• Publications
  • Influence
All The Cool Kids, How Do They Fit In?: Popularity and Demographic Biases in Recommender Evaluation and Effectiveness
TLDR
System evaluation protocols that explicitly quantify the degree to which the system is meeting the information needs of all its users are proposed, as well as the need for researchers and operators to move beyond evaluations that favor the needs of larger subsets of the user population while ignoring smaller subsets. Expand
Exploring author gender in book rating and recommendation
TLDR
This work measures the distribution of the genders of the authors of books in user rating profiles and recommendation lists produced from this data to find that common collaborative filtering algorithms differ in the gender distribution of their recommendation lists, and in the relationship of that output distribution to user profile distribution. Expand
Estimating Error and Bias in Offline Evaluation Results
TLDR
It is found 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. Expand
Quantifying Error in Recommender System Evaluations
Monte Carlo Estimates of Evaluation Metric Error and Bias
Traditional offline evaluations of recommender systems apply metrics from machine learning and information retrieval in settings where their underlying assumptions no longer hold. This results inExpand
Dynamic Prediction of Building HVAC Energy Consumption by Ensemble Learning Approach
TLDR
This work adopts ensemble learning approaches that consider the load model as a black box and use historical weather data as input and power consumption data as output, to train the HVAC power load model of a building. Expand
Counterfactual Learning to Rank using Heterogeneous Treatment Effect Estimation
TLDR
This work employs heterogeneous treatment effect estimation techniques to estimate position bias when intervention click data is limited and uses such estimations to debias the observed click distribution and re-draw a new de-biased data set, which can be used for any LTR algorithms. Expand