Challenges in Benchmarking Stream Learning Algorithms with Real-world Data
Streaming data are increasingly present in real-world applications such as sensor measurements, satellite data feed, stock market, and financial data. The main characteristics of these applications are the online arrival of data observations at high speed and the susceptibility to changes in the dat...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2020-06 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Souza, Vinicius M A dos Reis, Denis M Maletzke, Andre G Gustavo E A P A Batista |
description | Streaming data are increasingly present in real-world applications such as sensor measurements, satellite data feed, stock market, and financial data. The main characteristics of these applications are the online arrival of data observations at high speed and the susceptibility to changes in the data distributions due to the dynamic nature of real environments. The data stream mining community still faces some primary challenges and difficulties related to the comparison and evaluation of new proposals, mainly due to the lack of publicly available non-stationary real-world datasets. The comparison of stream algorithms proposed in the literature is not an easy task, as authors do not always follow the same recommendations, experimental evaluation procedures, datasets, and assumptions. In this paper, we mitigate problems related to the choice of datasets in the experimental evaluation of stream classifiers and drift detectors. To that end, we propose a new public data repository for benchmarking stream algorithms with real-world data. This repository contains the most popular datasets from literature and new datasets related to a highly relevant public health problem that involves the recognition of disease vector insects using optical sensors. The main advantage of these new datasets is the prior knowledge of their characteristics and patterns of changes to evaluate new adaptive algorithm proposals adequately. We also present an in-depth discussion about the characteristics, reasons, and issues that lead to different types of changes in data distribution, as well as a critical review of common problems concerning the current benchmark datasets available in the literature. |
doi_str_mv | 10.48550/arxiv.2005.00113 |
format | Article |
fullrecord | <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2005_00113</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2397932261</sourcerecordid><originalsourceid>FETCH-LOGICAL-a521-4db18cdeb0f94b4e38be4d9eed7f59381fbc5271cfebdd2f34ca320df20369cb3</originalsourceid><addsrcrecordid>eNotj11LwzAYhYMgOOZ-gFcGvO5M3jdZ28tZvxkIuvuSNG_Wzq6daef039ttXj0cOBzOw9iVFFOVaC1uTfipvqcghJ4KISWesREgyihRABds0nVrIQTMYtAaR-w1K01dU7OijlcNv6OmKDcmfFbNin_0gcyGL8iE5pDn9aoNVV9uOr4fwN_J1NG-DbXj96Y3l-zcm7qjyT_HbPn4sMyeo8Xb00s2X0RGg4yUszIpHFnhU2UVYWJJuZTIxV6nmEhvCw2xLDxZ58CjKgyCcB4EztLC4phdn2aPovk2VMPf3_wgnB-Fh8bNqbEN7deOuj5ft7vQDJ9ywDROEWAm8Q87g1lk</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2397932261</pqid></control><display><type>article</type><title>Challenges in Benchmarking Stream Learning Algorithms with Real-world Data</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Souza, Vinicius M A ; dos Reis, Denis M ; Maletzke, Andre G ; Gustavo E A P A Batista</creator><creatorcontrib>Souza, Vinicius M A ; dos Reis, Denis M ; Maletzke, Andre G ; Gustavo E A P A Batista</creatorcontrib><description>Streaming data are increasingly present in real-world applications such as sensor measurements, satellite data feed, stock market, and financial data. The main characteristics of these applications are the online arrival of data observations at high speed and the susceptibility to changes in the data distributions due to the dynamic nature of real environments. The data stream mining community still faces some primary challenges and difficulties related to the comparison and evaluation of new proposals, mainly due to the lack of publicly available non-stationary real-world datasets. The comparison of stream algorithms proposed in the literature is not an easy task, as authors do not always follow the same recommendations, experimental evaluation procedures, datasets, and assumptions. In this paper, we mitigate problems related to the choice of datasets in the experimental evaluation of stream classifiers and drift detectors. To that end, we propose a new public data repository for benchmarking stream algorithms with real-world data. This repository contains the most popular datasets from literature and new datasets related to a highly relevant public health problem that involves the recognition of disease vector insects using optical sensors. The main advantage of these new datasets is the prior knowledge of their characteristics and patterns of changes to evaluate new adaptive algorithm proposals adequately. We also present an in-depth discussion about the characteristics, reasons, and issues that lead to different types of changes in data distribution, as well as a critical review of common problems concerning the current benchmark datasets available in the literature.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2005.00113</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Adaptive algorithms ; Algorithms ; Benchmarks ; Computer Science - Learning ; Data transmission ; Datasets ; Evaluation ; Insects ; Machine learning ; Optical measuring instruments ; Proposals ; Public health ; Repositories ; Statistics - Machine Learning</subject><ispartof>arXiv.org, 2020-06</ispartof><rights>2020. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,784,885,27925</link.rule.ids><backlink>$$Uhttps://doi.org/10.48550/arXiv.2005.00113$$DView paper in arXiv$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.1007/s10618-020-00698-5$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink></links><search><creatorcontrib>Souza, Vinicius M A</creatorcontrib><creatorcontrib>dos Reis, Denis M</creatorcontrib><creatorcontrib>Maletzke, Andre G</creatorcontrib><creatorcontrib>Gustavo E A P A Batista</creatorcontrib><title>Challenges in Benchmarking Stream Learning Algorithms with Real-world Data</title><title>arXiv.org</title><description>Streaming data are increasingly present in real-world applications such as sensor measurements, satellite data feed, stock market, and financial data. The main characteristics of these applications are the online arrival of data observations at high speed and the susceptibility to changes in the data distributions due to the dynamic nature of real environments. The data stream mining