Improving recommendation lists through topic diversification

By Cai-Nicolas Ziegler, Sean M. McNee, J. Konstan, Georg Lausen. 2005

2,119 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Explaining machine learning classifiers through diverse counterfactual explanations (2019, background): “In other domains of information search such as search engines and recommendation systems, multiple studies [15, 22, 35, 45] show the benefits of presenting a diverse set of information items to a user.”
Ranking with Fairness Constraints (2017, background): “…w ij s to attempt to capture the amount of diversity item i would introduce at position k conditioned on the items that were placed at positions 1 , . . . , k − 1 (see [CG98, ZH08, CKC + 08, ZCL03, ZMKL05]), or casting it directly as an (unconstrained) multi-objective optimization problem [YS16].”
Evaluating Recommendation Systems (2011, methodology): “Then, we could measure the diversity of a list based on the sum, average, min, or max distance between item pairs, or measure the value of adding each item to the recommendation list as the new item’s diversity from the items already in the list [64, 6].”
fLDA: matrix factorization through latent dirichlet allocation (2010, neutral): “[31] collected book ratings from the site.”
Recommender Systems (2012, background): “Notably, intra-list diversity can be used to enhance improve recommendation lists by avoiding recommendation of excessively similar objects [37].”
Current challenges and visions in music recommender systems research (2017, methodology): “to compute pairwise distance between all items in the recommendation set, either averaged [196] or summed [173].”
Being accurate is not enough: how accuracy metrics have hurt recommender systems (2006, background): “We claim there are many different aspects to the recommendation process which current accuracy metrics do not measure.”
Novelty and Diversity in Recommender Systems (2015, methodology): “Being similar to the profile, recommended items are also similar among themselves, which explains the low EILD.”
Diversity in Big Data: A Review (2017, neutral): “Novelty-based diversity measures deﬁne diversity with respect to the elements seen in the past (Refs.”
Auralist: introducing serendipity into music recommendation (2012, methodology): “In particular, we support Ziegler [29] and Ratner [19] s .ndings that users are willing to sacri.ce some amount of accuracy for improved novelty/diversity/serendipity performance, and that such systems are more satisfying overall.”
Multistakeholder recommendation: Survey and research directions (2020, background): “See, for example, [1, 120, 141, 127, 67].”
Bridging Language and Items for Retrieval and Recommendation: Benchmarking LLMs as Semantic Encoders (2024, positive): “Additionally, we include the widely used Book-Crossing dataset ( BX ) (Ziegler et al., 2005).”
Time-aware recommender systems: a comprehensive survey and analysis of existing evaluation protocols (2013, background): “Apart from prediction accuracy and ranking precision, more recently, other recommendation properties are starting to be investigated, such as novelty and diversity (Vargas and Castells 2011), by means of metrics like Self Information (Zhou et al. 2010), and Intra List Similarity (Ziegler et al. 2005”