{"corpus_id":206784101,"paper_sha":"17296302ffb402ba3ec4b49a883224c7ceaa1ae7","doi":"10.1109/ICDM.2007.90","arxiv_id":null,"pmid":null,"pmcid":null,"mag_id":2172249709,"dblp_id":"conf/icdm/BellK07","acl_id":null,"title":"Scalable Collaborative Filtering with Jointly Derived Neighborhood Interpolation Weights","year":2007,"publication_date":"2007-10-28","venue":"Industrial Conference on Data Mining","journal":{"name":"Seventh IEEE International Conference on Data Mining (ICDM 2007)","pages":"43-52","volume":null},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle","Conference"],"pubmed_pub_types":null,"s2_fields_of_study":["Computer Science"],"reference_count":16,"citation_count":586,"influential_citation_count":52,"is_open_access":false,"arxiv_categories":null,"arxiv_license":null,"arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":null,"s2_open_access_landing_url":null,"s2_open_access_license":null,"s2_open_access_status":null,"pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"Recommender systems based on collaborative filtering predict user preferences for products or services by learning past user-item relationships. A predominant approach to collaborative filtering is neighborhood based (\"k-nearest neighbors\"), where a user-item preference rating is interpolated from ratings of similar items and/or users. We enhance the neighborhood-based approach leading to substantial improvement of prediction accuracy, without a meaningful increase in running time. First, we remove certain so-called \"global effects\" from the data to make the ratings more comparable, thereby improving interpolation accuracy. Second, we show how to simultaneously derive interpolation weights for all nearest neighbors, unlike previous approaches where each weight is computed separately. By globally solving a suitable optimization problem, this simultaneous interpolation accounts for the many interactions between neighbors leading to improved accuracy. Our method is very fast in practice, generating a prediction in about 0.2 milliseconds. Importantly, it does not require training many parameters or a lengthy preprocessing, making it very practical for large scale applications. Finally, we show how to apply these methods to the perceivably much slower user-oriented approach. To this end, we suggest a novel scheme for low dimensional embedding of the users. We evaluate these methods on the netflix dataset, where they deliver significantly better results than the commercial netflix cinematch recommender system.","claims":[{"public_id":"cl_fbc90c33c8ed180a2b7ae4d429d9c00f","status":"active","text":"A novel low-dimensional embedding scheme for users enables application of the neighborhood interpolation approach to the user-oriented method.","confidence":0.9,"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_fbc90c33c8ed180a2b7ae4d429d9c00f"},{"public_id":"cl_96e2acdc15dde55a546c65e7b3f8732d","status":"active","text":"Removing global effects from rating data improves interpolation accuracy in neighborhood-based collaborative filtering.","confidence":0.9,"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_96e2acdc15dde55a546c65e7b3f8732d"},{"public_id":"cl_b89d8894191df779a338e8e909014c29","status":"active","text":"Simultaneously deriving interpolation weights for all nearest neighbors via a global optimization problem improves prediction accuracy over separate weight computation.","confidence":0.95,"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_b89d8894191df779a338e8e909014c29"},{"public_id":"cl_9619e705e8173ce7a92f854fbfd55a2b","status":"active","text":"The methods deliver significantly better results than the commercial Netflix Cinematch recommender system on the Netflix dataset.","confidence":0.95,"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_9619e705e8173ce7a92f854fbfd55a2b"},{"public_id":"cl_c29614bccf2292066e652bd5b01b24e9","status":"active","text":"The proposed method generates a prediction in about 0.2 milliseconds and does not require lengthy preprocessing or training many parameters.","confidence":0.95,"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_c29614bccf2292066e652bd5b01b24e9"}],"concepts":[{"public_id":"co_06314eeefc41204dd4774b5560822ec7","status":"active","name":"simultaneous interpolation weights","description":"Interpolation weights for all nearest neighbors derived together by solving a global optimization problem, accounting for interactions among neighbors.","types":["method"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_06314eeefc41204dd4774b5560822ec7"},{"public_id":"co_1625075e709b424e2bd7b006ae91c7d8","status":"active","name":"nearest