{"@context":{"@vocab":"https://cir.nii.ac.jp/schema/1.0/","rdfs":"http://www.w3.org/2000/01/rdf-schema#","dc":"http://purl.org/dc/elements/1.1/","dcterms":"http://purl.org/dc/terms/","foaf":"http://xmlns.com/foaf/0.1/","prism":"http://prismstandard.org/namespaces/basic/2.0/","cinii":"http://ci.nii.ac.jp/ns/1.0/","datacite":"https://schema.datacite.org/meta/kernel-4/","ndl":"http://ndl.go.jp/dcndl/terms/","jpcoar":"https://github.com/JPCOAR/schema/blob/master/2.0/"},"@id":"https://cir.nii.ac.jp/crid/1360021392640267392.json","@type":"Article","productIdentifier":[{"identifier":{"@type":"DOI","@value":"10.1145/3474085.3475665"}},{"identifier":{"@type":"URI","@value":"https://dl.acm.org/doi/10.1145/3474085.3475665"}},{"identifier":{"@type":"URI","@value":"https://dl.acm.org/doi/pdf/10.1145/3474085.3475665"}},{"identifier":{"@type":"DOI","@value":"10.48550/arxiv.2107.05315"}}],"dc:title":[{"@value":"Contrastive Learning for Cold-Start Recommendation"}],"description":[{"notation":[{"@value":"Recommending cold-start items is a long-standing and fundamental challenge in recommender systems. Without any historical interaction on cold-start items, CF scheme fails to use collaborative signals to infer user preference on these items. To solve this problem, extensive studies have been conducted to incorporate side information into the CF scheme. Specifically, they employ modern neural network techniques (e.g., dropout, consistency constraint) to discover and exploit the coalition effect of content features and collaborative representations. However, we argue that these works less explore the mutual dependencies between content features and collaborative representations and lack sufficient theoretical supports, thus resulting in unsatisfactory performance. In this work, we reformulate the cold-start item representation learning from an information-theoretic standpoint. It aims to maximize the mutual dependencies between item content and collaborative signals. Specifically, the representation learning is theoretically lower-bounded by the integration of two terms: mutual information between collaborative embeddings of users and items, and mutual information between collaborative embeddings and feature representations of items. To model such a learning process, we devise a new objective function founded upon contrastive learning and develop a simple yet effective Contrastive Learning-based Cold-start Recommendation framework(CLCRec). In particular, CLCRec consists of three components: contrastive pair organization, contrastive embedding, and contrastive optimization modules. It allows us to preserve collaborative signals in the content representations for both warm and cold-start items. Through extensive experiments on four publicly accessible datasets, we observe that CLCRec achieves significant improvements over state-of-the-art approaches in both warm- and cold-start scenarios."}]},{"notation":[{"@value":"Accepted by ACM Multimedia 2021"}]}],"creator":[{"@id":"https://cir.nii.ac.jp/crid/1380021392640267392","@type":"Researcher","foaf:name":[{"@value":"Yinwei Wei"}],"jpcoar:affiliationName":[{"@value":"National University of Singapore, Singapore, Singapore"}]},{"@id":"https://cir.nii.ac.jp/crid/1380021392640267394","@type":"Researcher","foaf:name":[{"@value":"Xiang Wang"}],"jpcoar:affiliationName":[{"@value":"National University of Singapore, Singapore, Singapore"}]},{"@id":"https://cir.nii.ac.jp/crid/1380021392640267521","@type":"Researcher","foaf:name":[{"@value":"Qi