How does serendipity affect diversity in recommender systems? A serendipity-oriented greedy algorithm

How does serendipity affect diversity in recommender systems? A serendipity-oriented greedy algorithm. Kotkov, D., Veijalainen, J., & Wang, S. Computing, 102(2):393–411, February, 2020.

Paper doi abstract bibtex 3 downloads

Most recommender systems suggest items that are popular among all users and similar to items a user usually consumes. As a result, the user receives recommendations that she/he is already familiar with or would find anyway, leading to low satisfaction. To overcome this problem, a recommender system should suggest novel, relevant and unexpected i.e., serendipitous items. In this paper, we propose a serendipity-oriented, reranking algorithm called a serendipity-oriented greedy (SOG) algorithm, which improves serendipity of recommendations through feature diversification and helps overcome the overspecialization problem. To evaluate our algorithm, we employed the only publicly available dataset containing user feedback regarding serendipity. We compared our SOG algorithm with topic diversification, popularity baseline, singular value decomposition, serendipitous personalized ranking and Zheng’s algorithms relying on the above dataset. SOG outperforms other algorithms in terms of serendipity and diversity. It also outperforms serendipity-oriented algorithms in terms of accuracy, but underperforms accuracy-oriented algorithms in terms of accuracy. We found that the increase of diversity can hurt accuracy and harm or improve serendipity depending on the size of diversity increase.

@article{kotkov_how_2020,
	title = {How does serendipity affect diversity in recommender systems? {A} serendipity-oriented greedy algorithm},
	volume = {102},
	issn = {0144-3097},
	url = {http://link.springer.com/10.1007/s00607-018-0687-5},
	doi = {10.1007/s00607-018-0687-5},
	abstract = {Most recommender systems suggest items that are popular among all users
and similar to items a user usually consumes. As a result, the user
receives recommendations that she/he is already familiar with or would
find anyway, leading to low satisfaction. To overcome this problem, a
recommender system should suggest novel, relevant and unexpected i.e.,
serendipitous items. In this paper, we propose a serendipity-oriented,
reranking algorithm called a serendipity-oriented greedy (SOG) algorithm,
which improves serendipity of recommendations through feature
diversification and helps overcome the overspecialization problem. To
evaluate our algorithm, we employed the only publicly available dataset
containing user feedback regarding serendipity. We compared our SOG
algorithm with topic diversification, popularity baseline, singular value
decomposition, serendipitous personalized ranking and Zheng’s algorithms
relying on the above dataset. SOG outperforms other algorithms in terms of
serendipity and diversity. It also outperforms serendipity-oriented
algorithms in terms of accuracy, but underperforms accuracy-oriented
algorithms in terms of accuracy. We found that the increase of diversity
can hurt accuracy and harm or improve serendipity depending on the size of
diversity increase.},
	number = {2},
	journal = {Computing},
	author = {Kotkov, Denis and Veijalainen, Jari and Wang, Shuaiqiang},
	month = feb,
	year = {2020},
	pages = {393--411},
}

