Publications

Supply chain optimization is key to a healthy and profitable business. Many companies use online procurement systems to agree contracts with suppliers. It is vital that the most competitive suppliers are invited to bid for such contracts. In this work, we propose a recommender system to assist with supplier discovery in road freight online procurement. Our system is able to provide personalized supplier recommendations, taking into account customer needs and preferences. This is a novel application of recommender systems, calling for design choices that fit the unique requirements of online procurement. Our preliminary results, using real-world data, are promising.

My Ph.D Thesis.

Uncertainty is a characteristic of every data-driven application, including recommender systems. The quantification of uncertainty can be key to increasing user trust in recommendations or choosing which recommendations should be accompanied by an explanation; uncertainty estimates can be used to accomplish recommender tasks such as active learning and co-training. Many uncertainty estimators are available, but to date, the literature has lacked a comprehensive survey and a detailed comparison. In this article, we fulfill these needs. We review the existing methods for uncertainty estimation and metrics for evaluating uncertainty estimates, while also proposing some estimation methods and evaluation metrics of our own. Using two datasets, we compare the methods using the evaluation metrics that we describe, and we discuss their strengths and potential issues. The goal of this work is to provide a foundation to the field of uncertainty estimation in recommender systems, on which further research can be built.

An important feature of successful supervised machine learning applications is to be able to explain the predictions given by the regression or classification model being used. However, most state-of-the-art models that have good predictive power lead to predictions that are hard to interpret. Thus, several model-agnostic interpreters have been developed recently as a way of explaining black-box classifiers. In practice, using these methods is a slow process because a novel fitting is required for each new testing instance, and several non-trivial choices must be made. We develop NLS (neural local smoother), a method that is complex enough to give good predictions, and yet gives solutions that are easy to be interpreted without the need of using a separate interpreter. The key idea is to use a neural network that imposes a local linear shape to the output layer. We show that NLS leads to predictive power that is comparable to state-of-the-art machine learning models, and yet is easier to interpret.

A Recommender System’s recommendations will each carry a certain level of uncertainty. The quantification of this uncertainty can be useful in a variety of ways. Estimates of uncertainty might be used externally; for example, showing them to the user to increase user trust in the abilities of the system. They may also be used internally; for example, deciding the balance of ‘safe’ and less safe recommendations. In this work, we explore several methods for estimating uncertainty. The novelty comes from proposing methods that work in the implicit feedback setting. We use experiments on two datasets to compare a number of recommendation algorithms that are modified to perform uncertainty estimation. In our experiments, we show that some of these modified algorithms are less accurate than their unmodified counterparts, but others are actually more accurate. We also show which of these methods are best at enabling the recommender to be ‘aware’ of which of its recommendations are likely to be correct and which are likely to be wrong.

Despite its common practice, statistical hypothesis testing presents challenges in interpretation. For instance, in the standard frequentist framework there is no control of the type II error. As a result, the non-rejection of the null hypothesis (H₀) cannot reasonably be interpreted as its acceptance. We propose that this dilemma can be overcome by using agnostic hypothesis tests, since they can control the type I and II errors simultaneously. In order to make this idea operational, we show how to obtain agnostic hypothesis in typical models. For instance, we show how to build (unbiased) uniformly most powerful agnostic tests and how to obtain agnostic tests from standard p-values. Also, we present conditions such that the above tests can be made logically coherent. Finally, we present examples of consistent agnostic hypothesis tests.

Stacking methods improve the prediction performance of regression models. A simple way to stack base regressions estimators is by combining them linearly, as done by Breiman [1]. Even though this approach is useful from an interpretative perspective, it often does not lead to high predictive power. We propose the NN-Stacking method (NNS), which generalizes Breiman’s method by allowing the linear parameters to vary with input features. This improvement enables NNS to take advantage of the fact that distinct base models often perform better at different regions of the feature space. Our method uses neural networks to estimate the stacking coefficients. We show that while our approach keeps the interpretative features of Breiman’s method at a local level, it leads to better predictive power, especially in datasets with large sample sizes.

Although hypothesis tests play a prominent role in Science, their interpretation can be challenging. Three issues are (i) the difficulty in making an assertive decision based on the output of an hypothesis test, (ii) the logical contradictions that occur in multiple hypothesis testing, and (iii) the possible lack of practical importance when rejecting a precise hypothesis. These issues can be addressed through the use of agnostic tests and pragmatic hypotheses.

My master’s dissertation.