Filters








17,653 Hits in 13.2 sec

How Is a Data-Driven Approach Better than Random Choice in Label Space Division for Multi-Label Classification?

Piotr Szymański, Tomasz Kajdanowicz, Kristian Kersting
2016 Entropy  
We propose using five data-driven community detection approaches from social networks to partition the label space for the task of multi-label classification as an alternative to random partitioning into  ...  We compare educated approaches to label space divisions against random baselines on 12 benchmark data sets over five evaluation measures.  ...  In this paper, we propose a data-driven approach for label space partitioning in multi-label classification.  ... 
doi:10.3390/e18080282 fatcat:peteqv52grf4no6gq6eqwvqwhy

A scikit-based Python environment for performing multi-label classification [article]

Piotr Szymański, Tomasz Kajdanowicz
2018 arXiv   pre-print
It provides native Python implementations of popular multi-label classification methods alongside a novel framework for label space partitioning and division.  ...  scikit-multilearn is a Python library for performing multi-label classification. The library is compatible with the scikit/scipy ecosystem and uses sparse matrices for all internal operations.  ...  , Felipe Almeida for bugfixes and testing, and Fernando Benites for providing the implementation of ML-ARAM (Brucker et al., 2011)  ... 
arXiv:1702.01460v5 fatcat:lujpcxp7avaglchqxvcus25x3u

scikit-multilearn: A scikit-based Python environment for performing multi-label classification

Piotr Szymański, Tomasz Kajdanowicz
2020 Zenodo  
It provides native Python implementations of popular multi-label classification methods alongside a novel framework for label space partitioning and division.  ...  scikit-multilearn is a Python library for performing multi-label classification. The library is compatible with the scikit/scipy ecosystem and uses sparse matrices for all internal operations.  ...  , Felipe Almeida for bugfixes and testing, and Fernando Benites for pro-  ... 
doi:10.5281/zenodo.3670934 fatcat:5pfwx74vajfbjnqrl2hwdobzpq

Is a Data-Driven Approach still Better than Random Choice with Naive Bayes classifiers? [article]

Piotr Szymański, Tomasz Kajdanowicz
2017 arXiv   pre-print
We study the performance of data-driven, a priori and random approaches to label space partitioning for multi-label classification with a Gaussian Naive Bayes classifier.  ...  In case of F1 scores and Subset Accuracy - data driven approaches were more likely to perform better than random approaches than otherwise in the worst case.  ...  .: How is a data-driven approach better than random choice in label space division for multi-label classification? Entropy ( ), ( ), http://dx.doi.org/ . /e .  ... 
arXiv:1702.04013v1 fatcat:uabogeg74jbijhzl3g2lw3r3ey

Comprehensive Comparative Study of Multi-Label Classification Methods [article]

Jasmin Bogatinovski, Ljupčo Todorovski, Sašo Džeroski, Dragi Kocev
2021 arXiv   pre-print
Multi-label classification (MLC) has recently received increasing interest from the machine learning community.  ...  Several studies provide reviews of methods and datasets for MLC and a few provide empirical comparisons of MLC methods. However, they are limited in the number of methods and datasets considered.  ...  Chi-dep's has a statistical driven built-in mechanism for resolving the dependencies between labels.  ... 
arXiv:2102.07113v2 fatcat:jtjefamw35fetjtnatmjvjl544

A Quest for Structure: Jointly Learning the Graph Structure and Semi-Supervised Classification [article]

Xuan Wu, Lingxiao Zhao, Leman Akoglu
2019 arXiv   pre-print
In essence, (1) allows us to search around a (random) initial hyperparameter configuration for a better one with lower validation loss.  ...  How should one construct a graph from the input point-cloud data for graph-based SSL?  ...  ACKNOWLEDGMENTS This research is sponsored by NSF CAREER 1452425 and IIS 1408287. Any conclusions in this material are of the authors and do not necessarily reflect the views of the funding parties.  ... 
arXiv:1909.12385v1 fatcat:hhzam7rxabeynji4fx6skagcfe

Multi-label ECG Signal Classification Based on Ensemble Classifier

Zhanquan Sun, Chaoli Wang, Yangyang Zhao, Chao Yan
2020 IEEE Access  
To resolve the multi-label ECG signal classification problems, we propose a novel ensemble multi-label classification model in this paper.  ...  It provides a feasible analysis method for multi-label ECG signal automatic classification. INDEX TERMS Electrocardiogram, multi-label classification, ensemble classification, mutual information.  ...  Kersting, ''How is a data-driven approach better than random choice in label space division for multi-label classification?'' Entropy, vol. 18, no. 8, p. 282, Jul. 2016.  ... 
doi:10.1109/access.2020.3004908 fatcat:bjl5ugj7gbc3lhvs7tr5umlrn4

Multi-modal Semantic Place Classification

A. Pronobis, O. Martínez Mozos, B. Caputo, P. Jensfelt
2009 The international journal of robotics research  
Our multi-modal place classification approach can be used to obtain a real-time semantic space labeling system which integrates information over time and space.  ...  The ability to represent knowledge about space and its position therein is crucial for a mobile robot.  ...  Acknowledgments Special thanks go to Sagar Behere for his great help with running the integrated system on the robotic platform, data acquisition and videotaping.  ... 
doi:10.1177/0278364909356483 fatcat:ic26hk5ylvcdddyyxqbzdlxhmy

