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.

Recommender systems have been part of the Internet for almost two decades. Dozens of vendors have built recommendation technologies and taken them to market in two waves, roughly aligning with the web 1.0 and 2.0 revolutions. Today recommender systems are found in a multitude of online services. They have been developed using a variety of techniques and user interfaces. They have been nurtured with millions of users’ explicit and implicit preferences (most often with their permission). Frequently they provide relevant recommendations that increase the revenue or user engagement of the online services that operate them. However, when we evaluate the current generation of recommender systems from the point of view of the “recommendee,” we find that most recommender systems serve the goals of the business instead of their users’ interests. Thus we believe that the big promise of recommender systems has yet to be fulfilled. We foresee a third wave of recommender systems that act directly on behalf of their users across a range of domains instead of acting as a sales assistant. We also predict that such new recommender systems will better deal with information overload, take advantage of contextual clues from mobile devices, and utilize the vast information and computation stores available through cloud-computing services to maximize users’ long-term goals

We propose a scalable, efficient and statistically motivated computational framework for Graphical Lasso (Friedman et al., 2007b) - a covariance regularization framework that has received significant attention in the statistics community over the past few years. Existing algorithms have trouble in scaling to dimensions larger than a thousand. Our proposal significantly enhances the state-of-the-art for such moderate sized problems and gracefully scales to larger problems where other algorithms become practically infeasible. This requires a few key new ideas. We operate on the primal problem and use a subtle variation of block-coordinate-methods which drastically reduces the computational complexity by orders of magnitude. We provide rigorous theoretical guarantees on the convergence and complexity of our algorithm and demonstrate the effectiveness of our proposal via experiments. We believe that our framework extends the applicability of Graphical Lasso to large-scale modern applications like bioinformatics, collaborative filtering and social networks, among others.

Information diffusion and virus propagation are fundamental processes taking place in networks. While it is often possible to directly observe when nodes become infected with a virus or adopt the information, observing individual transmissions (i.e., who infects whom, or who influences whom) is typically very difficult. Furthermore, in many applications, the underlying network over which the diffusions and propagations spread is actually unobserved. We tackle these challenges by developing a method for tracing paths of diffusion and influence through networks and inferring the networks over which contagions propagate. Given the times when nodes adopt pieces of information or become infected, we identify the optimal network that best explains the observed infection times. Since the optimization problem is NP-hard to solve exactly, we develop an efficient approximation algorithm that scales to large datasets and finds provably near-optimal networks. We demonstrate the effectiveness of our approach by tracing information diffusion in a set of 170 million blogs and news articles over a one year period to infer how information flows through the online media space. We find that the diffusion network of news for the top 1,000 media sites and blogs tends to have a core-periphery structure with a small set of core media sites that diffuse information to the rest of the Web. These sites tend to have stable circles of influence with more general news media sites acting as connectors between them.

Conflict is an essential element of interesting stories. In previous work, we proposed a formal model of narrative conflict. We also described 7 dimensions which can be used to distinguish one conflict from another: participants, subject, duration, balance, directness, intensity, and resolution. This paper presents the results of an experiment designed to measure how well our metrics for balance, directness, intensity, and resolution predict the responses of human readers when asked to measure these same values in a set of four stories. We conclude that our metrics are able to rank stories similarly to human readers.

Interactive Narrative often involves dialogue with virtual dramatic characters. In this paper we compare two kinds of models of character style: one based on models derived from the Big Five theory personality, and the other derived from a corpus-based method applied to characters and films from the IMSDb archive. We apply these models to character utterances for a pilot narrative-based outdoor augmented reality game called Murder in the Arboretum . We use an objective quantitative metric to estimate the quality of a character model, with the aim of predicting model quality without perceptual experiments. We show that corpus-based character models derived from individual characters are often more detailed and specific than personality based models, but that there is a strong correlation between personality judgments of original character dialogue and personality judgments of utterances generated for Murder in the Arboretum that use the derived character models.

Many forms of interactive digital entertainment involve interacting with virtual dramatic characters. Our long term goal is to procedurally generate character dialogue behavior that automatically mimics, or blends, the style of existing characters. In this paper, we show how linguistic elements in character dialogue can define the style of characters in our RPG SpyFeet. We utilize a corpus of 862 film scripts from the IMSDb website, representing 7,400 characters, 664,000 lines of dialogue and 9,599,000 word tokens. We utilize counts of linguistic reflexes that have been used previously for personality or author recognition to discriminate different character types. With classification experiments, we show that different types of characters can be distinguished at accuracies up to 83% over a baseline of 20%. We discuss the characteristics of the learned models and show how they can be used to mimic particular film characters.

This paper applies machine learning and the mathematics of chaos to the task of designing indoor rock-climbing routes. Chaotic variation has been used to great advantage on music and dance, but the challenges here are quite different, beginning with the representation. We present a formalized system for transcribing rock climbing problems, then describe a variation generator that is designed to support human route-setters in designing new and interesting climbing problems. This variation generator, termed Strange Beta, combines chaos and machine learning, using the former to introduce novelty and the latter to smooth transitions in a manner that is consistent with the style of the climbs This entails parsing the domain-specific natural language that rock climbers use to describe routes and movement and then learning the patterns in the results. We validated this approach with a pilot study in a small university rock climbing gym, followed by a large blinded study in a commercial climbing gym, in cooperation with experienced climbers and expert route setters. The results show that {\sc Strange Beta} can help a human setter produce routes that are at least as good as, and in some cases better than, those produced in the traditional manner.

Modelling the real world complexity of music is a challenge for machine learning. We address the task of modeling melodic sequences from the same music genre. We perform a comparative analysis of two probabilistic models; a Dirichlet Variable Length Markov Model (Dirichlet-VMM) and a Time Convolutional Restricted Boltzmann Machine (TC-RBM). We show that the TC-RBM learns descriptive music features, such as underlying chords and typical melody transitions and dynamics. We assess the models for future prediction and compare their performance to a VMM, which is the current state of the art in melody generation. We show that both models perform significantly better than the VMM, with the Dirichlet-VMM marginally outperforming the TC-RBM. Finally, we evaluate the short order statistics of the models, using the Kullback-Leibler divergence between test sequences and model samples, and show that our proposed methods match the statistics of the music genre significantly better than the VMM.

An answer to a query has a well-defined lineage expression (alternatively called how-provenance) that explains how the answer was derived. Recent work has also shown how to compute the lineage of a non-answer to a query. However, the cause of an answer or non-answer is a more subtle notion and consists, in general, of only a fragment of the lineage. In this paper, we adapt Halpern, Pearl, and Chockler's recent definitions of causality and responsibility to define the causes of answers and non-answers to queries, and their degree of responsibility. Responsibility captures the notion of degree of causality and serves to rank potentially many causes by their relative contributions to the effect. Then, we study the complexity of computing causes and responsibilities for conjunctive queries. It is known that computing causes is NP-complete in general. Our first main result shows that all causes to conjunctive queries can be computed by a relational query which may involve negation. Thus, causality can be computed in PTIME, and very efficiently so. Next, we study computing responsibility. Here, we prove that the complexity depends on the conjunctive query and demonstrate a dichotomy between PTIME and NP-complete cases. For the PTIME cases, we give a non-trivial algorithm, consisting of a reduction to the max-flow computation problem. Finally, we prove that, even when it is in PTIME, responsibility is complete for LOGSPACE, implying that, unlike causality, it cannot be computed by a relational query.