Beyond PageRank: Machine Learning for Static Ranking

Beyond PageRank: Machine Learning for Static Ranking. Richardson, M., Prakash, A., & Brill, E. In Proceedings of the 15th International Conference on World Wide Web, of WWW '06, pages 707–715, New York, NY, USA, 2006. ACM.

Paper doi abstract bibtex

Since the publication of Brin and Page's paper on PageRank, many in the Web community have depended on PageRank for the static (query-independent) ordering of Web pages. We show that we can significantly outperform PageRank using features that are independent of the link structure of the Web. We gain a further boost in accuracy by using data on the frequency at which users visit Web pages. We use RankNet, a ranking machine learning algorithm, to combine these and other static features based on anchor text and domain characteristics. The resulting model achieves a static ranking pairwise accuracy of 67.3% (vs. 56.7% for PageRank or 50% for random).

@inproceedings{richardson_beyond_2006,
	address = {New York, NY, USA},
	series = {{WWW} '06},
	title = {Beyond {PageRank}: {Machine} {Learning} for {Static} {Ranking}},
	isbn = {978-1-59593-323-2},
	shorttitle = {Beyond {PageRank}},
	url = {http://doi.acm.org/10.1145/1135777.1135881},
	doi = {10.1145/1135777.1135881},
	abstract = {Since the publication of Brin and Page's paper on PageRank, many in the Web community have depended on PageRank for the static (query-independent) ordering of Web pages. We show that we can significantly outperform PageRank using features that are independent of the link structure of the Web. We gain a further boost in accuracy by using data on the frequency at which users visit Web pages. We use RankNet, a ranking machine learning algorithm, to combine these and other static features based on anchor text and domain characteristics. The resulting model achieves a static ranking pairwise accuracy of 67.3\% (vs. 56.7\% for PageRank or 50\% for random).},
	urldate = {2017-04-28},
	booktitle = {Proceedings of the 15th {International} {Conference} on {World} {Wide} {Web}},
	publisher = {ACM},
	author = {Richardson, Matthew and Prakash, Amit and Brill, Eric},
	year = {2006},
	keywords = {PageRank, RankNet, relevance, search engines, static ranking},
	pages = {707--715}
}

Downloads: 0

{"_id":"BvbN2SEMCMLEbk9ff","bibbaseid":"richardson-prakash-brill-beyondpagerankmachinelearningforstaticranking-2006","authorIDs":[],"author_short":["Richardson, M.","Prakash, A.","Brill, E."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"New York, NY, USA","series":"WWW '06","title":"Beyond PageRank: Machine Learning for Static Ranking","isbn":"978-1-59593-323-2","shorttitle":"Beyond PageRank","url":"http://doi.acm.org/10.1145/1135777.1135881","doi":"10.1145/1135777.1135881","abstract":"Since the publication of Brin and Page's paper on PageRank, many in the Web community have depended on PageRank for the static (query-independent) ordering of Web pages. We show that we can significantly outperform PageRank using features that are independent of the link structure of the Web. We gain a further boost in accuracy by using data on the frequency at which users visit Web pages. We use RankNet, a ranking machine learning algorithm, to combine these and other static features based on anchor text and domain characteristics. The resulting model achieves a static ranking pairwise accuracy of 67.3% (vs. 56.7% for PageRank or 50% for random).","urldate":"2017-04-28","booktitle":"Proceedings of the 15th International Conference on World Wide Web","publisher":"ACM","author":[{"propositions":[],"lastnames":["Richardson"],"firstnames":["Matthew"],"suffixes":[]},{"propositions":[],"lastnames":["Prakash"],"firstnames":["Amit"],"suffixes":[]},{"propositions":[],"lastnames":["Brill"],"firstnames":["Eric"],"suffixes":[]}],"year":"2006","keywords":"PageRank, RankNet, relevance, search engines, static ranking","pages":"707–715","bibtex":"@inproceedings{richardson_beyond_2006,\n\taddress = {New York, NY, USA},\n\tseries = {{WWW} '06},\n\ttitle = {Beyond {PageRank}: {Machine} {Learning} for {Static} {Ranking}},\n\tisbn = {978-1-59593-323-2},\n\tshorttitle = {Beyond {PageRank}},\n\turl = {http://doi.acm.org/10.1145/1135777.1135881},\n\tdoi = {10.1145/1135777.1135881},\n\tabstract = {Since the publication of Brin and Page's paper on PageRank, many in the Web community have depended on PageRank for the static (query-independent) ordering of Web pages. We show that we can significantly outperform PageRank using features that are independent of the link structure of the Web. We gain a further boost in accuracy by using data on the frequency at which users visit Web pages. We use RankNet, a ranking machine learning algorithm, to combine these and other static features based on anchor text and domain characteristics. The resulting model achieves a static ranking pairwise accuracy of 67.3\\% (vs. 56.7\\% for PageRank or 50\\% for random).},\n\turldate = {2017-04-28},\n\tbooktitle = {Proceedings of the 15th {International} {Conference} on {World} {Wide} {Web}},\n\tpublisher = {ACM},\n\tauthor = {Richardson, Matthew and Prakash, Amit and Brill, Eric},\n\tyear = {2006},\n\tkeywords = {PageRank, RankNet, relevance, search engines, static ranking},\n\tpages = {707--715}\n}\n\n","author_short":["Richardson, M.","Prakash, A.","Brill, E."],"key":"richardson_beyond_2006","id":"richardson_beyond_2006","bibbaseid":"richardson-prakash-brill-beyondpagerankmachinelearningforstaticranking-2006","role":"author","urls":{"Paper":"http://doi.acm.org/10.1145/1135777.1135881"},"keyword":["PageRank","RankNet","relevance","search engines","static ranking"],"downloads":0},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/jsan","creationDate":"2020-05-26T09:25:38.360Z","downloads":0,"keywords":["pagerank","ranknet","relevance","search engines","static ranking"],"search_terms":["beyond","pagerank","machine","learning","static","ranking","richardson","prakash","brill"],"title":"Beyond PageRank: Machine Learning for Static Ranking","year":2006,"dataSources":["h8ZDyzMApwwGDDKcf"]}