Tweeting about twenty: an analysis of interest, public sentiments and opinion about 20mph speed restrictions in two UK cities

Background

Twenty miles per hour (20mph) speed limits (equivalent to roughly 30kmh) have become part of public health policies to reduce urban road collisions and casualties, especially in Western countries. Public opinion plays a crucial role in opposition to and acceptance of policies that are advocated for improving public health. Twenty miles per hour speed limit policies were implemented in Edinburgh and Belfast from 2016 to 2018. In this paper, we extract public opinion and sentiments expressed about the new 20mph speed limits in those cities using publicly available Twitter data.

Methods

We analysed public sentiments from Twitter data and classified the public comments in plain English into the categories ‘positive’, ‘neutral’, and ‘negative’. We also explored the frequency and sources of the tweets.

Results

The total volume of tweets was higher for Edinburgh than for Belfast, but the volume of tweets followed a similar pattern, peaking around 2016, which is when the schemes were implemented. Overall, the tone of the tweets was positive or neutral towards the implementation of the speed limit policies. This finding was surprising as there is a perception among policymakers that there would have been public backlash against these sorts of policy changes. The commonly used hashtags focused largely on road safety and other potential benefits, for example to air pollution.

Conclusions

Overall, public attitudes towards the policies were positive, thus policymakers should be less anxious about potential public backlash when considering the scale-up of 20mph speed restrictions.

Understanding Concerns, Sentiments, and Disparities Among Population Groups During the COVID-19 Pandemic Via Twitter Data Mining: Large-scale Cross-sectional Study

Background

Since the beginning of the COVID-19 pandemic in late 2019, its far-reaching impacts have been witnessed globally across all aspects of human life, such as health, economy, politics, and education. Such widely penetrating impacts cast significant and profound burdens on all population groups, incurring varied concerns and sentiments among them.

Objective

This study aims to identify the concerns, sentiments, and disparities of various population groups during the COVID-19 pandemic through a cross-sectional study conducted via large-scale Twitter data mining infoveillance.

Methods

This study consisted of three steps: first, tweets posted during the pandemic were collected and preprocessed on a large scale; second, the key population attributes, concerns, sentiments, and emotions were extracted via a collection of natural language processing procedures; third, multiple analyses were conducted to reveal concerns, sentiments, and disparities among population groups during the pandemic. Overall, this study implemented a quick, effective, and economical approach for analyzing population-level disparities during a public health event. The source code developed in this study was released for free public use at GitHub.

Results

A total of 1,015,655 original English tweets posted from August 7 to 12, 2020, were acquired and analyzed to obtain the following results. Organizations were significantly more concerned about COVID-19 (odds ratio [OR] 3.48, 95% CI 3.39-3.58) and expressed more fear and depression emotions than individuals. Females were less concerned about COVID-19 (OR 0.73, 95% CI 0.71-0.75) and expressed less fear and depression emotions than males. Among all age groups (ie, ≤18, 19-29, 30-39, and ≥40 years of age), the attention ORs of COVID-19 fear and depression increased significantly with age. It is worth noting that not all females paid less attention to COVID-19 than males. In the age group of 40 years or older, females were more concerned than males, especially regarding the economic and education topics. In addition, males 40 years or older and 18 years or younger were the least positive. Lastly, in all sentiment analyses, the sentiment polarities regarding political topics were always the lowest among the five topics of concern across all population groups.

Conclusions

Through large-scale Twitter data mining, this study revealed that meaningful differences regarding concerns and sentiments about COVID-19-related topics existed among population groups during the study period. Therefore, specialized and varied attention and support are needed for different population groups. In addition, the efficient analysis method implemented by our publicly released code can be utilized to dynamically track the evolution of each population group during the pandemic or any other major event for better informed public health research and interventions.

Deep learning for pollen allergy surveillance from twitter in Australia

BACKGROUND

The paper introduces a deep learning-based approach for real-time detection and insights generation about one of the most prevalent chronic conditions in Australia - Pollen allergy. The popular social media platform is used for data collection as cost-effective and unobtrusive alternative for public health monitoring to complement the traditional survey-based approaches.

METHODS

The data was extracted from Twitter based on pre-defined keywords (i.e. 'hayfever' OR 'hay fever') throughout the period of 6 months, covering the high pollen season in Australia. The following deep learning architectures were adopted in the experiments: CNN, RNN, LSTM and GRU. Both default (GloVe) and domain-specific (HF) word embeddings were used in training the classifiers. Standard evaluation metrics (i.e. Accuracy, Precision and Recall) were calculated for the results validation. Finally, visual correlation with weather variables was performed.

RESULTS

The neural networks-based approach was able to correctly identify the implicit mentions of the symptoms and treatments, even unseen previously (accuracy up to 87.9% for GRU with GloVe embeddings of 300 dimensions).

CONCLUSIONS

The system addresses the shortcomings of the conventional machine learning techniques with manual feature-engineering that prove limiting when exposed to a wide range of non-standard expressions relating to medical concepts. The case-study presented demonstrates an application of 'black-box' approach to the real-world problem, along with its internal workings demonstration towards more transparent, interpretable and reproducible decision-making in health informatics domain.

