TY - GEN
T1 - Visual Representation for Capturing Creator Theme in Brand-Creator Marketplace
AU - Duanis, Sarel
AU - Gaiger, Keren
AU - Cohen, Ravid
AU - Zychlinski, Shaked
AU - Greenstein-Messica, Asnat
N1 - Publisher Copyright:
© 2023 Owner/Author.
PY - 2023/9/14
Y1 - 2023/9/14
N2 - Providing cold start recommendations in a brand-creator marketplace is challenging as brands' preferences extend beyond the mere objects depicted in the creator's content and encompass the creator's individual theme consistently thatresonates across images shared on her social media profile. Furthermore, brands often use textual keywords to describe their campaign's aesthetic appeal, with which creators must align. To address these challenges, we propose two methods: SAME (Same Account Media Embedding), a novel creator representation employing a Siamese network to capture the unique creator theme and OAAR (Object-Agnostic Adjective Representation), enabling filtering creators based on textual adjectives that relate to aesthetic qualities through zero-shot learning. These two methods utilize CLIP, a state-of-the-art language-image model, and improve it in addressing the aforementioned challenges.
AB - Providing cold start recommendations in a brand-creator marketplace is challenging as brands' preferences extend beyond the mere objects depicted in the creator's content and encompass the creator's individual theme consistently thatresonates across images shared on her social media profile. Furthermore, brands often use textual keywords to describe their campaign's aesthetic appeal, with which creators must align. To address these challenges, we propose two methods: SAME (Same Account Media Embedding), a novel creator representation employing a Siamese network to capture the unique creator theme and OAAR (Object-Agnostic Adjective Representation), enabling filtering creators based on textual adjectives that relate to aesthetic qualities through zero-shot learning. These two methods utilize CLIP, a state-of-the-art language-image model, and improve it in addressing the aforementioned challenges.
KW - cold-start recommendations
KW - two-sided marketplace
UR - http://www.scopus.com/inward/record.url?scp=85174522535&partnerID=8YFLogxK
U2 - 10.1145/3604915.3610237
DO - 10.1145/3604915.3610237
M3 - Conference contribution
AN - SCOPUS:85174522535
T3 - Proceedings of the 17th ACM Conference on Recommender Systems, RecSys 2023
SP - 1027
EP - 1030
BT - Proceedings of the 17th ACM Conference on Recommender Systems, RecSys 2023
PB - Association for Computing Machinery, Inc
T2 - 17th ACM Conference on Recommender Systems, RecSys 2023
Y2 - 18 September 2023 through 22 September 2023
ER -