community still faces some primary challenges and difficulties related to the comparison and evaluation of new proposals, mainly due to the lack of publicly available non-stationary real-world datasets. The comparison of stream algorithms proposed in the literature is not an easy task, as authors do not always follow the same recommendations, experimental evaluation procedures, datasets, and assumptions. In this paper, we mitigate problems related to the choice of datasets in the experimental evaluation of stream classifiers and drift detectors. To that end, we propose a new public data repository for benchmarking stream algorithms with real-world data. This repository contains the most popular datasets from literature and new datasets related to a highly relevant public health problem that involves the recognition of disease vector insects using optical sensors. The main advantage of these new datasets is the prior knowledge of their characteristics and patterns of changes to evaluate new adaptive algorithm proposals adequately. We also present an in-depth discussion about the characteristics, reasons, and issues that lead to different types of changes in data distribution, as well as a critical review of common problems concerning the current benchmark datasets available in the literature.</description><subject>Adaptive algorithms</subject><subject>Algorithms</subject><subject>Benchmarks</subject><subject>Computer Science - Learning</subject><subject>Data transmission</subject><subject>Datasets</subject><subject>Evaluation</subject><subject>Insects</subject><subject>Machine learning</subject><subject>Optical measuring instruments</subject><subject>Proposals</subject><subject>Public health</subject><subject>Repositories</subject><subject>Statistics - Machine Learning</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNotj11LwzAYhYMgOOZ-gFcGvO5M3jdZ28tZvxkIuvuSNG_Wzq6daef039ttXj0cOBzOw9iVFFOVaC1uTfipvqcghJ4KISWesREgyihRABds0nVrIQTMYtAaR-w1K01dU7OijlcNv6OmKDcmfFbNin_0gcyGL8iE5pDn9aoNVV9uOr4fwN_J1NG-DbXj96Y3l-zcm7qjyT_HbPn4sMyeo8Xb00s2X0RGg4yUszIpHFnhU2UVYWJJuZTIxV6nmEhvCw2xLDxZ58CjKgyCcB4EztLC4phdn2aPovk2VMPf3_wgnB-Fh8bNqbEN7deOuj5ft7vQDJ9ywDROEWAm8Q87g1lk</recordid><startdate>20200630</startdate><enddate>20200630</enddate><creator>Souza, Vinicius M A</creator><creator>dos Reis, Denis M</creator><creator>Maletzke, Andre G</creator><creator>Gustavo E A P A Batista</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20200630</creationdate><title>Challenges in Benchmarking Stream Learning Algorithms with Real-world Data</title><author>Souza, Vinicius M A ; dos Reis, Denis M ; Maletzke, Andre G ; Gustavo E A P A Batista</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a521-4db18cdeb0f94b4e38be4d9eed7f59381fbc5271cfebdd2f34ca320df20369cb3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Adaptive algorithms</topic><topic>Algorithms</topic><topic>Benchmarks</topic><topic>Computer Science - Learning</topic><topic>Data transmission</topic><topic>Datasets</topic><topic>Evaluation</topic><topic>Insects</topic><topic>Machine learning</topic><topic>Optical measuring instruments</topic><topic>Proposals</topic><topic>Public health</topic><topic>Repositories</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Souza, Vinicius M A</creatorcontrib><creatorcontrib>dos Reis, Denis M</creatorcontrib><creatorcontrib>Maletzke, Andre G</creatorcontrib><creatorcontrib>Gustavo E A P A Batista</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Souza, Vinicius M A</au><au>dos Reis, Denis M</au><au>Maletzke, Andre G</au><au>Gustavo E A P A Batista</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Challenges in Benchmarking Stream Learning Algorithms with Real-world Data</atitle><jtitle>arXiv.org</jtitle><date>2020-06-30</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>Streaming data are increasingly present in real-world applications such as sensor measurements, satellite data feed, stock market, and financial data. The main characteristics of these applications are the online arrival of data observations at high speed and the susceptibility to changes in the data distributions due to the dynamic nature of real environments. The data stream mining community still faces some primary challenges and difficulties related to the comparison and evaluation of new proposals, mainly due to the lack of publicly available non-stationary real-world datasets. The comparison of stream algorithms proposed in the literature is not an easy task, as authors do not always follow the same recommendations, experimental evaluation procedures, datasets, and assumptions. In this paper, we mitigate problems related to the choice of datasets in the experimental evaluation of stream classifiers and drift detectors. To that end, we propose a new public data repository for benchmarking stream algorithms with real-world data. This repository contains the most popular datasets from literature and new datasets related to a highly relevant public health problem that involves the recognition of disease vector insects using optical sensors. The main advantage of these new datasets is the prior knowledge of their characteristics and patterns of changes to evaluate new adaptive algorithm proposals adequately. We also present an in-depth discussion about the characteristics, reasons, and issues that lead to different types of changes in data distribution, as well as a critical review of common problems concerning the current benchmark datasets available in the literature.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2005.00113</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | EISSN: 2331-8422 |
ispartof | arXiv.org, 2020-06 |
issn | 2331-8422 |
language | eng |
recordid | cdi_arxiv_primary_2005_00113 |
source | arXiv.org; Free E- Journals |
subjects | Adaptive algorithms Algorithms Benchmarks Computer Science - Learning Data transmission Datasets Evaluation Insects Machine learning Optical measuring instruments Proposals Public health Repositories Statistics - Machine Learning |
title | Challenges in Benchmarking Stream Learning Algorithms with Real-world Data |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-26T06%3A16%3A21IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Challenges%20in%20Benchmarking%20Stream%20Learning%20Algorithms%20with%20Real-world%20Data&rft.jtitle=arXiv.org&rft.au=Souza,%20Vinicius%20M%20A&rft.date=2020-06-30&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2005.00113&rft_dat=%3Cproquest_arxiv%3E2397932261%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2397932261&rft_id=info:pmid/&rfr_iscdi=true |