neighbors","description":"Items or users most similar to a target item or user, used for rating interpolation.","types":["data entity"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_1625075e709b424e2bd7b006ae91c7d8"},{"public_id":"co_1857b2d2b1d4571baecff501395f132a","status":"active","name":"interpolation accuracy","description":"The quality of predicted ratings derived from interpolating ratings of similar items or users.","types":["metric"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_1857b2d2b1d4571baecff501395f132a"},{"public_id":"co_311493035599548cbb92f3430b2cfb84","status":"active","name":"global optimization","description":"An optimization problem solved over all nearest neighbors simultaneously to derive interpolation weights.","types":["method"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_311493035599548cbb92f3430b2cfb84"},{"public_id":"co_37f01ada1350ac2a871990e70863239c","status":"active","name":"user-oriented approach","description":"A collaborative filtering variant that predicts ratings based on similarities among users rather than items.","types":["method"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_37f01ada1350ac2a871990e70863239c"},{"public_id":"co_4985af999c85171e2d6458ab74f2b8c5","status":"active","name":"Netflix Cinematch recommender system","description":"The commercial recommender system from Netflix used as a baseline for comparison.","types":["baseline"],"aliases":["Cinematch"],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_4985af999c85171e2d6458ab74f2b8c5"},{"public_id":"co_8e70947f4531d455deb72a97e5d5dbf5","status":"active","name":"neighborhood-based collaborative filtering","description":"A collaborative filtering approach that predicts preferences by interpolating ratings from similar items or users (k-nearest neighbors).","types":["method"],"aliases":["k-nearest neighbors"],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_8e70947f4531d455deb72a97e5d5dbf5"},{"public_id":"co_9522b290729f18e6107f3740f7d1ddf6","status":"active","name":"Netflix dataset","description":"The evaluation dataset used to compare the proposed methods against the Netflix Cinematch recommender.","types":["dataset"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_9522b290729f18e6107f3740f7d1ddf6"},{"public_id":"co_95bab07d2c0e8214cf3595ed9d03df45","status":"active","name":"practicality for large scale applications","description":"The method's suitability for large-scale use due to fast prediction and minimal preprocessing or parameter training.","types":["characteristic"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_95bab07d2c0e8214cf3595ed9d03df45"},{"public_id":"co_9bc649f1b9f2d15f654b844d16d12e66","status":"active","name":"prediction speed","description":"The runtime of generating a single prediction, reported as about 0.2 milliseconds.","types":["performance metric"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_9bc649f1b9f2d15f654b844d16d12e66"},{"public_id":"co_a79852e9a3b4be2b6853ddc64a944f56","status":"active","name":"global effects","description":"Systematic biases or patterns in rating data that are removed to make ratings more comparable.","types":["data preprocessing"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_a79852e9a3b4be2b6853ddc64a944f56"},{"public_id":"co_fa7768d09ecd3b5bdbc318c13e5ef319","status":"active","name":"low-dimensional embedding of users","description":"A novel scheme that maps users into a low-dimensional space to enable efficient user-oriented collaborative filtering.","types":["method"],"aliases":[],"contributors":[{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["extraction"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["review"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_fa7768d09ecd3b5bdbc318c13e5ef319"}],"external_ids":{"DOI":"10.1109/ICDM.2007.90","ArXiv":null,"PubMed":null,"PubMedCentral":null,"MAG":2172249709,"DBLP":"conf/icdm/BellK07","ACL":null},"open_access":{"is_open_access":false,"pdf_url":null,"landing_url":"https://sah.borca.ai/papers/206784101","source":null,"pdf_url_source":null,"license":null,"reason":"pdf_url_not_indexed"},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":false,"full_text_source":null,"count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":632838,"paper_uid":"eedd3285-ff9c-4d98-a6a7-512b066e1dc5","canonical_identity":{"paper_id":632838,"paper_uid":"eedd3285-ff9c-4d98-a6a7-512b066e1dc5","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/206784101"}