Li"}],"jpcoar:affiliationName":[{"@value":"Shandong University, Qingdao, China"}]},{"@id":"https://cir.nii.ac.jp/crid/1380021392640267522","@type":"Researcher","foaf:name":[{"@value":"Liqiang Nie"}],"jpcoar:affiliationName":[{"@value":"Shandong University, Qingdao, China"}]},{"@id":"https://cir.nii.ac.jp/crid/1380021392640267393","@type":"Researcher","foaf:name":[{"@value":"Yan Li"}],"jpcoar:affiliationName":[{"@value":"Beijing Kuaishou Technology Co., Ltd., Beijing, China"}]},{"@id":"https://cir.nii.ac.jp/crid/1380021392640267523","@type":"Researcher","foaf:name":[{"@value":"Xuanping Li"}],"jpcoar:affiliationName":[{"@value":"Beijing Kuaishou Technology Co., Ltd., Beijing, China"}]},{"@id":"https://cir.nii.ac.jp/crid/1380021392640267520","@type":"Researcher","foaf:name":[{"@value":"Tat-Seng Chua"}],"jpcoar:affiliationName":[{"@value":"National University of Singapore, Singapore, Singapore"}]}],"publication":{"prism:publicationName":[{"@value":"Proceedings of the 29th ACM International Conference on Multimedia"}],"dc:publisher":[{"@value":"ACM"}],"prism:publicationDate":"2021-10-17","prism:startingPage":"5382","prism:endingPage":"5390"},"reviewed":"false","dcterms:accessRights":"http://purl.org/coar/access_right/c_abf2","dc:rights":["https://www.acm.org/publications/policies/copyright_policy#Background"],"url":[{"@id":"https://dl.acm.org/doi/10.1145/3474085.3475665"},{"@id":"https://dl.acm.org/doi/pdf/10.1145/3474085.3475665"}],"createdAt":"2021-10-18","modifiedAt":"2025-06-17","foaf:topic":[{"@id":"https://cir.nii.ac.jp/all?q=FOS:%20Computer%20and%20information%20sciences","dc:title":"FOS: Computer and information sciences"},{"@id":"https://cir.nii.ac.jp/all?q=Computer%20Science%20-%20Multimedia","dc:title":"Computer Science - Multimedia"},{"@id":"https://cir.nii.ac.jp/all?q=Information%20Retrieval%20(cs.IR)","dc:title":"Information Retrieval (cs.IR)"},{"@id":"https://cir.nii.ac.jp/all?q=Computer%20Science%20-%20Information%20Retrieval","dc:title":"Computer Science - Information Retrieval"},{"@id":"https://cir.nii.ac.jp/all?q=Multimedia%20(cs.MM)","dc:title":"Multimedia (cs.MM)"}],"relatedProduct":[{"@id":"https://cir.nii.ac.jp/crid/1360025429443104000","@type":"Article","resourceType":"学術雑誌論文(journal article)","relationType":["isReferencedBy"],"jpcoar:relatedTitle":[{"@value":"EEDN: Enhanced Encoder-Decoder Network with Local and Global Context Learning for POI Recommendation"}]},{"@id":"https://cir.nii.ac.jp/crid/1360025431117497344","@type":"Article","resourceType":"学術雑誌論文(journal article)","relationType":["isReferencedBy"],"jpcoar:relatedTitle":[{"@value":"Trend Analysis with Interpretability and Cold-Start Problems for Recommender Systems"}]},{"@id":"https://cir.nii.ac.jp/crid/1360302865725124096","@type":"Article","resourceType":"学術雑誌論文(journal article)","relationType":["isReferencedBy"],"jpcoar:relatedTitle":[{"@value":"Four-level knowledge graph contrastive learning structure for smart-phone application recommendation"}]}],"dataSourceIdentifier":[{"@type":"CROSSREF","@value":"10.1145/3474085.3475665"},{"@type":"OPENAIRE","@value":"doi_dedup___::19a4ba2a1cea03878c5ab97178403b89"},{"@type":"CROSSREF","@value":"10.1145/3539618.3591678_references_DOI_VxNzEeSHvNZvnYAOHRqKZwe5KRN"},{"@type":"CROSSREF","@value":"10.1007/s12626-024-00168-0_references_DOI_VxNzEeSHvNZvnYAOHRqKZwe5KRN"},{"@type":"CROSSREF","@value":"10.1109/iwcmc58020.2023.10182890_references_DOI_VxNzEeSHvNZvnYAOHRqKZwe5KRN"}]}