Data-driven label space division usingcommunity detection methods

Piotr Szymański
2017 Figshare  
Presentation of progress in label space division for multi-label classification at Josef Stefan Insitute.  ...  fewer than 2 |L| combinations for Label Powerset there is a structure in how label co-occur the structure in a real world data set is not random (try running entropy checks) k-means is not an effective  ...  RAkELd than a priori methods in the worst case Yes Yes Yes Yes Yes RH4: Data-driven approach is more likely to per- form better than RAkELd in the worst case, than otherwise Yes Yes Yes  ... 
doi:10.6084/m9.figshare.4659010.v1 fatcat:c2pe5goclfeslnpbxlltsdrx6m

Boosting for tumor classification with gene expression data

M. Dettling, P. Buhlmann
2003 Bioinformatics  
In particular, we present a feature preselection method, a more robust boosting procedure and a new approach for multicategorical problems.  ...  Accurate supervised classification of tissue samples in such high-dimensional problems is difficult but often crucial for successful diagnosis and treatment.  ...  ACKNOWLEDGEMENTS Many thanks are due to Rainer Spang, Mike West and Joe Nevins for providing the estrogen and nodal datasets, and to Jane Fridlyand for providing the preprocessed NCI data.  ... 
doi:10.1093/bioinformatics/btf867 pmid:12801866 fatcat:i7dge2a3arbr7e42h5gwohitsm

Feature-Based Classification of Networks [article]

Ian Barnett, Nishant Malik, Marieke L. Kuijjer, Peter J. Mucha, Jukka-Pekka Onnela
2016 arXiv   pre-print
Here we describe and demonstrate a new, hybrid approach that combines manual selection of features of potential interest with existing automated classification methods.  ...  of special utility in the presence of feature collinearity, we find that we achieve higher accuracy, in shorter computation time, with greater interpretability of the network classification results.  ...  Acknowledgements We thank Kenth Eng-Monsen at Telenor Research for making the CDR-data available for this research.  ... 
arXiv:1610.05868v1 fatcat:5sqeo7be6bg53p2mxo46pmtt7e

Classification of drivers manoeuvre for road intersection crossing with synthethic and real data

Mathieu Barbier, Christian Laugier, Olivier Simonin, Javier Ibanez-Guzman
2017 2017 IEEE Intelligent Vehicles Symposium (IV)  
To address this problem, a hybrid data set was constructed. It is built in a simulated environment and completed with real data after has driven multiple times across the intersection.  ...  We show how our approach performs compared to other classifiers and space discretization. In addition, we demonstrate the impact and the usefulness of the mixture between simulated and real data.  ...  ACKNOWLEDGMENT The authors would like to thanks Jérôme Lussereau, Jean-Alix David for their support for the experimental part. This work is partially supported by the european project. ENABLE-S3  ... 
doi:10.1109/ivs.2017.7995724 dblp:conf/ivs/BarbierL0G17 fatcat:g2b74mkrjjdsnc2xbmi5segmkq

A Network Perspective on Stratification of Multi-Label Data [article]

Piotr Szymański, Tomasz Kajdanowicz
2017 arXiv   pre-print
We present a new approach to stratifying multi-label data for classification purposes based on the iterative stratification approach proposed by Sechidis et. al. in an ECML PKDD 2011 paper.  ...  In the recent years, we have witnessed the development of multi-label classification methods which utilize the structure of the label space in a divide and conquer approach to improve classification performance  ...  Introduction In our recent work [9] we proposed a data-driven community detection approach to partition the label space for the multi-label classification.  ... 
arXiv:1704.08756v1 fatcat:i44xx67tfjh4ferjwoxnfuclja

Partial Classifier Chains with Feature Selection by Exploiting Label Correlation in Multi-Label Classification

Zhenwu Wang, Tielin Wang, Benting Wan, Mengjie Han
2020 Entropy  
How to exploit the resulting label correlations is the key issue in MLC problems. The classifier chain (CC) is a well-known MLC approach that can learn complex coupling relationships between labels.  ...  Multi-label classification (MLC) is a supervised learning problem where an object is naturally associated with multiple concepts because it can be described from various dimensions.  ...  Using a data-driven approach, Wang et al.  ... 
doi:10.3390/e22101143 pmid:33286912 fatcat:6xwh6lngingmvluhgoirtirbne

Polyphonic Music Information Retrieval Based on Multi-label Cascade Classification System [chapter]

Wenxin Jiang, Amanda Cohen, Zbigniew W. Raś
2009 Studies in Computational Intelligence  
Polyphonic music information retrieval based on multi-label cascade classification system. (Under the direction of DR. ZBIGNIEW W.  ...  RAS) Recognition and separation of sounds played by various instruments is very useful in labeling audio files with semantic information.  ...  Figure 3 . 3 4 shows how the multi-label classification works.  ... 
doi:10.1007/978-3-642-04141-9_6 fatcat:7fek3uzphzcktewg75i7ider4m
« Previous Showing results 1 — 15 out of 17,653 results