Gediminas Adomavicius

Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions

This paper presents an overview of the field of recommender systems and describes the current generation of recommendation methods that are usually classified into the following three main categories: content-based, collaborative, and hybrid recommendation approaches. This paper also describes various limitations of current recommendation methods and discusses possible extensions that can improve recommendation capabilities and make recommender systems applicable to an even broader range of applications. These extensions include, among others, an improvement of understanding of users and items, incorporation of the contextual information into the recommendation process, support for multicriteria ratings, and a provision of more flexible and less intrusive types of recommendations.

Incorporating contextual information in recommender systems using a multidimensional approach

The article presents a multidimensional (MD) approach to recommender systems that can provide recommendations based on additional contextual information besides the typical information on users and items used in most of the current recommender systems. This approach supports multiple dimensions, profiling information, and hierarchical aggregation of recommendations. The article also presents a multidimensional rating estimation method capable of selecting two-dimensional segments of ratings pertinent to the recommendation context and applying standard collaborative filtering or other traditional two-dimensional rating estimation techniques to these segments. A comparison of the multidimensional and two-dimensional rating estimation approaches is made, and the tradeoffs between the two are studied. Moreover, the article introduces a combined rating estimation method, which identifies the situations where the MD approach outperforms the standard two-dimensional approach and uses the MD approach in those situations and the standard two-dimensional approach elsewhere. Finally, the article presents a pilot empirical study of the combined approach, using a multidimensional movie recommender system that was developed for implementing this approach and testing its performance.

Context-Aware Recommender Systems

Context-aware recommender systems (CARS) generate more relevant recommendations by adapting them to the specific contextual situation of the user. This article explores how contextual information can be used to create more intelligent and useful recommender systems. It provides an overview of the multifaceted notion of context, discusses several approaches for incorporating contextual information in recommendation process, and illustrates the usage of such approaches in several application areas where different types of contexts are exploited. The article concludes by discussing the challenges and future research directions for context-aware recommender systems.

New Recommendation Techniques for Multicriteria Rating Systems

Personalization technologies and recommender systems help online consumers avoid information overload by making suggestions regarding which information is most relevant to them. Most online shopping sites and many other applications now use recommender systems. Two new recommendation techniques leverage multicriteria ratings and improve recommendation accuracy as compared with single-rating recommendation approaches. Taking full advantage of multicriteria ratings in personalization applications requires new recommendation techniques. In this article, we propose several new techniques for extending recommendation technologies to incorporate and leverage multicriteria rating information.

Improving Aggregate Recommendation Diversity Using Ranking-Based Techniques

Recommender systems are becoming increasingly important to individual users and businesses for providing personalized recommendations. However, while the majority of algorithms proposed in recommender systems literature have focused on improving recommendation accuracy (as exemplified by the recent Netflix Prize competition), other important aspects of recommendation quality, such as the diversity of recommendations, have often been overlooked. In this paper, we introduce and explore a number of item ranking techniques that can generate substantially more diverse recommendations across all users while maintaining comparable levels of recommendation accuracy. Comprehensive empirical evaluation consistently shows the diversity gains of the proposed techniques using several real-world rating data sets and different rating prediction algorithms.

A Parallel Multilevel Method for Adaptively Refined Cartesian Grids with Embedded Boundaries

. Asrecently as five to ten years ago, mesh generation was fre-quently the most time consuming task in a typical CFD cycle.Adaptive Cartesian mesh generation methods are capable ofproducing millions of cells around complex geometries inminutes and have substantially removed this bottleneck.Why write yet another Euler solver? With robust mesh gener-ation largely in-hand, solution time resurfaces as the pacingitem in the CFD cycle. The current work attacks this issue bydesigning a scalable, accurate Cartesian solver with robustmultigrid convergence acceleration. Our primary motivationis to gain efficiency by capitalizing on the simplifications andspecialized data structures available on Cartesian grids. Sig-nificant savings in both CPU time and storage may be real-ized by taking advantage of the fact that cell faces arecoordinate aligned. In addition, higher-order methods withgood limiters are generally easier to design and perform morerobustly on uniform Cartesian meshes.Secondly, in any embedded-boundary Cartesian solver, thebody-intersectingcut-cellsdemand special attention. Thesecells can impose a substantial burden on the numerical dis-cretization since the arbitrary nature of geometric intersec-tion implies that a cut-cell may be orders of magnitudesmaller than its neighboring cells. This fact contrasts sharplywith the comparatively smooth meshes that are generallyfound on a good quality structured or unstructured mesh.Substantial research into these cut-cell issues have been stud-ied by references [9],[10],[6],[8], and [12] (among others)and we hope to take advantage of this investment.Thirdly, this work investigates a multigrid strategy that isspecialized for adaptively refined Cartesian meshes. In ourapproach, all grids in the multigrid hierarchy cover the entiredomain and include cells at many refinement levels. Thesmoother therefore iterates over the entire domain when it isinvoked on any grid in the hierarchy. In this respect, theapproach shares more with agglomeration or algebraic multi-grid techniques than with many other Cartesian or AMRmethods which iterate over only cells at the same level ofrefinement

