Leveraging Transfer Learning to Analyze Opinions, Attitudes, and Behavioral Intentions Toward COVID-19 Vaccines: Social Media Content and Temporal Analysis (Preprint)

BACKGROUND The COVID-19 vaccine is considered to be the most promising approach to alleviate the pandemic. However, in recent surveys, acceptance of the COVID-19 vaccine has been low. To design more effective outreach interventions, there is an urgent need to understand public perceptions of COVID-19 vaccines. OBJECTIVE Our objective was to analyze the potential of leveraging transfer learning to detect tweets containing opinions, attitudes, and behavioral intentions toward COVID-19 vaccines, and to explore temporal trends as well as automatically extract topics across a large number of tweets. METHODS We developed machine learning and transfer learning models to classify tweets, followed by temporal analysis and topic modeling on a dataset of COVID-19 vaccine–related tweets posted from November 1, 2020 to January 31, 2021. We used the F1 values as the primary outcome to compare the performance of machine learning and transfer learning models. The statistical values and P values from the Augmented Dickey-Fuller test were used to assess whether users’ perceptions changed over time. The main topics in tweets were extracted by latent Dirichlet allocation analysis. RESULTS We collected 2,678,372 tweets related to COVID-19 vaccines from 841,978 unique users and annotated 5000 tweets. The F1 values of transfer learning models were 0.792 (95% CI 0.789-0.795), 0.578 (95% CI 0.572-0.584), and 0.614 (95% CI 0.606-0.622) for these three tasks, which significantly outperformed the machine learning models (logistic regression, random forest, and support vector machine). The prevalence of tweets containing attitudes and behavioral intentions varied significantly over time. Specifically, tweets containing positive behavioral intentions increased significantly in December 2020. In addition, we selected tweets in the following categories: positive attitudes, negative attitudes, positive behavioral intentions, and negative behavioral intentions. We then identified 10 main topics and relevant terms for each category. CONCLUSIONS Overall, we provided a method to automatically analyze the public understanding of COVID-19 vaccines from real-time data in social media, which can be used to tailor educational programs and other interventions to effectively promote the public acceptance of COVID-19 vaccines.

Download Full-text

Microblog credibility indicators regarding misinformation of genetically modified food on Weibo

PLoS ONE ◽

10.1371/journal.pone.0252392 ◽

2021 ◽

Vol 16 (6) ◽

pp. e0252392

Author(s):

Jiaojiao Ji ◽

Naipeng Chao ◽

Shitong Wei ◽

George A. Barnett

Keyword(s):

Machine Learning ◽

Social Media ◽

Science Communication ◽

Predictive Power ◽

Genetically Modified ◽

Public Understanding ◽

Supervised Machine Learning ◽

Genetically Modified Food ◽

The Public ◽

Gm Food

The considerable amount of misinformation on social media regarding genetically modified (GM) food will not only hinder public understanding but also mislead the public to make unreasoned decisions. This study discovered a new mechanism of misinformation diffusion in the case of GM food and applied a framework of supervised machine learning to identify effective credibility indicators for the misinformation prediction of GM food. Main indicators are proposed, including user identities involved in spreading information, linguistic styles, and propagation dynamics. Results show that linguistic styles, including sentiment and topics, have the dominant predictive power. In addition, among the user identities, engagement, and extroversion are effective predictors, while reputation has almost no predictive power in this study. Finally, we provide strategies that readers should be aware of when assessing the credibility of online posts and suggest improvements that Weibo can use to avoid rumormongering and enhance the science communication of GM food.

Download Full-text

Analisis Sentimen Wacana Pemindahan Ibu Kota Indonesia Menggunakan Algoritma Support Vector Machine (SVM)

Jurnal Teknologi Informasi dan Ilmu Komputer ◽

10.25126/jtiik.0813944 ◽

2021 ◽

Vol 8 (1) ◽

pp. 147

Author(s):

Primandani Arsi ◽

Retno Waluyo

Keyword(s):

