Neural attention with character embeddings for hay fever detection from twitter

Abstract

The paper aims to leverage the highly unstructured user-generated content in the context of pollen allergy surveillance using neural networks with character embeddings and the attention mechanism. Currently, there is no accurate representation of hay fever prevalence, particularly in real-time scenarios. Social media serves as an alternative to extract knowledge about the condition, which is valuable for allergy sufferers, general practitioners, and policy makers. Despite tremendous potential offered, conventional natural language processing methods prove limited when exposed to the challenging nature of user-generated content. As a result, the detection of actual hay fever instances among the number of false positives, as well as the correct identification of non-technical expressions as pollen allergy symptoms poses a major problem. We propose a deep architecture enhanced with character embeddings and neural attention to improve the performance of hay fever-related content classification from Twitter data. Improvement in prediction is achieved due to the character-level semantics introduced, which effectively addresses the out-of-vocabulary problem in our dataset where the rate is approximately 9%. Overall, the study is a step forward towards improved real-time pollen allergy surveillance from social media with state-of-art technology.

Translational health technology and system schemes: enhancing the dynamics of health informatics

“…Neural networks model combined with data and local social media is utilized successfully in the classification of allergy symptoms. [ 53 ] In the field of dermatology AI has the potential to assist in the diagnostics and at the interface between primary and secondary care [ 54 ]. AI applications serve in clinical practice also in pathology and by robot-assisted surgery, in precision robotic treatment, and in virtual reality-enabled robotics [ 42 , 45 ].…”

FEECA: Fuzzy based Energy Efficient Clustering Approach in Wireless Sensor Network

“…The main dissimilarity between Gupta protocol [7] and LEACH is in the setup phase where the sink wants to collect energy and position information for every SN and calculates the chance value of a SN with the help of fuzzy interference system (FIS) to become a CH or not, the sink selects the SN that has the greatest possibility of becoming a CH on the basis of their chance values. Gupta protocol [7] does not consider the farness between sink and SN.CH selection has many ambiguities as the number of selection parameters increases day by day in order to minimize the ambiguity in clustering process. To overcome this problem, we introduce fuzzy based clustering.…”

“…WSN can be deployed in various areas like battlefield, military surveillance, home monitoring, health monitoring, agriculture monitoring, and vehicle tracing etc. [4]- [7]. In some of the applications, multiple sensors are deployed and arranged intelligently to capture sensitive information which is to be forwarded to the sink but in some applications, it is very difficult to deploy sensors deterministically [8] [9].…”

Investigation of Social Behaviour Patterns using Location-Based Data – A Melbourne Case Study

“…Even though such data has been used extensively in the past to study urban functions, urban planning, and geographical information systems, it has limitations in regard to the time and effort required to collect, process and analyze it. On the other hand, data generated by a 'bottom-up approach' offers a better alternative and includes usercreated data in the form of online blogs, check-ins, and social media posts [36][37][38]. Researchers are not required to create a questionnaire or a survey to collect data in this form, which is usually a very rich source of information [11].…”