PhD , University of Cambridge
What are the difficulties that engineers experience in searching for required information?
I am working on the reuse and management of design information. Design can be viewed as an information-rich process, translating technical requirements into product specifications. Studies show that engineers spend significant time on acquiring information throughout this process. To better understand what difficulties designers encounter in information-seeking activities, studies with professional practitioners were conducted to investigate their level of satisfaction with existing search methods/tools. Issues with information searches were identified in the areas of retrieval systems and human sources. Preliminary results show that engineers need assistance to formulate the right questions to ask when conducting information queries, and b) they wish to have systems that can better understand their information needs.
Abstract: Recommendations are treatments. While todays recommender systems attempt to emulate the naturally occurring user behaviour by predicting either missing entries in the user-item matrix or computing the most likely continuation of user sessions, we need to start thinking of recommendations in terms of optimal interventions with respect to specific goals, such as the increase of number of user conversions on a E-Commerce website. This objective is known as Incremental Treatment Effect prediction (ITE) in the causal community. We propose a new way of factorizing user-item matrices created from a large sample of biased data collected using a control recommendation policy and from limited randomized recommendation data collected using a treatment recommendation policy in order to jointly optimize the prediction of outcomes of the treatment policy and its incremental treatment effect with respect to the control policy. We compare our method against both state-of-the-art factorization methods and against new approaches of causal recommendation and show significant improvements in performance.
Pub.: 23 Jun '17, Pinned: 29 Jun '17
Abstract: Recommendations can greatly benefit from good representations of the user state at recommendation time. Recent approaches that leverage Recurrent Neural Networks (RNNs) for session-based recommendations have shown that Deep Learning models can provide useful user representations for recommendation. However, current RNN modeling approaches summarize the user state by only taking into account the sequence of items that the user has interacted with in the past, without taking into account other essential types of context information such as the associated types of user-item interactions, the time gaps between events and the time of day for each interaction. To address this, we propose a new class of Contextual Recurrent Neural Networks for Recommendation (CRNNs) that can take into account the contextual information both in the input and output layers and modifying the behavior of the RNN by combining the context embedding with the item embedding and more explicitly, in the model dynamics, by parametrizing the hidden unit transitions as a function of context information. We compare our CRNNs approach with RNNs and non-sequential baselines and show good improvements on the next event prediction task.
Pub.: 23 Jun '17, Pinned: 29 Jun '17
Abstract: Ontology is an interdisciplinary field that involves both the use of philosophical principles and the development of computational artifacts. As artifacts, ontologies can have diverse applications in knowledge management, information retrieval, and information systems, to mention a few. They have been largely applied to organize information in complex fields like Biomedicine. In this article, we present the OntoNeo Ontology, an initiative to build a formal ontology in the obstetrics and neonatal domain. OntoNeo is a resource that has been designed to serve as a comprehensive infrastructure providing scientific research and healthcare professionals with access to relevant information. The goal of OntoNeo is twofold: (a) to organize specialized medical knowledge, and (b) to provide a potential consensual representation of the medical information found in electronic health records and medical information systems. To describe our initiative, we first provide background information about distinct theories underlying ontology, top-level computational ontologies and their applications in Biomedicine. Then, we present the methodology employed in the development of OntoNeo and the results obtained to date. Finally, we discuss the applicability of OntoNeo by presenting a proof of concept that illustrates its potential usefulness in the realm of healthcare information systems.
Pub.: 22 Jun '17, Pinned: 29 Jun '17