Named Entity Recognition (NER) in social media posts is challenging since texts are usually short and contexts are lacking. Most recent works show that visual information can boost the NER performance since images can provide complementary contextual information for texts. However, the image-level features ignore the mapping relations between fine-grained visual objects and textual entities, which results in error detection in entities with different types. To better exploit visual and textual information in NER, we propose an adversarial gated bilinear attention neural network (AGBAN). The model jointly extracts entity-related features from both visual objects and texts, and leverages an adversarial training to map two different representations into a shared representation. As a result, domain information contained in an image can be transferred and applied for extracting named entities in the text associated with the image. Experimental results on Tweets dataset demonstrate that our model outperforms the state-of-the-art methods. Moreover, we systematically evaluate the effectiveness of the proposed gated bilinear attention network in capturing the interactions of mutimodal features visual objects and textual words. Our results indicate that the adversarial training can effectively exploit commonalities across heterogeneous data sources, which leads to improved performance in NER when compared to models purely exploiting text data or combining the image-level visual features.

Background: Mobile health apps are increasingly available and used in a clinical context to monitor young people’s mood and mental health. Despite the benefits of accessibility and cost-effectiveness, consumer engagement remains a hurdle for uptake and continued use. Hundreds of mood-monitoring apps are publicly available to young people on app stores; however, few studies have examined consumer perspectives. App store reviews held on Google and Apple platforms provide a large, rich source of naturally generated, publicly available user reviews. Although commercial developers use these data to modify and improve their apps, to date, there has been very little in-depth evaluation of app store user reviews within scientific research, and our current understanding of what makes apps engaging and valuable to young people is limited. Objective: This study aims to gain a better understanding of what app users consider useful to encourage frequent and prolonged use of mood-monitoring apps appropriate for young people. Methods: A systematic approach was applied to the selection of apps and reviews. We identified mood-monitoring apps (n=53) by a combination of automated application programming interface (API) methods. We only included apps appropriate for young people based on app store age categories (apps available to those younger than 18 years). We subsequently downloaded all available user reviews via API data scraping methods and selected a representative subsample of reviews (n=1803) for manual qualitative content analysis. Results: The qualitative content analysis revealed 8 main themes: accessibility (34%), flexibility (21%), recording and representation of mood (18%), user requests (17%), reflecting on mood (16%), technical features (16%), design (13%), and health promotion (11%). A total of 6 minor themes were also identified: notification and reminders; recommendation; privacy, security, and transparency; developer; adverts; and social/community. Conclusions: Users value mood-monitoring apps that can be personalized to their needs, have a simple and intuitive design, and allow accurate representation and review of complex and fluctuating moods. App store reviews are a valuable repository of user engagement feedback and provide a wealth of information about what users value in an app and what user needs are not being met. Users perceive mood-monitoring apps positively, but over 20% of reviews identified the need for improvement.

Distractor generation task aims to generate incorrect options (i.e., distractors) for multiple choice questions from an article.Existing methods for this task often utilize a standard encoder-decoder framework. However, these methods often tend to generate semantically similar distractors, since the same article representations are used to generate different distractors. Multiple generated distractors with similar semantics are considered equivalent. Because the correct answer is unique, students can eliminate these distractors even without reading the article. In this paper, we propose a multi-selector generation network (MSG-Net) that generates distractors with rich semantics based on different sentences in an article. MSG-Net adopts a multi-selector mechanism to select multiple different sentences in an article that are useful to generate diverse distractors. Specifically, a question-aware and answer-aware mechanism are introduced to assist in selecting useful key sentences, where each key sentence is coherent with the question and not equivalent to the answer. MSG-Net can generate diverse distractors based on each selected key sentence with different semantics. Extensive experiments on the RACE dataset and Cosmos QA dataset show that the proposed model outperforms the state-of-the-art models in generating diverse distractors.

Implementation Of A Real-time Psychosis Risk Detection And Alerting System Based On Electronic Health Records Using Cogstack. JoVE publishes peer-reviewed scientific video protocols to accelerate biological, medical, chemical and physical research. Watch our scientific video articles. JoVE is the world-leading producer and provider of science videos with the mission to improve scientific research, scientific journals, and education. Millions of scientists, educators and students at thousands of universities, colleges, hospitals and biopharmaceutical companies worldwide use JoVE for their research, teaching and learning.

Joint multimodal entity-relation extraction (JMERE) is a challenging task that involves two joint subtasks, i.e., named entity recognition and relation extraction, from multimodal data such as text sentences with associated images. Previous JMERE methods have primarily employed 1) pipeline models, which apply pre-trained unimodal models separately and ignore the interaction between tasks, or 2) word-pair relation tagging methods, which neglect neighboring word pairs. To address these limitations, we propose a fine-grained network for JMERE. Specifically, we introduce a fine-grained alignment module that utilizes a phrase-patch to establish connections between text phrases and visual objects. This module can learn consistent multimodal representations from multimodal data. Furthermore, we address the task-irrelevant image information issue by proposing a gate fusion module, which mitigates the impact of image noise and ensures a balanced representation between image objects and text representations. Furthermore, we design a multi-word decoder that enables ensemble prediction of tags for each word pair. This approach leverages the predicted results of neighboring word pairs, improving the ability to extract multi-word entities. Evaluation results from a series of experiments demonstrate the superiority of our proposed model over state-of-the-art models in JMERE.