Medical coding is the task of assigning medical codes to clinical free-text documentation. Healthcare professionals manually assign such codes to track patient diagnoses and treatments. Automated medical coding can considerably alleviate this administrative burden. In this paper, we reproduce, compare, and analyze state-of-the-art automated medical coding machine learning models. We show that several models underperform due to weak configurations, poorly sampled train-test splits, and insufficient evaluation. In previous work, the macro F1 score has been calculated sub-optimally, and our correction doubles it. We contribute a revised model comparison using stratified sampling and identical experimental setups, including hyperparameters and decision boundary tuning. We analyze prediction errors to validate and falsify assumptions of previous works. The analysis confirms that all models struggle with rare codes, while long documents only have a negligible impact. Finally, we present the first comprehensive results on the newly released MIMIC-IV dataset using the reproduced models. We release our code, model parameters, and new MIMIC-III and MIMIC-IV training and evaluation pipelines to accommodate fair future comparisons.

In this paper we present a formal framework to define and study the properties of utility-oriented measurements of retrieval effectiveness, like AP, RBP, ERR and many other popular IR evaluation measures. The proposed framework is laid in the wake of the representational theory of measurement, which provides the foundations of the modern theory of measurement in both physical and social sciences, thus contributing to explicitly link IR evaluation to a broader context. The proposed framework is minimal, in the sense that it relies on just one axiom, from which other properties are derived. Finally, it contributes to a better understanding and a clear separation of what issues are due to the inherent problems in comparing systems in terms of retrieval effectiveness and what others are due to the expected numerical properties of a measurement.

Fairness is an emerging and challenging topic in recommender systems. In recent years, various ways of evaluating and therefore improving fairness have emerged. In this study, we examine existing evaluation measures of fairness in recommender systems. Specifically, we focus solely on exposure-based fairness measures of individual items that aim at quantifying the disparity in how individual items are recommended to users, separate from item relevance to users. We gather all such measures and we critically analyse their theoretical properties. We identify a series of limitations in each of them, which collectively may render the affected measures hard or impossible to interpret, to compute, or to use for comparing recommendations. We resolve these limitations by redefining or correcting the affected measures, or we argue why certain limitations cannot be resolved. We further perform a comprehensive empirical analysis of both the original and our corrected versions of these fairness measures, using real-world and synthetic datasets. Our analysis provides novel insights into the relationship between measures based on different fairness concepts, and different levels of measure sensitivity and strictness. We conclude with practical suggestions of which fairness measures should be used and when. Our code is publicly available. To our knowledge, this is the first critical comparison of individual item fairness measures in recommender systems.

We propose a family of new evaluation measures, called Markov Precision (MP), which exploits continuous-time and discrete-time Markov chains in order to inject user models into precision. Continuous-time MP behaves like time-calibrated measures, bringing the time spent by the user into the evaluation of a system; discrete-time MP behaves like traditional evaluation measures. Being part of the same Markovian framework, the time-based and rank-based versions of MP produce values that are directly comparable. We show that it is possible to re-create average precision using specific user models and this helps in providing an explanation of Average Precision (AP) in terms of user models more realistic than the ones currently used to justify it. We also propose several alternative models that take into account different possible behaviors in scanning a ranked result list. Finally, we conduct a thorough experimental evaluation of MP on standard TREC collections in order to show that MP is as reliable as other measures and we provide an example of calibration of its time parameters based on click logs from Yandex.

We propose theAssessor-driven Weighted Averages for Retrieval Evaluation (AWARE)probabilistic framework, a novel methodology for dealing with multiple crowd assessors that may be contradictory and/or noisy. By modeling relevance judgements and crowd assessors as sources of uncertainty, AWARE takes the expectation of a generic performance measure, like Average Precision, composed with these random variables. In this way, it approaches the problem of aggregating different crowd assessors from a new perspective, that is, directly combining the performance measures computed on the ground truth generated by the crowd assessors instead of adopting some classification technique to merge the labels produced by them. We propose several unsupervised estimators that instantiate the AWARE framework and we compare them with state-of-the-art approaches, that is,Majoriity Vote and Expectation Maximization, on TREC collections. We found that AWARE approaches improve in terms of their capability of correctly ranking systems and predicting their actual performance scores.