This paper investigates recent research on active learning for (geo) text and image classification, with an emphasis on methods that combine visual analytics and/or deep learning. Deep learning has attracted substantial attention across many domains of science and practice, because it can find intricate patterns in big data; but successful application of the methods requires a big set of labeled data. Active learning, which has the potential to address the data labeling challenge, has already had success in geospatial applications such as trajectory classification from movement data and (geo) text and image classification. This review is intended to be particularly relevant for extension of these methods to GISience, to support work in domains such as geographic information retrieval from text and image repositories, interpretation of spatial language, and related geo-semantics challenges. Specifically, to provide a structure for leveraging recent advances, we group the relevant work into five categories: active learning, visual analytics, active learning with visual analytics, active deep learning, plus GIScience and Remote Sensing (RS) using active learning and active deep learning. Each category is exemplified by recent influential work. Based on this framing and our systematic review of key research, we then discuss some of the main challenges of integrating active learning with visual analytics and deep learning, and point out research opportunities from technical and application perspectives-for application-based opportunities, with emphasis on those that address big data with geospatial components.
When an emergency occurs, citizens can be a helpful support for the operation centers involved in the response activities. As witnesses to a crisis, they initially can share updated and detailed information about what is going on. Moreover, thanks to the current technological evolution people are able to quickly and easily gather rich information and transmit it through different communication channels. Indeed, modern mobile devices embed several sensors such as GPS receivers, Wi-Fi, accelerometers or cameras that can transform users into well-equipped human sensors. For these reasons, emergency organizations and small and medium enterprises have demonstrated a growing interest in developing smart applications for reporting any exceptional circumstances. In this paper, we present a practical study about this kind of applications for identifying both limitations and common features. Based on a study of relevant existent contributions in this area and our personal direct experience in developing and evaluating emergency management solutions, our aim is to propose several findings about how to design effective and efficient mobile emergency notification applications. For this purpose we have exploited the basic sensors of modern mobile devices and the users’ aptitude for using them. The evaluation consists of a practical and a theoretical part. In the practical part, we have simulated a traffic accident as closely as possible to a real scenario, with a victim lying on the ground near a car in the middle of a street. For the theoretical part, we have interviewed some emergency experts for collecting their opinions about the utility of the proposed solution. Results from this evaluation phase confirm the positive impact that EN application have for both operators’ and citizens’ perspective. Moreover, we collected several findings useful for future design challenges in the same area, as shown in the final redesign of the proposed application.
Abstract. Providing alert communication in emergency situations is vital to reduce the number of victims. However, this is a challenging goal for researchers and professionals due to the diverse pool of prospective users, e.g. people with disabilities as well as other vulnerable groups. Moreover, in the event of an emergency situation, many people could become vulnerable because of exceptional circumstances such as stress, an unknown environment or even visual impairment (e.g. fire causing smoke).Within this scope, a crucial activity is to notify affected people about safe places and available evacuation routes. In order to address this need, we propose to extend an ontology, called SEMA4A (Simple EMergency Alert 4 [for] All), developed in a previous work for managing knowledge about accessibility guidelines, emergency situations and communication technologies. In this paper, we introduce a semi-automatic technique for knowledge acquisition and modeling on accessible evacuation routes. We introduce a use case to show applications of the ontology and conclude with an evaluation involving several experts in evacuation procedures.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.