Downloads: 3

{"_id":"3S9FQ4kYrWbLJCkWb","bibbaseid":"kotkov-veijalainen-wang-howdoesserendipityaffectdiversityinrecommendersystemsaserendipityorientedgreedyalgorithm-2020","authorIDs":[],"author_short":["Kotkov, D.","Veijalainen, J.","Wang, S."],"bibdata":{"bibtype":"article","type":"article","title":"How does serendipity affect diversity in recommender systems? A serendipity-oriented greedy algorithm","volume":"102","issn":"0144-3097","url":"http://link.springer.com/10.1007/s00607-018-0687-5","doi":"10.1007/s00607-018-0687-5","abstract":"Most recommender systems suggest items that are popular among all users and similar to items a user usually consumes. As a result, the user receives recommendations that she/he is already familiar with or would find anyway, leading to low satisfaction. To overcome this problem, a recommender system should suggest novel, relevant and unexpected i.e., serendipitous items. In this paper, we propose a serendipity-oriented, reranking algorithm called a serendipity-oriented greedy (SOG) algorithm, which improves serendipity of recommendations through feature diversification and helps overcome the overspecialization problem. To evaluate our algorithm, we employed the only publicly available dataset containing user feedback regarding serendipity. We compared our SOG algorithm with topic diversification, popularity baseline, singular value decomposition, serendipitous personalized ranking and Zheng’s algorithms relying on the above dataset. SOG outperforms other algorithms in terms of serendipity and diversity. It also outperforms serendipity-oriented algorithms in terms of accuracy, but underperforms accuracy-oriented algorithms in terms of accuracy. We found that the increase of diversity can hurt accuracy and harm or improve serendipity depending on the size of diversity increase.","number":"2","journal":"Computing","author":[{"propositions":[],"lastnames":["Kotkov"],"firstnames":["Denis"],"suffixes":[]},{"propositions":[],"lastnames":["Veijalainen"],"firstnames":["Jari"],"suffixes":[]},{"propositions":[],"lastnames":["Wang"],"firstnames":["Shuaiqiang"],"suffixes":[]}],"month":"February","year":"2020","pages":"393–411","bibtex":"@article{kotkov_how_2020,\n\ttitle = {How does serendipity affect diversity in recommender systems? {A} serendipity-oriented greedy algorithm},\n\tvolume = {102},\n\tissn = {0144-3097},\n\turl = {http://link.springer.com/10.1007/s00607-018-0687-5},\n\tdoi = {10.1007/s00607-018-0687-5},\n\tabstract = {Most recommender systems suggest items that are popular among all users\nand similar to items a user usually consumes. As a result, the user\nreceives recommendations that she/he is already familiar with or would\nfind anyway, leading to low satisfaction. To overcome this problem, a\nrecommender system should suggest novel, relevant and unexpected i.e.,\nserendipitous items. In this paper, we propose a serendipity-oriented,\nreranking algorithm called a serendipity-oriented greedy (SOG) algorithm,\nwhich improves serendipity of recommendations through feature\ndiversification and helps overcome the overspecialization problem. To\nevaluate our algorithm, we employed the only publicly available dataset\ncontaining user feedback regarding serendipity. We compared our SOG\nalgorithm with topic diversification, popularity baseline, singular value\ndecomposition, serendipitous personalized ranking and Zheng’s algorithms\nrelying on the above dataset. SOG outperforms other algorithms in terms of\nserendipity and diversity. It also outperforms serendipity-oriented\nalgorithms in terms of accuracy, but underperforms accuracy-oriented\nalgorithms in terms of accuracy. We found that the increase of diversity\ncan hurt accuracy and harm or improve serendipity depending on the size of\ndiversity increase.},\n\tnumber = {2},\n\tjournal = {Computing},\n\tauthor = {Kotkov, Denis and Veijalainen, Jari and Wang, Shuaiqiang},\n\tmonth = feb,\n\tyear = {2020},\n\tpages = {393--411},\n}\n\n","author_short":["Kotkov, D.","Veijalainen, J.","Wang, S."],"key":"kotkov_how_2020","id":"kotkov_how_2020","bibbaseid":"kotkov-veijalainen-wang-howdoesserendipityaffectdiversityinrecommendersystemsaserendipityorientedgreedyalgorithm-2020","role":"author","urls":{"Paper":"http://link.springer.com/10.1007/s00607-018-0687-5"},"metadata":{"authorlinks":{}},"downloads":3},"bibtype":"article","biburl":"https://api.zotero.org/users/6655/collections/TJPPJ92X/items?key=VFvZhZXIoHNBbzoLZ1IM2zgf&format=bibtex&limit=100","creationDate":"2020-10-23T07:37:04.823Z","downloads":3,"keywords":[],"search_terms":["serendipity","affect","diversity","recommender","systems","serendipity","oriented","greedy","algorithm","kotkov","veijalainen","wang"],"title":"How does serendipity affect diversity in recommender systems? A serendipity-oriented greedy algorithm","year":2020,"dataSources":["5Dp4QphkvpvNA33zi","jfoasiDDpStqkkoZB","BiuuFc45aHCgJqDLY"]}