As digital streaming media matures, consumers and publishers have become interested in short duration ads (i.e., <10 s). However, previous research on memory retention for short-duration ad content utilized video or audio-video experiences where attention is directed toward the ad. Therefore, it is unclear whether this study is relevant to audio-only content, and it is unknown if similar memory retention may be achieved when attention is not directed toward the ad (i.e., incidental memory). To study the incidental memory of short audio ads, participants were recruited to play a driving game while listening to music with periodic ad breaks (pods). Each pod contained a 2, 4, and 8-s ad. Results showed each ad duration was able to leave an impression on both recall and recognition memory. Whereas the duration of the ad generally did not affect incidental memory, ads placed first or last in a pod were remembered best. Together, the data indicates that audio ads less than 10-s are a viable option to raise awareness for brand name, product name and company location. Moreover, these results indicate audio ads may leave an impression even when the listener is engaged in other tasks, such as driving or playing a video game.

This corrects the article DOI: 10.1103/PhysRevLett.92.062301.

This corrects the article DOI: 10.1103/PhysRevLett.89.132301.

We address the following problem: How do we incorporate user item interaction signals as part of the relevance model in a large-scale personalized recommendation system such that, (1) the ability to interpret the model and explain recommendations is retained, and (2) the existing infrastructure designed for the (user profile) content-based model can be leveraged? We propose Dionysius, a hierarchical graphical model based framework and system for incorporating user interactions into recommender systems, with minimal change to the underlying infrastructure. We learn a hidden fields vector for each user by considering the hierarchy of interaction signals, and replace the user profile-based vector with this learned vector, thereby not expanding the feature space at all. Thus, our framework allows the use of existing recommendation infrastructure that supports content based features. We implemented and deployed this system as part of the recommendation platform at LinkedIn for more than one year. We validated the efficacy of our approach through extensive offline experiments with different model choices, as well as online A/B testing experiments. Our deployment of this system as part of the job recommendation engine resulted in significant improvement in the quality of retrieved results, thereby generating improved user experience and positive impact for millions of users.

Recommender systems typically leverage two types of signals to effectively recommend items to users: user activities and content matching between user and item profiles, and recommendation models in literature are usually categorized into collaborative filtering models, content-based models and hybrid models. In practice, when rich profiles about users and items are available, and user activities are sparse (cold-start), effective content matching signals become much more important in the relevance of the recommendation. The de-facto method to measure similarity between two pieces of text is computing the cosine similarity of the two bags of words, and each word is weighted by TF (term frequency within the document) x IDF (inverted document frequency of the word within the corpus). In general sense, TF can represent any local weighting scheme of the word within each document, and IDF can represent any global weighting scheme of the word across the corpus. In this paper, we focus on the latter, i.e., optimizing the global term weights, for a particular recommendation domain by leveraging supervised approaches. The intuition is that some frequent words (lower IDF, e.g. ``database'') can be essential and predictive for relevant recommendation, while some rare words (higher IDF, e.g. the name of a small company) could have less predictive power. Given plenty of observed activities between users and items as training data, we should be able to learn better domain-specific global term weights, which can further improve the relevance of recommendation. We propose a unified method that can simultaneously learn the weights of multiple content matching signals, as well as global term weights for specific recommendation tasks. Our method is efficient to handle large-scale training data generated by production recommender systems. And experiments on LinkedIn job recommendation data justify the effectiveness of our approach.

Downside management is an important topic in the field of recommender systems. User satisfaction increases when good items are recommended, but satisfaction drops significantly when bad recommendations are pushed to them. For example, a parent would be disappointed if violent movies are recommended to their kids and may stop using the recommendation system entirely. A vegetarian would feel steak-house recommendations useless. A CEO in a mid-sized company would feel offended by receiving intern-level job recommendations. Under circumstances where there is penalty for a bad recommendation, a bad recommendation is worse than no recommendation at all. While most existing work focuses on upside management (recommending the best items to users), this paper emphasizes on achieving better downside management (reducing the recommendation of irrelevant or offensive items to users). The approach we propose is general and can be applied to any scenario or domain where downside management is key to the system. To tackle the problem, we design a user latent preference model to predict the user preference in a specific dimension, say, the dietary restrictions of the user, the acceptable level of adult content in a movie, or the geographical preference of a job seeker. We propose to use multinomial regression as the core model and extend it with a hierarchical Bayesian framework to address the problem of data sparsity. After the user latent preference is predicted, we leverage it to filter out downside items. We validate the soundness of our approach by evaluating it with an anonymous job application dataset on LinkedIn. The effectiveness of the latent preference model was demonstrated in both offline experiments and online A/B testings. The user latent preference model helps to improve the VPI (views per impression) and API (applications per impression) significantly which in turn achieves a higher user satisfaction.

This study investigated whether volunteer experience compensates for a gap in employment that occurs either early or late in one's career. Recruiters (n=82) evaluated resumes of fictitious applicants with either early or late employment gaps, plus one of three types of volunteer experience: career-related, career-unrelated, and none. For applicants with an employment gap, resumes with volunteer experience - regardless of its career-relatedness - were not rated significantly higher than resumes without volunteer experience. Although not statistically significant, resumes with late employment gaps were rated highest when they had career-related volunteer experience and lowest when they had no volunteer experience. In line with human capital theory, applicants late in their career were rated higher than applicants early in their career.

Generally discussed herein are methods, systems, and apparatuses for determining a latent preference of a user. One or more embodiments, discussed herein regard determining a latent preference of a user's propensity to relocate for a job. According to an example, a method can include receiving one or more characteristics of a user of a web service, estimating a probability corresponding to a latent preference of the user, and/or determining whether the probability indicates the user has the latent preference.

The NA44 Collaboration has measured yields and differential distributions of K, Preprint submitted to Elsevier Preprint 28 January 2013 K−, π, π− in transverse kinetic energy and rapidity, around the center-of-mass rapidity in 158 A GeV/c Pb+Pb collisions at the CERN SPS. A considerable enhancement of K production per π is observed, as compared to p + p collisions at this energy. To illustrate the importance of secondary hadron rescattering as an enhancement mechanism, we compare strangeness production at the SPS and AGS with predictions of the transport model RQMD.

Bright has built an automated system for ranking job candidates against job descriptions. The candidate's resume and social media profiles are interwoven to build an augmented user profile. Similarly, the job description is augmented by external databases and user-generated content to build an enhanced job profile. These augmented user and job profiles are then analyzed in order to develop numerical overlap features each with strong discriminating power, and in sum with maximal coverage. The resulting feature scores are then combined into a single Bright Score using a custom algorithm, where the feature weights are derived from a nation-wide and controlled study in which we collected a large sample of human judgments on real resume-job pairings. We demonstrate that the addition of social media profile data and external data improves the classification accuracy dramatically in terms of identifying the most qualified candidates.