Expert-Driven Validation of Rule-Based User Models in Personalization Applications

In many e-commerce applications, ranging from dynamic Web content presentation, to personalized ad targeting, to individual recommendations to the customers, it is important to build personalized profiles of individual users from their transactional histories. These profiles constitute models of individual user behavior and can be specified with sets of rules learned from user transactional histories using various data mining techniques. Since many discovered rules can be spurious, irrelevant, or trivial, one of the main problems is how to perform post-analysis of the discovered rules, i.e., how to validate user profiles by separating “good” rules from the “bad.” This validation process should be done with an explicit participation of the human expert. However, complications may arise because there can be very large numbers of rules discovered in the applications that deal with many users, and the expert cannot perform the validation on a rule-by-rule basis in a reasonable period of time. This paper presents a framework for building behavioral profiles of individual users. It also introduces a new approach to expert-driven validation of a very large number of rules pertaining to these users. In particular, it presents several types of validation operators, including rule grouping, filtering, browsing, and redundant rule elimination operators, that allow a human expert validate many individual rules at a time. By iteratively applying such operators, the human expert can validate a significant part of all the initially discovered rules in an acceptable time period. These validation operators were implemented as a part of a one-to-one profiling system. The paper also presents a case study of using this system for validating individual user rules discovered in a marketing application.

User profiling in personalization applications through rule discovery and validation

Gediminas Adomavicius New York University adomavic@cs.nyu.edu In many applications, ranging from recommender systems to one-to-one marketing to Web browsing, it is important to build personalized profiles of individual users from their transactional histories. These profiles describe individual behavior of users and can be specified with sets of rules learned from user transactional histories using various data mining techniques. Since many discovered rules can be spurious, irrelevant, or trivial, one of the main problems is how to perform post-analysis of the discovered rules, i.e., how to validate customer profiles by separating “good” rules from the “bad.” This paper presents a method for validating such rules with an explicit participation of a human expert

Multi-Criteria Recommender Systems

This chapter aims to provide an overview of the class of multi-criteria recommender systems, i.e., the category of recommender systems that use multi-criteria preference ratings. Traditionally, the vast majority of recommender systems literature has focused on providing recommendations by modelling a user’s utility (or preference) for an item as a single preference rating. However, where possible, capturing richer user preferences along several dimensions—for example, capturing not only the user’s overall preference for a given movie but also her preferences for specific movie aspects (such as acting, story, or visual effects)—can provide opportunities for further improvements in recommendation quality. As a result, a number of recommendation techniques that attempt to take advantage of such multi-criteria preference information have been developed in recent years. A review of current algorithms that use multi-criteria ratings for calculating predictions and generating recommendations is provided. The chapter concludes with a discussion on open issues and future challenges for the class of multi-criteria rating recommenders.

Making sense of technology trends in the information technology landscape: a design science approach

A major problem for firms making information technology investment decisions is predicting and understanding the effects of future technological developments on the value of present technologies. Failure to adequately address this problem can result in wasted organization resources in acquiring, developing, managing, and training employees in the use of technologies that are short-lived and fail to produce adequate return on investment. The sheer number of available technologies and the complex set of relationships among them make IT landscape analysis extremely challenging. Most IT-consuming firms rely on third parties and suppliers for strategic recommendations on IT investments, which can lead to biased and generic advice. We address this problem by defining a new set of constructs and methodologies upon which we develop an IT ecosystem model. The objective of these artifacts is to provide a formal problem representation structure for the analysis of information technology development trends and to reduce the complexity of the IT landscape for practitioners making IT investment decisions. We adopt a process theory perspective and use a combination of visual mapping and quantification strategies to develop our artifacts and a state diagram-based technique to represent evolutionary transitions over time. We illustrate our approach using two exemplars: digital music technologies and wireless networking technologies. We evaluate the utility of our approach by conducting in-depth interviews with IT industry experts and demonstrate the contribution of our approach relative to existing techniques for technology forecasting.