Artificial intelligence (AI) is expected to play a transformational role in health and wellbeing. Search (i.e. information retrieval) technologies already play a significant role in healthcare research and practice. Relevance feedback in Search is vital for system evaluation and improvements. However, in small user scale contexts, the exploitation of user behaviors may not infer valid relevance judgments. Therefore, engaging users to provide such feedback explicitly is essential for improving search performance (i.e. effectiveness). However, previous research has found that users are generally reluctant to provide explicit feedback in digital environments, and the willingness decreases overtime in some experiments. In collaboration with myTomorrows, an Amsterdam-based pharma-tech company, this Master thesis aims to find answers to the challenge mentioned above through a specific context of myTomororws AI-powered treatment Search which has the urgent need for engaging healthcare professionals (HCPs) in providing relevance feedback on search results (e.g. Clinical Trials and Expanded Access Programs) for system evaluation and improvements. Through Human Centered Design methods such as interviews, observations, and speed dates, the project yielded a future myTomorrows Search design enhanced with three relevance feedback collection concepts. As research materials, the concepts were tested and evaluated by nine HCPs from three countries (the Netherlands, China, and Brazil). The user study results indicate that embedding utility, as the motivator, in relevance feedback collection appeals to HCPs more than using motivators such as altruism or enjoyment. Moreover, the best point of user engagement is identified as the moment between users finishing the examination of information and starting the next ones. Additionally, this study generalized the project process and user study insights into a four-stage guide for designing explicit feedback collection in text-base Search. Although it remains unvalidated, this guide has the potential to apply to other small user scale contexts, guiding or inspiring user researchers and designers to design explicit user feedback collection in Search.

Machine Learning models are increasingly used to assist or replace humans in a variety of decision-making domains. However, a lot of concerns have been raised about the impact of these decisions on people’s lives. In this work we focus on two main problems. The first one is that there might be discrimination between different groups of people with respect to their protected attributes in the aforementioned Machine Learning decision making applications. The second one is that there is lack of methods that actually interpret and explain the predictions of these Machine Learning systems which are then used to help decision making. Particularly, we focus on a specific aspect of these decision making applications, namely in Machine Learning training data for the problem of classification. Most research tackles solely individual aspects like trying to detect and mitigate bias and does not pay any attention to explain their reasoning in a human interpretable way. Also, the main method that they use is to balance the distribution of the training data with respect to the protected attribute. However, as we show, this is not always the solution to the problem. On the contrary, we study concurrently three steps (detection, semantic interpretation and mitigation of bias) to overcome these shortcomings and limitations and show that there is the presence of specific visual clues that leads to that bias. Finally, we perform an extensive evaluation of our method in order to verify its efficiency and effectiveness on the particular use-case Gender Bias in Profession Prediction from Images.

Training data for segmentation tasks are often available only on a small scale. Transferring learned representations from pre-trained classification models is therefore widely adopted by convolutional neural networks for semantic segmentation. In domains where the representations from the classification models are not directly applicable, we propose to train representations with segmentation datasets that potentially contains label errors. Our experiments demonstrate that label errors, such as mislabeled segments and missing segmentations, have negative influences to the learned representations. To alleviate the negative effects of object mislabelling, we propose to discard the object labels and instead train foreground/background segmentation. The learned representations with binary segmentation achieve a fine-tuning performance comparable to the representations learned with ``gold'' standard segmentations. In the existence of missing segmentations, a sigmoid loss for the background class is proposed to achieve high recall while keeping the precision better than simply weighting the classes. The proposed class dependent, sigmoid loss obtains better segmentation performance as well as better representations than the weighting the classes in the presence of missing segmentations. To summerize, we propose to learn representations with foreground/background segmentation and with a sigmoid loss for the background class when there exist missing segmentations for objects.