Serendipity in recommender systems

Serendipity in recommender systems. Kotkov, D. University of Jyväskylä, 2018.

The number of goods and services (such as accommodation or music streaming) offered by e-commerce websites does not allow users to examine all the available options in a reasonable amount of time. Recommender systems are auxiliary systems designed to help users ﬁnd interesting goods or services (items) on a website when the number of available items is overwhelming. Traditionally, recommender systems have been optimized for accuracy, which indicates how often a user consumed the items recommended by system. To increase accuracy, recommender systems often suggest items that are popular and suitably similar to items these users have consumed in the past. As a result, users often lose interest in using these systems, as they either know about the recommended items already or can easily ﬁnd these items themselves. One way to increase user satisfaction and user retention is to suggest serendipitous items. These items are items that users would not ﬁnd themselves or even look for, but would enjoy consuming. Serendipity in recommender systems has not been thoroughly investigated. There is not even a consensus on the concept’s deﬁnition. In this dissertation, serendipitous items are deﬁned as relevant, novel and unexpected to a user. In this dissertation, we (a) review different deﬁnitions of the concept and evaluate them in a user study, (b) assess the proportion of serendipitous items in a typical recommender system, (c) review ways to measure and improve serendipity, (d) investigate serendipity in cross-domain recommender systems (systems that take advantage of multiple domains, such as movies, songs and books) and (e) discuss challenges and future directions concerning this topic. We applied a Design Science methodology as the framework for this study and developed four artifacts: (1) a collection of eight variations of serendipity deﬁnition, (2) a measure of the serendipity of suggested items, (3) an algorithm that generates serendipitous suggestions, (4) a dataset of user feedback regarding serendipitous movies in the recommender system MovieLens. These artifacts are evaluated using suitable methods and communicated through publications.

@book{kotkov_serendipity_2018,
	title = {Serendipity in recommender systems},
	isbn = {978-951-39-7438-1},
	url = {https://jyx.jyu.fi/handle/123456789/58207},
	abstract = {The number of goods and services (such as accommodation or music
streaming) offered by e-commerce websites does not allow users to examine
all the available options in a reasonable amount of time. Recommender
systems are auxiliary systems designed to help users ﬁnd interesting goods
or services (items) on a website when the number of available items is
overwhelming. Traditionally, recommender systems have been optimized for
accuracy, which indicates how often a user consumed the items recommended
by system. To increase accuracy, recommender systems often suggest items
that are popular and suitably similar to items these users have consumed
in the past. As a result, users often lose interest in using these
systems, as they either know about the recommended items already or can
easily ﬁnd these items themselves. One way to increase user satisfaction
and user retention is to suggest serendipitous items. These items are
items that users would not ﬁnd themselves or even look for, but would
enjoy consuming. Serendipity in recommender systems has not been
thoroughly investigated. There is not even a consensus on the concept’s
deﬁnition. In this dissertation, serendipitous items are deﬁned as
relevant, novel and unexpected to a user. In this dissertation, we (a)
review different deﬁnitions of the concept and evaluate them in a user
study, (b) assess the proportion of serendipitous items in a typical
recommender system, (c) review ways to measure and improve serendipity,
(d) investigate serendipity in cross-domain recommender systems (systems
that take advantage of multiple domains, such as movies, songs and books)
and (e) discuss challenges and future directions concerning this topic. We
applied a Design Science methodology as the framework for this study and
developed four artifacts: (1) a collection of eight variations of
serendipity deﬁnition, (2) a measure of the serendipity of suggested
items, (3) an algorithm that generates serendipitous suggestions, (4) a
dataset of user feedback regarding serendipitous movies in the recommender
system MovieLens. These artifacts are evaluated using suitable methods and
communicated through publications.},
	urldate = {2018-07-06},
	publisher = {University of Jyväskylä},
	author = {Kotkov, Denis},
	year = {2018},
}

