MEDIC: a multi-task learning dataset for disaster image classification

<p>Recent research in disaster informatics demonstrates a practical and important use case of artificial intelligence to save human lives and suffering during natural disasters based on social media contents (text and images). While notable progress has been made using texts, research on explo...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Firoj Alam (14158866) (author)
مؤلفون آخرون: Tanvirul Alam (14150628) (author), Md. Arid Hasan (14150631) (author), Abul Hasnat (5561213) (author), Muhammad Imran (282621) (author), Ferda Ofli (8983517) (author)
منشور في: 2022
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513568007782400
author Firoj Alam (14158866)
author2 Tanvirul Alam (14150628)
Md. Arid Hasan (14150631)
Abul Hasnat (5561213)
Muhammad Imran (282621)
Ferda Ofli (8983517)
author2_role author
author
author
author
author
author_facet Firoj Alam (14158866)
Tanvirul Alam (14150628)
Md. Arid Hasan (14150631)
Abul Hasnat (5561213)
Muhammad Imran (282621)
Ferda Ofli (8983517)
author_role author
dc.creator.none.fl_str_mv Firoj Alam (14158866)
Tanvirul Alam (14150628)
Md. Arid Hasan (14150631)
Abul Hasnat (5561213)
Muhammad Imran (282621)
Ferda Ofli (8983517)
dc.date.none.fl_str_mv 2022-09-03T06:00:00Z
dc.identifier.none.fl_str_mv 10.1007/s00521-022-07717-0
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/MEDIC_a_multi-task_learning_dataset_for_disaster_image_classification/21597078
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Earth sciences
Physical geography and environmental geoscience
Information and computing sciences
Artificial intelligence
Computer vision and multimedia computation
Machine learning
Software engineering
Multi-task learning
Social media images
Image classification
Natural disasters
Crisis informatics
Deep learning
Dataset
dc.title.none.fl_str_mv MEDIC: a multi-task learning dataset for disaster image classification
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p>Recent research in disaster informatics demonstrates a practical and important use case of artificial intelligence to save human lives and suffering during natural disasters based on social media contents (text and images). While notable progress has been made using texts, research on exploiting the images remains relatively under-explored. To advance image-based approaches, we propose MEDIC (https://crisisnlp.qcri.org/medic/index.html), which is the largest social media image classification dataset for humanitarian response consisting of 71,198 images to address four different tasks in a multi-task learning setup. This is the first dataset of its kind: social media images, disaster response, and multi-task learning research. An important property of this dataset is its high potential to facilitate research on multi-task learning, which recently receives much interest from the machine learning community and has shown remarkable results in terms of memory, inference speed, performance, and generalization capability. Therefore, the proposed dataset is an important resource for advancing image-based disaster management and multi-task machine learning research. We experiment with different deep learning architectures and report promising results, which are above the majority baselines for all tasks. Along with the dataset, we also release all relevant scripts (https://github.com/firojalam/medic).</p><h2>Other Information</h2> <p> Published in: Neural Computing and Applications<br> License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="http://dx.doi.org/10.1007/s00521-022-07717-0" target="_blank">http://dx.doi.org/10.1007/s00521-022-07717-0</a></p>
eu_rights_str_mv openAccess
id Manara2_19b7bb146134eec52aa1e0d69308f464
identifier_str_mv 10.1007/s00521-022-07717-0
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/21597078
publishDate 2022
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling MEDIC: a multi-task learning dataset for disaster image classificationFiroj Alam (14158866)Tanvirul Alam (14150628)Md. Arid Hasan (14150631)Abul Hasnat (5561213)Muhammad Imran (282621)Ferda Ofli (8983517)Earth sciencesPhysical geography and environmental geoscienceInformation and computing sciencesArtificial intelligenceComputer vision and multimedia computationMachine learningSoftware engineeringMulti-task learningSocial media imagesImage classificationNatural disastersCrisis informaticsDeep learningDataset<p>Recent research in disaster informatics demonstrates a practical and important use case of artificial intelligence to save human lives and suffering during natural disasters based on social media contents (text and images). While notable progress has been made using texts, research on exploiting the images remains relatively under-explored. To advance image-based approaches, we propose MEDIC (https://crisisnlp.qcri.org/medic/index.html), which is the largest social media image classification dataset for humanitarian response consisting of 71,198 images to address four different tasks in a multi-task learning setup. This is the first dataset of its kind: social media images, disaster response, and multi-task learning research. An important property of this dataset is its high potential to facilitate research on multi-task learning, which recently receives much interest from the machine learning community and has shown remarkable results in terms of memory, inference speed, performance, and generalization capability. Therefore, the proposed dataset is an important resource for advancing image-based disaster management and multi-task machine learning research. We experiment with different deep learning architectures and report promising results, which are above the majority baselines for all tasks. Along with the dataset, we also release all relevant scripts (https://github.com/firojalam/medic).</p><h2>Other Information</h2> <p> Published in: Neural Computing and Applications<br> License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="http://dx.doi.org/10.1007/s00521-022-07717-0" target="_blank">http://dx.doi.org/10.1007/s00521-022-07717-0</a></p>2022-09-03T06:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1007/s00521-022-07717-0https://figshare.com/articles/journal_contribution/MEDIC_a_multi-task_learning_dataset_for_disaster_image_classification/21597078CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/215970782022-09-03T06:00:00Z
spellingShingle MEDIC: a multi-task learning dataset for disaster image classification
Firoj Alam (14158866)
Earth sciences
Physical geography and environmental geoscience
Information and computing sciences
Artificial intelligence
Computer vision and multimedia computation
Machine learning
Software engineering
Multi-task learning
Social media images
Image classification
Natural disasters
Crisis informatics
Deep learning
Dataset
status_str publishedVersion
title MEDIC: a multi-task learning dataset for disaster image classification
title_full MEDIC: a multi-task learning dataset for disaster image classification
title_fullStr MEDIC: a multi-task learning dataset for disaster image classification
title_full_unstemmed MEDIC: a multi-task learning dataset for disaster image classification
title_short MEDIC: a multi-task learning dataset for disaster image classification
title_sort MEDIC: a multi-task learning dataset for disaster image classification
topic Earth sciences
Physical geography and environmental geoscience
Information and computing sciences
Artificial intelligence
Computer vision and multimedia computation
Machine learning
Software engineering
Multi-task learning
Social media images
Image classification
Natural disasters
Crisis informatics
Deep learning
Dataset