Vivek V. Datla

Adverse Drug Event Detection in Tweets with Semi-Supervised Convolutional Neural Networks

Current Adverse Drug Events (ADE) surveillance systems are often associated with a sizable time lag before such events are published. Online social media such as Twitter could describe adverse drug events in real-time, prior to official reporting. Deep learning has significantly improved text classification performance in recent years and can potentially enhance ADE classification in tweets. However, these models typically require large corpora with human expert-derived labels, and such resources are very expensive to generate and are hardly available. Semi-supervised deep learning models, which offer a plausible alternative to fully supervised models, involve the use of a small set of labeled data and a relatively larger collection of unlabeled data for training. Traditionally, these models are trained on labeled and unlabeled data from similar topics or domains. In reality, millions of tweets generated daily often focus on disparate topics, and this could present a challenge for building deep learning models for ADE classification with random Twitter stream as unlabeled training data. In this work, we build several semi-supervised convolutional neural network (CNN) models for ADE classification in tweets, specifically leveraging different types of unlabeled data in developing the models to address the problem. We demonstrate that, with the selective use of a variety of unlabeled data, our semi-supervised CNN models outperform a strong state-of-the-art supervised classification model by +9.9% F1-score. We evaluated our models on the Twitter data set used in the PSB 2016 Social Media Shared Task. Our results present the new state-of-the-art for this data set.

Attention-Based Medical Caption Generation with Image Modality Classification and Clinical Concept Mapping

This paper proposes an attention-based deep learning framework for caption generation from medical images. We also propose to utilize the same framework for clinical concept prediction to improve caption generation by formulating the task as a case of sequence-to-sequence learning. The predicted concept IDs are then mapped to corresponding terms in a clinical ontology to generate an image caption. We also investigate if learning to classify images based on the modality e.g. CT scan, MRI etc. can aid in generating precise captions.

Studying Military Community Health, Well-Being, and Discourse Through the Social Media Lens

Social media can provide a resource for characterizing communities and targeted populations through activities and content shared online. For instance, studying the armed forces’ use of social media may provide insights into their health and well-being. In this paper, we address three broad research questions: (1) How do military populations use social media? (2) What topics do military users discuss in social media? (3) Do military users talk about health and well-being differently than civilians? Military Twitter users were identified through keywords in the profile description of users who posted geo-tagged tweets at military installations. These military tweets were compared with the tweets from remaining population. Our analysis indicates that military users talk more about military related responsibilities and events, whereas nonmilitary users talk more about school, work, and leisure activities. A significant difference in online content generated by both populations was identified, involving sentiment, health, language, and social media features.