complexity, Computational modeling, Concept Evolution, Data models, Infinitely Delayed Labels, Multi-label Stream Classification, Novelty Detection, Task analysis, Training, Training data","pages":"144–149","bibtex":"@inproceedings{costa_junior_novelty_2019,\n\ttitle = {Novelty {Detection} for {Multi}-{Label} {Stream} {Classification}},\n\tdoi = {10.1109/BRACIS.2019.00034},\n\tabstract = {In Multi-Label Stream Classification (MLSC) examples arriving in a stream can be simultaneously classified into multiple classes. This is a very challenging task, especially considering that new classes can emerge during the stream (Concept Evolution), and known classes can change over time (Concept Drift). In real situations, these characteristics come together with a scenario with Infinitely Delayed Labels, where we can never access the true class labels of the examples to update classifiers. In order to overcome these issues, this paper proposes a new method called MultI-label learNing Algorithm for Data Streams with Binary Relevance transformation (MINAS-BR). Our proposal uses a new Novelty Detection (ND) procedure to detect concept evolution and concept drift, being updated in an unsupervised fashion. We also propose a new methodology to evaluate MLSC methods in scenarios with Infinitely Delayed Labels. Experiments over synthetic data sets attested the potential of MINAS-BR, which was able to adapt to different concept drift and concept evolution scenarios, obtaining superior or competitive performances in comparison to literature baselines.},\n\tbooktitle = {2019 8th {Brazilian} {Conference} on {Intelligent} {Systems} ({BRACIS})},\n\tauthor = {Costa Júnior, Joel D. and Faria, Elaine R. and Silva, Jonathan A. and Gama, João and Cerri, Ricardo},\n\tmonth = oct,\n\tyear = {2019},\n\tnote = {ISSN: 2643-6264},\n\tkeywords = {Adaptation models, Computational complexity, Computational modeling, Concept Evolution, Data models, Infinitely Delayed Labels, Multi-label Stream Classification, Novelty Detection, Task analysis, Training, Training data},\n\tpages = {144--149},\n}\n\n","author_short":["Costa Júnior, J. D.","Faria, E. R.","Silva, J. A.","Gama, J.","Cerri, R."],"key":"costa_junior_novelty_2019","id":"costa_junior_novelty_2019","bibbaseid":"costajnior-faria-silva-gama-cerri-noveltydetectionformultilabelstreamclassification-2019","role":"author","urls":{},"keyword":["Adaptation models","Computational complexity","Computational modeling","Concept Evolution","Data models","Infinitely Delayed Labels","Multi-label Stream Classification","Novelty Detection","Task analysis","Training","Training data"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/mh_lenguyen","dataSources":["iwKepCrWBps7ojhDx"],"keywords":["adaptation models","computational complexity","computational modeling","concept evolution","data models","infinitely delayed labels","multi-label stream classification","novelty detection","task analysis","training","training data"],"search_terms":["novelty","detection","multi","label","stream","classification","costa júnior","faria","silva","gama","cerri"],"title":"Novelty Detection for Multi-Label Stream Classification","year":2019}