Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigation
We investigate the potential of a new citizen science paradigm that facilitates collaborative learning between humans and artificial intelligence (AI). Recognising the potential of AI to support and empower rather than replace human participation, we explore the integration of image recognition as a...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ubiquity Press
2024-12-01
|
| Series: | Citizen Science: Theory and Practice |
| Subjects: | |
| Online Access: | https://account.theoryandpractice.citizenscienceassociation.org/index.php/up-j-cstp/article/view/735 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841554979401236480 |
|---|---|
| author | Nirwan Sharma Laura Colucci-Gray Poppy Lakeman-Fraser Annie Robinson Julie Newman René Van der Wal Stefan Rueger Advaith Siddharthan |
| author_facet | Nirwan Sharma Laura Colucci-Gray Poppy Lakeman-Fraser Annie Robinson Julie Newman René Van der Wal Stefan Rueger Advaith Siddharthan |
| author_sort | Nirwan Sharma |
| collection | DOAJ |
| description | We investigate the potential of a new citizen science paradigm that facilitates collaborative learning between humans and artificial intelligence (AI). Recognising the potential of AI to support and empower rather than replace human participation, we explore the integration of image recognition as a ‘dialogic AI partner’ in citizen science (CS) projects, interacting with participants in real time. We study this in the context of a biodiversity monitoring project that relies on volunteers to identify biological species from images taken in the wild. Guided by the idea of Bakhtin’s dialogism and Bayesian inference principles, we developed a web interface that integrated an image recognition model, fine-tuned for classifying 22 UK bumblebee species, into an interactive interface based on visual feature keys to enable real-time dialogue between humans and AI. We report a significant improvement in identification accuracy for both humans and AI when they engage in such dialogue and retain the ability to reach independent conclusions rather than achieve consensus. Given the inherent need for convergence in decision-making within scientific processes such as species identification tasks, we augmented the dialogic process with a Bayesian model that unifies potentially divergent human and AI perspectives post collaboration to achieve a more accurate consensus decision than that achieved by either AI or citizens. Our work provides new understandings around the design of a dialogic space for CS practice that effectively builds on the complementary strengths of human and AI visual recognition approaches. |
| format | Article |
| id | doaj-art-fb4d466a3b7d48968dd3619e7f83220c |
| institution | Kabale University |
| issn | 2057-4991 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Ubiquity Press |
| record_format | Article |
| series | Citizen Science: Theory and Practice |
| spelling | doaj-art-fb4d466a3b7d48968dd3619e7f83220c2025-01-08T07:54:40ZengUbiquity PressCitizen Science: Theory and Practice2057-49912024-12-0191353510.5334/cstp.735717Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigationNirwan Sharma0https://orcid.org/0000-0002-6576-3848Laura Colucci-Gray1https://orcid.org/0000-0003-0390-7364Poppy Lakeman-Fraser2https://orcid.org/0000-0002-1221-0094Annie Robinson3Julie Newman4https://orcid.org/0009-0008-4111-7500René Van der Wal5https://orcid.org/0000-0002-9175-0266Stefan Rueger6https://orcid.org/0000-0002-6013-9018Advaith Siddharthan7https://orcid.org/0000-0003-0796-8826The Open UniversityUniversity of EdinburghImperial College LondonUniversity of AberdeenSt Alban’s C of E Primary SchoolSwedish University of Agricultural Sciences (SLU)The Open UniversityThe Open UniversityWe investigate the potential of a new citizen science paradigm that facilitates collaborative learning between humans and artificial intelligence (AI). Recognising the potential of AI to support and empower rather than replace human participation, we explore the integration of image recognition as a ‘dialogic AI partner’ in citizen science (CS) projects, interacting with participants in real time. We study this in the context of a biodiversity monitoring project that relies on volunteers to identify biological species from images taken in the wild. Guided by the idea of Bakhtin’s dialogism and Bayesian inference principles, we developed a web interface that integrated an image recognition model, fine-tuned for classifying 22 UK bumblebee species, into an interactive interface based on visual feature keys to enable real-time dialogue between humans and AI. We report a significant improvement in identification accuracy for both humans and AI when they engage in such dialogue and retain the ability to reach independent conclusions rather than achieve consensus. Given the inherent need for convergence in decision-making within scientific processes such as species identification tasks, we augmented the dialogic process with a Bayesian model that unifies potentially divergent human and AI perspectives post collaboration to achieve a more accurate consensus decision than that achieved by either AI or citizens. Our work provides new understandings around the design of a dialogic space for CS practice that effectively builds on the complementary strengths of human and AI visual recognition approaches.https://account.theoryandpractice.citizenscienceassociation.org/index.php/up-j-cstp/article/view/735biodiversity citizen sciencedialogismhuman-ai collaborationspecies identificationmachine learningbayesian reasoningimage recognition |
| spellingShingle | Nirwan Sharma Laura Colucci-Gray Poppy Lakeman-Fraser Annie Robinson Julie Newman René Van der Wal Stefan Rueger Advaith Siddharthan Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigation Citizen Science: Theory and Practice biodiversity citizen science dialogism human-ai collaboration species identification machine learning bayesian reasoning image recognition |
| title | Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigation |
| title_full | Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigation |
| title_fullStr | Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigation |
| title_full_unstemmed | Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigation |
| title_short | Image Recognition as a “Dialogic AI Partner” Within Biodiversity Citizen Science—an empirical investigation |
| title_sort | image recognition as a dialogic ai partner within biodiversity citizen science an empirical investigation |
| topic | biodiversity citizen science dialogism human-ai collaboration species identification machine learning bayesian reasoning image recognition |
| url | https://account.theoryandpractice.citizenscienceassociation.org/index.php/up-j-cstp/article/view/735 |
| work_keys_str_mv | AT nirwansharma imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation AT lauracoluccigray imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation AT poppylakemanfraser imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation AT annierobinson imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation AT julienewman imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation AT renevanderwal imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation AT stefanrueger imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation AT advaithsiddharthan imagerecognitionasadialogicaipartnerwithinbiodiversitycitizenscienceanempiricalinvestigation |