skip to main content

Attention:

The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 11:00 PM ET on Friday, November 15 until 2:00 AM ET on Saturday, November 16 due to maintenance. We apologize for the inconvenience.


Title: Distributionally-Informed Recommender System Evaluation
Current practice for evaluating recommender systems typically focuses on point estimates of user-oriented effectiveness metrics or business metrics, sometimes combined with additional metrics for considerations such as diversity and novelty. In this paper, we argue for the need for researchers and practitioners to attend more closely to various distributions that arise from a recommender system (or other information access system) and the sources of uncertainty that lead to these distributions. One immediate implication of our argument is that both researchers and practitioners must report and examine more thoroughly the distribution of utility between and within different stakeholder groups. However, distributions of various forms arise in many more aspects of the recommender systems experimental process, and distributional thinking has substantial ramifications for how we design, evaluate, and present recommender systems evaluation and research results. Leveraging and emphasizing distributions in the evaluation of recommender systems is a necessary step to ensure that the systems provide appropriate and equitably-distributed benefit to the people they affect.  more » « less
Award ID(s):
1751278
NSF-PAR ID:
10461937
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
ACM Transactions on Recommender Systems
ISSN:
2770-6699
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. While other areas of machine learning have seen more and more automation, designing a high-performing recommender system still requires a high level of human effort. Furthermore, recent work has shown that modern recommender system algorithms do not always improve over well-tuned baselines. A natural follow-up question is, "how do we choose the right algorithm for a new dataset and performance metric?" In this work, we start by giving the first large-scale study of recommender system approaches by comparing 24 algorithms and 100 sets of hyperparameters across 85 datasets and 315 metrics. We find that the best algorithms and hyperparameters are highly dependent on the dataset and performance metric. However, there is also a strong correlation between the performance of each algorithm and various meta-features of the datasets. Motivated by these findings, we create RecZilla, a meta-learning approach to recommender systems that uses a model to predict the best algorithm and hyperparameters for new, unseen datasets. By using far more meta-training data than prior work, RecZilla is able to substantially reduce the level of human involvement when faced with a new recommender system application. We not only release our code and pretrained RecZilla models, but also all of our raw experimental results, so that practitioners can train a RecZilla model for their desired performance metric: https://github.com/naszilla/reczilla. 
    more » « less
  2. null (Ed.)
    The growing amount of online information today has increased opportunity to discover interesting and useful information. Various recommender systems have been designed to help people discover such information. No matter how accurately the recommender algorithms perform, users’ engagement with recommended results has been complained being less than ideal. In this study, we touched on two human-centered objectives for recommender systems: user satisfaction and curiosity, both of which are believed to play roles in maintaining user engagement and sustain such engagement in the long run. Specifically, we leveraged the concept of surprise and used an existing computational model of surprise to identify relevantly surprising health articles aiming at improving user satisfaction and inspiring their curiosity. We designed a user study to first test the validity of the surprise model in a health news recommender system, called LuckyFind. Then user satisfaction and curiosity were evaluated. We find that the computational surprise model helped identify surprising recommendations at little cost of user satisfaction. Users gave higher ratings on interestingness than usefulness for those surprising recommendations. Curiosity was inspired more for those individuals who have a larger capacity to experience curiosity. Over half of the users have changed their preferences after using LuckyFind, either discovering new areas, reinforcing their existing interests, or stopping following those they did not want anymore. The insights of the research will make researchers and practitioners rethink the objectives of today’s recommender systems as being more human-centered beyond algorithmic accuracy. 
    more » « less
  3. Today’s recommender systems are criticized for recommending items that are too obvious to arouse users’ interest. That is why the recommender systems research community has advocated some ”beyond accuracy” evaluation metrics such as novelty, diversity, coverage, and serendipity with the hope of promoting information discovery and sustain users’ interest over a long period of time. While bringing in new perspectives, most of these evaluation metrics have not considered individual users’ difference: an open-minded user may favor highly novel or diversified recommendations whereas a conservative user’s appetite for novelty or diversity may not be that large. In this paper, we developed a model to approximate an individual’s curiosity distribution over different levels of stimuli guided by the well-known Wundt curve in Psychology. We measured an item’s surprise level to assess the stimulation level and whether it is in the range of the user’s appetite for stimulus. We then proposed a recommendation system framework that considers both user preference and appetite for stimulus where the curiosity is maximally aroused. Our framework differs from a typical recommender system in that it leverages human’s curiosity to promote intrinsic interest with the system. A series of evaluation experiments have been conducted to show that our framework is able to rank higher the items with not only high ratings but also high response likelihood. The recommendation list generated by our algorithm has higher potential of inspiring user curiosity compared to traditional approaches. The personalization factor for assessing the stimulus (surprise) strength further helps the recommender achieve smaller (better) inter-user similarity. 
    more » « less
  4. Offline evaluation protocols for recommender systems are intended to estimate users' satisfaction with recommendations using static data from prior user interactions. These evaluations allow researchers and production developers to carry out first-pass estimates of the likely performance of a new system and weed out bad ideas before presenting them to users. However, offline evaluations cannot accurately assess novel, relevant recommendations, because the most novel recommendations items that were previously unknown to the user; such items are missing from the historical data, so they cannot be judged as relevant. A breakthrough that reliably produces novel, relevant recommendations would score poorly with current offline evaluation techniques. While the existence of this problem is noted in the literature, its extent is not well-understood. We present a simulation study to estimate the error that such missing data causes in commonly-used evaluation metrics in order to assess its prevalence and impact. We find that missing data in the rating or observation process causes the evaluation protocol to systematically mis-estimate metric values, and in some cases erroneously determine that a popularity-based recommender outperforms even a perfect personalized recommender. Substantial breakthroughs in recommendation quality, therefore, will be difficult to assess with existing offline techniques. 
    more » « less
  5. null (Ed.)
    LensKit is an open-source toolkit for building, researching, and learning about recommender systems. First released in 2010 as a Java framework, it has supported diverse published research, small-scale production deployments, and education in both MOOC and traditional classroom settings. In this paper, I present the next generation of the LensKit project, re-envisioning the original tool's objectives as flexible Python package for supporting recommender systems research and development. LensKit for Python (LKPY) enables researchers and students to build robust, flexible, and reproducible experiments that make use of the large and growing PyData and Scientific Python ecosystem, including scikit-learn, and TensorFlow. To that end, it provides classical collaborative filtering implementations, recommender system evaluation metrics, data preparation routines, and tools for efficiently batch running recommendation algorithms, all usable in any combination with each other or with other Python software. This paper describes the design goals, use cases, and capabilities of LKPY, contextualized in a reflection on the successes and failures of the original LensKit for Java software. 
    more » « less