Novelty and Diversity in Recommender Systems

Castells, Pablo; Hurley, Neil; Vargas, Saúl

doi:10.1007/978-1-0716-2197-4_16

Cited by 50 publications

(30 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Participants' ratings were used to calculate the pairwise cosine similarity matrix for the movies, and the inverse of the average similarity of movies in a user's profile was used to indicate the diversity of their movie viewing profile. (Castells et al, 2015; Eskandanian et al, 2016; Hurley & Zhang, 2011).

italicILD ((), L) goodbreak= \frac{1}{(||, L) ((), (||, L) goodbreak- 1)} \sum_{i \in L} \sum_{j \in L} \frac{1}{similarity ((), i, j)},

…”

Section: Resultsmentioning

confidence: 99%

“…Traditionally, evaluations of recommender systems have largely focused on prediction accuracy. The accuracy paradigm is efficient for the evaluation of recommendation algorithms, yet accuracy‐based measures fail to fully reflect other values users might derive from the recommender systems (Castells et al, 2015; Herlocker et al, 2004; Konstan & Riedl, 2012; McNee et al, 2006). Lack of diversity in the recommendation set is often identified (Bradley & Smyth, 2001).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Preference diversity and openness to novelty: Scales construction from the perspective of movie recommendation

Tang

Liao

2022

Asso for Info Science & Tech

View full text Add to dashboard Cite

In response to calls for recommender systems to balance accuracy and alternative measures such as diversity and novelty, we propose that recommendation strategies should be applied adaptively according to users' preference traits. Psychological scales for "preference diversity" and "openness to novelty" were developed to measure users' willingness to accept diverse and novel recommendations, respectively. To validate the scales empirically, a user study was conducted in which 293 regular moviegoers were asked to judge a set of 220 movies representing both mainstream and "long-tail" appeals. The judgment task involved indicating and rating movies they had seen, heard of but not seen, and not known previously. Correlatoin analyses were then conducted

show abstract

italicILD ((), L) goodbreak= \frac{1}{(||, L) ((), (||, L) goodbreak- 1)} \sum_{i \in L} \sum_{j \in L} \frac{1}{similarity ((), i, j)},

…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Preference diversity and openness to novelty: Scales construction from the perspective of movie recommendation

Tang

Liao

2022

Asso for Info Science & Tech

View full text Add to dashboard Cite

show abstract

“…While diversity preference is overlooked by existing link recommendation methods, diversity has been studied in the related area of recommender systems (Castells et al, 2015;Kaminskas and Bridge, 2016;Wu et al, 2019). Recommender systems recommend items (e.g., books) to users (Adomavicius and Tuzhilin, 2005), whereas link recommendation methods recommend users (e.g., friends) to users (Li et al, 2017b).…”

Section: Diversification Methods For Recommender Systemsmentioning

confidence: 99%

“…Because the DPA-LR problem is a new research problem, no existing method has been developed for this problem. A closely related problem is the diversification problem in the field of recommender systems, and various diversification methods have been proposed to diversify recommendations by recommender systems (Castells et al, 2015;Wu et al, 2019). Moreover, prior research has adapted diversification methods to improve the diversity of link recommendations (Sanz-Cruzado and Castells, 2018).…”

Section: Data and Benchmark Methodsmentioning

confidence: 99%

Diversity Preference-Aware Link Recommendation for Online Social Networks

Yin¹,

Xiao²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Link recommendation, which recommends links to connect unlinked online social network users, is a fundamental social network analytics problem with ample business implications. Existing link recommendation methods tend to recommend similar friends to a user but overlook the user's diversity preference, although social psychology theories suggest the criticality of diversity preference to link recommendation performance. In recommender systems, a field related to link recommendation, a number of diversification methods have been proposed to improve the diversity of recommended items. Nevertheless, diversity preference is distinct from diversity studied by diversification methods. To address these research gaps, we define and operationalize the concept of diversity preference for link recommendation and propose a new link recommendation problem: the diversity preference-aware link recommendation problem. We then analyze key properties of the new link recommendation problem and develop a novel link recommendation method to solve the problem. Using two large-scale online social network data sets, we conduct extensive empirical evaluations to demonstrate the superior performance of our method over representative diversification methods adapted for link recommendation as well as state-of-the-art link recommendation methods.

show abstract

“…This means that, unlike in search, the utility from an item depends on whether or how frequently an item has been experienced by the user. Additionally, the diversity of the recommendations may also be considered [18].…”

Section: Evaluation Criteriamentioning

confidence: 99%

On Natural Language User Profiles for Transparent and Scrutable Recommendation

Radlinski,

Balog,

Diaz

et al. 2022

Preprint

View full text Add to dashboard Cite

Natural interaction with recommendation and personalized search systems has received tremendous attention in recent years. We focus on the challenge of supporting people's understanding and control of these systems and explore a fundamentally new way of thinking about representation of knowledge in recommendation and personalization systems. Specifically, we argue that it may be both desirable and possible for algorithms that use natural language representations of users' preferences to be developed. We make the case that this could provide significantly greater transparency, as well as affordances for practical actionable interrogation of, and control over, recommendations. Moreover, we argue that such an approach, if successfully applied, may enable a major step towards systems that rely less on noisy implicit observations while increasing portability of knowledge of one's interests.

show abstract

Novelty and Diversity in Recommender Systems

Cited by 50 publications

References 64 publications

Preference diversity and openness to novelty: Scales construction from the perspective of movie recommendation

Preference diversity and openness to novelty: Scales construction from the perspective of movie recommendation

Diversity Preference-Aware Link Recommendation for Online Social Networks

On Natural Language User Profiles for Transparent and Scrutable Recommendation

Contact Info

Product

Resources

About