Scalable Similarity-based Neighborhood Methods with MapReduce

Scalable Similarity-based Neighborhood Methods with MapReduce. Schelter, S., Boden, C., & Markl, V. In RecSys '12, pages 163–170, New York, NY, USA, 2012. ACM. Journal Abbreviation: RecSys '12

Paper doi abstract bibtex 1 download

Similarity-based neighborhood methods, a simple and popular approach to collaborative filtering, infer their predictions by finding users with similar taste or items that have been similarly rated. If the number of users grows to millions, the standard approach of sequentially examining each item and looking at all interacting users does not scale. To solve this problem, we develop a MapReduce algorithm for the pairwise item comparison and top-N recommendation problem that scales linearly with respect to a growing number of users. This parallel algorithm is able to work on partitioned data and is general in that it supports a wide range of similarity measures. We evaluate our algorithm on a large dataset consisting of 700 million song ratings from Yahoo! Music.

@inproceedings{schelter_scalable_2012,
	address = {New York, NY, USA},
	title = {Scalable {Similarity}-based {Neighborhood} {Methods} with {MapReduce}},
	url = {http://doi.acm.org/10.1145/2365952.2365984},
	doi = {10.1145/2365952.2365984},
	abstract = {Similarity-based neighborhood methods, a simple and popular approach to
collaborative filtering, infer their predictions by finding users with
similar taste or items that have been similarly rated. If the number of
users grows to millions, the standard approach of sequentially examining
each item and looking at all interacting users does not scale. To solve
this problem, we develop a MapReduce algorithm for the pairwise item
comparison and top-N recommendation problem that scales linearly with
respect to a growing number of users. This parallel algorithm is able to
work on partitioned data and is general in that it supports a wide range
of similarity measures. We evaluate our algorithm on a large dataset
consisting of 700 million song ratings from Yahoo! Music.},
	urldate = {2015-09-23},
	booktitle = {{RecSys} '12},
	publisher = {ACM},
	author = {Schelter, Sebastian and Boden, Christoph and Markl, Volker},
	year = {2012},
	note = {Journal Abbreviation: RecSys '12},
	pages = {163--170},
}

Downloads: 1

{"_id":"xP4oDPQbZTTWzzvt6","bibbaseid":"schelter-boden-markl-scalablesimilaritybasedneighborhoodmethodswithmapreduce-2012","authorIDs":[],"author_short":["Schelter, S.","Boden, C.","Markl, V."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"New York, NY, USA","title":"Scalable Similarity-based Neighborhood Methods with MapReduce","url":"http://doi.acm.org/10.1145/2365952.2365984","doi":"10.1145/2365952.2365984","abstract":"Similarity-based neighborhood methods, a simple and popular approach to collaborative filtering, infer their predictions by finding users with similar taste or items that have been similarly rated. If the number of users grows to millions, the standard approach of sequentially examining each item and looking at all interacting users does not scale. To solve this problem, we develop a MapReduce algorithm for the pairwise item comparison and top-N recommendation problem that scales linearly with respect to a growing number of users. This parallel algorithm is able to work on partitioned data and is general in that it supports a wide range of similarity measures. We evaluate our algorithm on a large dataset consisting of 700 million song ratings from Yahoo! Music.","urldate":"2015-09-23","booktitle":"RecSys '12","publisher":"ACM","author":[{"propositions":[],"lastnames":["Schelter"],"firstnames":["Sebastian"],"suffixes":[]},{"propositions":[],"lastnames":["Boden"],"firstnames":["Christoph"],"suffixes":[]},{"propositions":[],"lastnames":["Markl"],"firstnames":["Volker"],"suffixes":[]}],"year":"2012","note":"Journal Abbreviation: RecSys '12","pages":"163–170","bibtex":"@inproceedings{schelter_scalable_2012,\n\taddress = {New York, NY, USA},\n\ttitle = {Scalable {Similarity}-based {Neighborhood} {Methods} with {MapReduce}},\n\turl = {http://doi.acm.org/10.1145/2365952.2365984},\n\tdoi = {10.1145/2365952.2365984},\n\tabstract = {Similarity-based neighborhood methods, a simple and popular approach to\ncollaborative filtering, infer their predictions by finding users with\nsimilar taste or items that have been similarly rated. If the number of\nusers grows to millions, the standard approach of sequentially examining\neach item and looking at all interacting users does not scale. To solve\nthis problem, we develop a MapReduce algorithm for the pairwise item\ncomparison and top-N recommendation problem that scales linearly with\nrespect to a growing number of users. This parallel algorithm is able to\nwork on partitioned data and is general in that it supports a wide range\nof similarity measures. We evaluate our algorithm on a large dataset\nconsisting of 700 million song ratings from Yahoo! Music.},\n\turldate = {2015-09-23},\n\tbooktitle = {{RecSys} '12},\n\tpublisher = {ACM},\n\tauthor = {Schelter, Sebastian and Boden, Christoph and Markl, Volker},\n\tyear = {2012},\n\tnote = {Journal Abbreviation: RecSys '12},\n\tpages = {163--170},\n}\n\n","author_short":["Schelter, S.","Boden, C.","Markl, V."],"key":"schelter_scalable_2012","id":"schelter_scalable_2012","bibbaseid":"schelter-boden-markl-scalablesimilaritybasedneighborhoodmethodswithmapreduce-2012","role":"author","urls":{"Paper":"http://doi.acm.org/10.1145/2365952.2365984"},"metadata":{"authorlinks":{}},"downloads":1},"bibtype":"inproceedings","biburl":"https://api.zotero.org/users/6655/collections/TJPPJ92X/items?key=VFvZhZXIoHNBbzoLZ1IM2zgf&format=bibtex&limit=100","creationDate":"2020-03-27T02:34:35.296Z","downloads":1,"keywords":[],"search_terms":["scalable","similarity","based","neighborhood","methods","mapreduce","schelter","boden","markl"],"title":"Scalable Similarity-based Neighborhood Methods with MapReduce","year":2012,"dataSources":["5Dp4QphkvpvNA33zi","jfoasiDDpStqkkoZB","BiuuFc45aHCgJqDLY"]}