Recommender systems: from algorithms to user experience
Konstan, J. & Riedl, J.
Since their introduction in the early 1990’s, automated recommender systems have revolutionized the marketing and delivery of commerce and content by providing personalized recommendations and predictions over a variety of large and complex product offerings. In this article, we review the key advances in collaborative filtering recommender systems, focusing on the evolution from research concentrated purely on algorithms to research concentrated on the rich set of questions around the user experience with the recommender. We show through examples that the embedding of the algorithm in the user experience dramatically affects the value to the user of the recommender. We argue that evaluating the user experience of a recommender requires a broader set of measures than have been commonly used, and suggest additional measures that have proven effective. Based on our analysis of the state of the field, we identify the most important open research problems, and outline key challenges slowing the advance of the state of the art, and in some cases limiting the relevance of research to real-world applications.
Empirical Comparison of Algorithms for Network Community Detection
Leskovec, J.; Lang, K. J. & Mahoney, M. W.
Detecting clusters or communities in large real-world graphs such as large
cial or information networks is a problem of considerable interest. In
actice, one typically chooses an objective function that captures the
tuition of a network cluster as set of nodes with better internal
nnectivity than external connectivity, and then one applies approximation
gorithms or heuristics to extract sets of nodes that are related to the
jective function and that "look like" good communities for the application of
terest. In this paper, we explore a range of network community detection
thods in order to compare them and to understand their relative performance
d the systematic biases in the clusters they identify. We evaluate several
mmon objective functions that are used to formalize the notion of a network
mmunity, and we examine several different classes of approximation algorithms
at aim to optimize such objective functions. In addition, rather than simply
xing an objective and asking for an approximation to the best cluster of any
ze, we consider a size-resolved version of the optimization problem.
nsidering community quality as a function of its size provides a much finer
ns with which to examine community detection algorithms, since objective
nctions and approximation algorithms often have non-obvious size-dependent
havior.
Evaluation of Collaborative Filtering Algorithms for Recommending Articles on CiteULike
Parra, D. & Brusilovsky, P.
Motivated by the potential use of collaborative tagging systems to develop new recommender systems, we have implemented and compared three variants of user-based collaborative filtering algorithms to provide recommendations of articles on CiteULike. On our first approach, Classic Collaborative filtering (CCF), we use Pearson correlation to calculate similarity between users and a classic adjusted ratings formula to rank the recommendations. Our second approach, Neighbor-weighted Collaborative Filtering (NwCF), incorporates the amount of raters in the ranking formula of the recommendations. A modified version of the Okapi BM25 IR model over users ’ tags is implemented on our third approach to form the user neighborhood. Our results suggest that incorporating the number of raters into the algorithms leads to an improvement of precision, and they also support that tags can be considered as an alternative to Pearson correlation to calculate the similarity between users and their neighbors in a collaborative tagging system.
Prediction and ranking algorithms for event-based network data
O'Madadhain, J.; Hutchins, J. & Smyth, P.
Event-based network data consists of sets of events over time, each of which may involve multiple entities. Examples include email traffic, telephone calls, and research publications (interpreted as co-authorship events). Traditional network analysis techniques, such as social network models, often aggregate the relational information from each event into a single static network. In contrast, in this paper we focus on the temporal nature of such data. In particular, we look at the problems of temporal link prediction and node ranking, and describe new methods that illustrate opportunities for data mining and machine learning techniques in this context. Experimental results are discussed for a large set of co-authorship events measured over multiple years, and a large corporate email data set spanning 21 months.