14 October 2021 in IEEE Access
Countless possibilities of recipe combinations challenge us to determine which additional ingredient goes well with others. In this work, we propose RecipeBowl which is a cooking recommendation system that takes a set of ingredients and cooking tags as input and suggests possible ingredient and recipe choices. We formulate a recipe completion task to train RecipeBowl on our constructed dataset where the model predicts a target ingredient previously eliminated from the original recipe. The RecipeBowl consists of a set encoder and a 2-way decoder for prediction. For the set encoder, we utilize the Set Transformer that builds meaningful set representations. Overall, our model builds a set representation of an leave-one-out recipe and maps it to the ingredient and recipe embedding space. Experimental results demonstrate the effectiveness of our approach. Furthermore, analysis on model predictions and interpretations show interesting insights related to cooking knowledge.
5 January 2018 in JMIR Medical Informatics
Clinical informatics, decision support for health professionals, electronic health records, and ehealth infrastructures.
14 September 2021 in PLOS Computational Biology
A continuing challenge in modern medicine is the identification of safer and more efficacious drugs. Precision therapeutics, which have one molecular target, have been long promised to be safer and more effective than traditional therapies. This approach has proven to be challenging for multiple reasons including lack of efficacy, rapidly acquired drug resistance, and narrow patient eligibility criteria. An alternative approach is the development of drugs that address the overall disease network by targeting multiple biological targets (‘polypharmacology’). Rational development of these molecules will require improved methods for predicting single chemical structures that target multiple drug targets. To address this need, we developed the Multi-Targeting Drug DREAM Challenge, in which we challenged participants to predict single chemical entities that target pro-targets but avoid anti-targets for two unrelated diseases: RET-based tumors and a common form of inherited Tauopathy. Here, we report the results of this DREAM Challenge and the development of two neural network-based machine learning approaches that were applied to the challenge of rational polypharmacology. Together, these platforms provide a potentially useful first step towards developing lead therapeutic compounds that address disease complexity through rational polypharmacology.
20 July 2020 in IEEE Access
A variety of real-world systems can be formulated as bipartite link prediction problems where two different types of nodes exist and no links connect nodes of the same type. In link prediction, triadic closure is an important property that describes how new links are formed. However, triadic closure is difficult to apply to bipartite link prediction tasks because the triadic closure property, which states that new edges tend to form triangles, does not hold true in bipartite settings. In this paper, we introduce Intra-class Connection based Triadic Closure (ICTC) which is a method that can use triadic closure even when the nodes in the same set are unconnected. ICTC aggregates the link probabilities of many local triads, which are edges between triples of nodes, to predict the probability of a link existing between nodes. Specifically, the probability of an edge in a triangle is calculated by multiplying the probabilities of two other edges. The experimental results on eight real-world datasets show that our method outperforms state-of-the-art methods in most cases.
9 January 2019 in IEEE Access
To measure the ability of a machine to understand professional-level scientific articles, we construct a scientific question answering task called PaperQA. The PaperQA task is based on more than 80,000 "fill-in-the-blank" type questions on articles from reputed scientific journals such as Nature and Science. We perform fine-grained linguistic analysis and evaluation to compare PaperQA and other conventional QA(question & answering) tasks on general literature (e.g., books, news articles, Wikipedia texts, and so on). The results indicate that the PaperQA task is the most difficult QA task for both humans (lay people) and machines (deep-learning models). Moreover, humans generally outperform machines in conventional QA tasks, but we found that advanced deep-learning models outperform humans by 3- 13% on average in the PaperQA task. The PaperQA dataset used in our study is publicly available at http://dmis.korea.ac.kr/downloads?id=PaperQA.