BackgroundMedical concepts are inherently ambiguous and error-prone due to human fallibility, which makes it hard for them to be fully used by classical machine learning methods (eg, for tasks like early stage disease prediction).ObjectiveOur work was to create a new machine-friendly representation that resembles the semantics of medical concepts. We then developed a sequential predictive model for medical events based on this new representation.MethodsWe developed novel contextual embedding techniques to combine different medical events (eg, diagnoses, prescriptions, and labs tests). Each medical event is converted into a numerical vector that resembles its “semantics,” via which the similarity between medical events can be easily measured. We developed simple and effective predictive models based on these vectors to predict novel diagnoses.ResultsWe evaluated our sequential prediction model (and standard learning methods) in estimating the risk of potential diseases based on our contextual embedding representation. Our model achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.79 on chronic systolic heart failure and an average AUC of 0.67 (over the 80 most common diagnoses) using the Medical Information Mart for Intensive Care III (MIMIC-III) dataset.ConclusionsWe propose a general early prognosis predictor for 80 different diagnoses. Our method computes numeric representation for each medical event to uncover the potential meaning of those events. Our results demonstrate the efficiency of the proposed method, which will benefit patients and physicians by offering more accurate diagnosis.
Question semantic similarity is a challenging and active research problem that is very useful in many NLP applications, such as detecting duplicate questions in community question answering platforms such as Quora. Arabic is considered to be an under-resourced language, has many dialects, and rich in morphology. Combined together, these challenges make identifying semantically similar questions in Arabic even more difficult. In this paper, we introduce a novel approach to tackle this problem, and test it on two benchmarks; one for Modern Standard Arabic (MSA), and another for the 24 major Arabic dialects. We are able to show that our new system outperforms state-of-the-art approaches by achieving 93% F1-score on the MSA benchmark and 82% on the dialectical one. This is achieved by utilizing contextualized word representations (ELMo embeddings) trained on a text corpus containing MSA and dialectic sentences. This in combination with a pairwise fine-grained similarity layer, helps our question-to-question similarity model to generalize predictions on different dialects while being trained only on question-to-question MSA data.
Arabic dialect identification is an inherently complex problem, as Arabic dialect taxonomy is convoluted and aims to dissect a continuous space rather than a discrete one. In this work, we present machine and deep learning approaches to predict 21 fine-grained dialects form a set of given tweets per user. We adopted numerous feature extraction methods most of which showed improvement in the final model, such as word embedding, Tf-idf, and other tweet features. Our results show that a simple LinearSVC can outperform any complex deep learning model given a set of curated features. With a relatively complex user voting mechanism, we were able to achieve a Macro-Averaged F1-score of 71.84% on MADAR shared subtask-2. Our best submitted model ranked second out of all participating teams.
This paper proposes a novel approach to an automatic estimation of three speaker traits from Arabic speech: gender, emotion, and dialect. After showing promising results on different text classification tasks, the multi-task learning (MTL) approach is used in this paper for Arabic speech classification tasks. The dataset was assembled from six publicly available datasets. First, The datasets were edited and thoroughly divided into train, development, and test sets (open to the public), and a benchmark was set for each task and dataset throughout the paper. Then, three different networks were explored: Long Short Term Memory (LSTM), Convolutional Neural Network (CNN), and Fully-Connected Neural Network (FCNN) on five different types of features: two raw features (MFCC and MEL) and three pre-trained vectors (i-vectors, d-vectors, and x-vectors). LSTM and CNN networks were implemented using raw features: MFCC and MEL, where FCNN was explored on the pre-trained vectors while varying the hyper-parameters of these networks to obtain the best results for each dataset and task. MTL was evaluated against the single task learning (STL) approach for the three tasks and six datasets, in which the MTL and pre-trained vectors almost constantly outperformed STL. All the data and pre-trained models used in this paper are available and can be acquired by the public.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.