Downloads: 0

{"_id":"zZ9GTG9vMvFPJhnxn","bibbaseid":"kotkov-serendipityinrecommendersystems-2018","authorIDs":[],"author_short":["Kotkov, D."],"bibdata":{"bibtype":"book","type":"book","title":"Serendipity in recommender systems","isbn":"978-951-39-7438-1","url":"https://jyx.jyu.fi/handle/123456789/58207","abstract":"The number of goods and services (such as accommodation or music streaming) offered by e-commerce websites does not allow users to examine all the available options in a reasonable amount of time. Recommender systems are auxiliary systems designed to help users ﬁnd interesting goods or services (items) on a website when the number of available items is overwhelming. Traditionally, recommender systems have been optimized for accuracy, which indicates how often a user consumed the items recommended by system. To increase accuracy, recommender systems often suggest items that are popular and suitably similar to items these users have consumed in the past. As a result, users often lose interest in using these systems, as they either know about the recommended items already or can easily ﬁnd these items themselves. One way to increase user satisfaction and user retention is to suggest serendipitous items. These items are items that users would not ﬁnd themselves or even look for, but would enjoy consuming. Serendipity in recommender systems has not been thoroughly investigated. There is not even a consensus on the concept’s deﬁnition. In this dissertation, serendipitous items are deﬁned as relevant, novel and unexpected to a user. In this dissertation, we (a) review different deﬁnitions of the concept and evaluate them in a user study, (b) assess the proportion of serendipitous items in a typical recommender system, (c) review ways to measure and improve serendipity, (d) investigate serendipity in cross-domain recommender systems (systems that take advantage of multiple domains, such as movies, songs and books) and (e) discuss challenges and future directions concerning this topic. We applied a Design Science methodology as the framework for this study and developed four artifacts: (1) a collection of eight variations of serendipity deﬁnition, (2) a measure of the serendipity of suggested items, (3) an algorithm that generates serendipitous suggestions, (4) a dataset of user feedback regarding serendipitous movies in the recommender system MovieLens. These artifacts are evaluated using suitable methods and communicated through publications.","urldate":"2018-07-06","publisher":"University of Jyväskylä","author":[{"propositions":[],"lastnames":["Kotkov"],"firstnames":["Denis"],"suffixes":[]}],"year":"2018","bibtex":"@book{kotkov_serendipity_2018,\n\ttitle = {Serendipity in recommender systems},\n\tisbn = {978-951-39-7438-1},\n\turl = {https://jyx.jyu.fi/handle/123456789/58207},\n\tabstract = {The number of goods and services (such as accommodation or music\nstreaming) offered by e-commerce websites does not allow users to examine\nall the available options in a reasonable amount of time. Recommender\nsystems are auxiliary systems designed to help users ﬁnd interesting goods\nor services (items) on a website when the number of available items is\noverwhelming. Traditionally, recommender systems have been optimized for\naccuracy, which indicates how often a user consumed the items recommended\nby system. To increase accuracy, recommender systems often suggest items\nthat are popular and suitably similar to items these users have consumed\nin the past. As a result, users often lose interest in using these\nsystems, as they either know about the recommended items already or can\neasily ﬁnd these items themselves. One way to increase user satisfaction\nand user retention is to suggest serendipitous items. These items are\nitems that users would not ﬁnd themselves or even look for, but would\nenjoy consuming. Serendipity in recommender systems has not been\nthoroughly investigated. There is not even a consensus on the concept’s\ndeﬁnition. In this dissertation, serendipitous items are deﬁned as\nrelevant, novel and unexpected to a user. In this dissertation, we (a)\nreview different deﬁnitions of the concept and evaluate them in a user\nstudy, (b) assess the proportion of serendipitous items in a typical\nrecommender system, (c) review ways to measure and improve serendipity,\n(d) investigate serendipity in cross-domain recommender systems (systems\nthat take advantage of multiple domains, such as movies, songs and books)\nand (e) discuss challenges and future directions concerning this topic. We\napplied a Design Science methodology as the framework for this study and\ndeveloped four artifacts: (1) a collection of eight variations of\nserendipity deﬁnition, (2) a measure of the serendipity of suggested\nitems, (3) an algorithm that generates serendipitous suggestions, (4) a\ndataset of user feedback regarding serendipitous movies in the recommender\nsystem MovieLens. These artifacts are evaluated using suitable methods and\ncommunicated through publications.},\n\turldate = {2018-07-06},\n\tpublisher = {University of Jyväskylä},\n\tauthor = {Kotkov, Denis},\n\tyear = {2018},\n}\n\n","author_short":["Kotkov, D."],"key":"kotkov_serendipity_2018","id":"kotkov_serendipity_2018","bibbaseid":"kotkov-serendipityinrecommendersystems-2018","role":"author","urls":{"Paper":"https://jyx.jyu.fi/handle/123456789/58207"},"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"book","biburl":"https://api.zotero.org/users/6655/collections/TJPPJ92X/items?key=VFvZhZXIoHNBbzoLZ1IM2zgf&format=bibtex&limit=100","creationDate":"2020-03-27T02:34:35.399Z","downloads":0,"keywords":[],"search_terms":["serendipity","recommender","systems","kotkov"],"title":"Serendipity in recommender systems","year":2018,"dataSources":["5Dp4QphkvpvNA33zi","jfoasiDDpStqkkoZB","BiuuFc45aHCgJqDLY"]}