Mining Social Media to Identify Heat Waves

Heat waves are one of the deadliest of natural hazards and their frequency and intensity will likely increase as the climate continues to warm. A challenge in studying these phenomena is the lack of a universally accepted quantitative definition that captures both temperature anomalies and associated mortality. We test the hypothesis that social media mining can be used to identify heat wave mortality. Applying the approach to India, we find that the number of heat-related tweets correlates with heat-related mortality much better than traditional climate-based indicators, especially at larger scales, which identify many heat wave days that do not lead to excess mortality. We conclude that social media based heat wave identification can complement climatic data and can be used to: (1) study heat wave impacts at large scales or in developing countries, where mortality data are difficult to obtain and uncertain, and (2) to track dangerous heat wave events in real time.

Twitter sentiment classification for measuring public health concerns

An important task of public health officials is to keep track of health issues, such as spreading epidemics. In this paper, we are addressing the issue of spreading public concern about epidemics. Public concern about a communicable disease can be seen as a problem of its own. Keeping track of trends in concern about public health and identifying peaks of public concern are therefore crucial tasks. However, monitoring public health concerns is not only expensive with traditional surveillance systems, but also suffers from limited coverage and significant delays. To address these problems, we are using Twitter messages, which are available free of cost, are generated world-wide, and are posted in real time. We are measuring public concern using a two-step sentiment classification approach. In the first step, we distinguish Personal tweets from News (i.e., Non-Personal) tweets. In the second step, we further separate Personal Negative from Personal Non-Negative tweets. Both these steps consist themselves of two sub-steps. In the first sub-step (of both steps), our programs automatically generate training data using an emotion-oriented, clue-based method. In the second sub-step, we are training and testing three different Machine Learning (ML) models with the training data from the first sub-step; this allows us to determine the best ML model for different datasets. Furthermore, we are testing the already trained ML models with a human annotated, disjoint dataset. Based on the number of tweets classified as Personal Negative, we compute a Measure of Concern (MOC) and a timeline of the MOC. We attempt to correlate peaks of the MOC timeline to peaks of the News (Non-Personal) timeline. Our best accuracy results are achieved using the two-step method with a Naïve Bayes classifier for the Epidemic domain (six datasets) and the Mental Health domain (three datasets).

Twitter mining for fine-grained syndromic surveillance

BACKGROUND

Digital traces left on the Internet by web users, if properly aggregated and analyzed, can represent a huge information dataset able to inform syndromic surveillance systems in real time with data collected directly from individuals. Since people use everyday language rather than medical jargon (e.g. runny nose vs. respiratory distress), knowledge of patients' terminology is essential for the mining of health related conversations on social networks.

OBJECTIVES

In this paper we present a methodology for early detection and analysis of epidemics based on mining Twitter messages. In order to reliably trace messages of patients that actually complain of a disease, first, we learn a model of naïve medical language, second, we adopt a symptom-driven, rather than disease-driven, keyword analysis. This approach represents a major innovation compared to previous published work in the field.

METHOD

We first developed an algorithm to automatically learn a variety of expressions that people use to describe their health conditions, thus improving our ability to detect health-related "concepts" expressed in non-medical terms and, in the end, producing a larger body of evidence. We then implemented a Twitter monitoring instrument to finely analyze the presence and combinations of symptoms in tweets.

RESULTS

We first evaluate the algorithm's performance on an available dataset of diverse medical condition synonyms, then, we assess its utility in a case study of five common syndromes for surveillance purposes. We show that, by exploiting physicians' knowledge on symptoms positively or negatively related to a given disease, as well as the correspondence between patients' "naïve" terminology and medical jargon, not only can we analyze large volumes of Twitter messages related to that disease, but we can also mine micro-blogs with complex queries, performing fine-grained tweets classification (e.g. those reporting influenza-like illness (ILI) symptoms vs. common cold or allergy).

CONCLUSIONS

Our approach yields a very high level of correlation with flu trends derived from traditional surveillance systems. Compared with Google Flu, another popular tool based on query search volumes, our method is more flexible and less sensitive to changes in web search behaviors.

Towards Large-scale Twitter Mining for Drug-related Adverse Events

Drug-related adverse events pose substantial risks to patients who consume post-market or Drug-related adverse events pose substantial risks to patients who consume post-market or investigational drugs. Early detection of adverse events benefits not only the drug regulators, but also the manufacturers for pharmacovigilance. Existing methods rely on patients' "spontaneous" self-reports that attest problems. The increasing popularity of social media platforms like the Twitter presents us a new information source for finding potential adverse events. Given the high frequency of user updates, mining Twitter messages can lead us to real-time pharmacovigilance. In this paper, we describe an approach to find drug users and potential adverse events by analyzing the content of twitter messages utilizing Natural Language Processing (NLP) and to build Support Vector Machine (SVM) classifiers. Due to the size nature of the dataset (i.e., 2 billion Tweets), the experiments were conducted on a High Performance Computing (HPC) platform using MapReduce, which exhibits the trend of big data analytics. The results suggest that daily-life social networking data could help early detection of important patient safety issues.