Machine Learning ◽

Social Media ◽

Support Vector Machine ◽

Sentiment Analysis ◽

Public Discourse ◽

Capital City ◽

Support Vector ◽

The Public ◽

Machine Learning Methods ◽

Classification Technique

Dewasa ini, media sosial berkembang pesat di internet, salah satu yang banyak digemari adalah Twitter. Berbagai topik ramai diperbincangkan di Twitter mulai dari ekonomi, politik, sosial, budaya, hukum dan lain-lain. Salah satu topik yang ramai diperbincangkan di Twitter adalah terkait isu pemindahan ibu kota Indonesia. Namun dibalik hal tersebut terdapat kontroversi dari pihak yang merasa pro dan kontra, masing-masing memiiki sudut pandang yang berbeda. Hal ini menyebabkan munculnya fenomena perdebatan khususnya di Twitter yang sebenarnya menunjukkan perhatian kolektif mengenai wacana publik tersebut. Analisis sentimen adalah proses mengekstraksi, memahami dan mengolah data berupa teks yang tidak terstruktur secara otomatis guna mendapatkan informasi sentimen yang terdapat pada sebuah kalimat pendapat atau opini. Dalam penerapan analisis sentimen menggunakan metode machine learning terdapat beberapa metode yang sering digunakan. Dalam penelitian ini diusulkan metode Support Vector Machine (SVM) untuk diterapkan pada tweets topik pemindahan ibu kota Indonesia untuk tujuan klasifikasi kelas sentimen pada media sosial twitter. Teknis klasifikasi dilakukan dengan cara mengklasifikasikan menjadi 2 kelas yakni positif dan negatif. Berdasarkan hasil pengujian yang dilakukan terhadap tweets sentimen pemindahan ibu kota dari media sosial twitter sebanyak 1.236 tweets (404 positif dan 832 negatif) menggunakan SVM diperoleh akurasi =96,68%, precision=95.82%, recall=94.04% dan AUC = 0,979. AbstractToday, social media is growing fast on the internet.One of the most popular social media is Twitter. Many topics are discussed on Twitter such as economic, politic, social, culture, and law. One of the hot topics discussed on Twitter is the issue of relocating Indonesia's capital city. However, there is controversy from supporters and opponents. They have different views. This issue leads to a phenomenon of debate on Twitter that actually shows a collective concern about the public discourse. Sentiment analysis is a process of extracting, understanding and processing unstructured data to get sentiment information which is found in an opinion sentence. Application of sentiment analysis using machine learning methods shows that there are several methods that are often used. In this study, the Support Vector Machine (SVM) method is proposed to be applied to tweets on the topic of relocating Indonesia's capital city for sentiment classification on social media twitter. The classification technique is carried out into 2 classes, namely positive and negative. Based on testing on the sentiment of relocating Indonesia's capital city from social media twitter from 1,116 tweets (404 positive and 832 negative) using SVM obtained accuracy = 96.68%, precision = 95.82%, recall = 94.04% and AUC = 0.979.

Download Full-text

Comparison of Deep Transfer Learning Techniques in Human Skin Burns Discrimination

Applied System Innovation ◽

10.3390/asi3020020 ◽

2020 ◽

Vol 3 (2) ◽

pp. 20 ◽

Cited By ~ 3

Author(s):

Aliyu Abubakar ◽

Mohammed Ajuji ◽

Ibrahim Usman Yahya

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Transfer Learning ◽

Fine Tuning ◽

Machine Learning Techniques ◽

Support Vector ◽

Learning Models ◽

Skin Injuries ◽

Learning Techniques ◽

Injured Skin

While visual assessment is the standard technique for burn evaluation, computer-aided diagnosis is increasingly sought due to high number of incidences globally. Patients are increasingly facing challenges which are not limited to shortage of experienced clinicians, lack of accessibility to healthcare facilities and high diagnostic cost. Certain number of studies were proposed in discriminating burn and healthy skin using machine learning leaving a huge and important gap unaddressed; whether burns and related skin injuries can be effectively discriminated using machine learning techniques. Therefore, we specifically use transfer learning by leveraging pre-trained deep learning models due to deficient dataset in this paper, to discriminate two classes of skin injuries—burnt skin and injured skin. Experiments were extensively conducted using three state-of-the-art pre-trained deep learning models that includes ResNet50, ResNet101 and ResNet152 for image patterns extraction via two transfer learning strategies—fine-tuning approach where dense and classification layers were modified and trained with features extracted by base layers and in the second approach support vector machine (SVM) was used to replace top-layers of the pre-trained models, trained using off-the-shelf features from the base layers. Our proposed approach records near perfect classification accuracy in categorizing burnt skin ad injured skin of approximately 99.9%.

Download Full-text

Abstract 472: Organizations Unite in “Call-Push-Shock” Movement to Impact Public Understanding That Drives Action

Circulation ◽

10.1161/circ.140.suppl_2.472 ◽

2019 ◽

Vol 140 (Suppl_2) ◽

Author(s):

Mary M Newman ◽

Martha E Lopez-Anderson ◽

Jennifer F Chap ◽

Carissa B Caramanis ◽

Maureen Legg ◽

...

Keyword(s):

Social Media ◽

Public Understanding ◽

Social Media Analytics ◽

Media Campaign ◽

Loved One ◽

The Public ◽

Bystander Cpr ◽

User Friendly ◽

The Impact ◽

Over Time

Introduction: The incidence of OHCA in the U.S. is high (>350,000); the survival rate is low, about 1 in 10. Less than half of victims receive bystander CPR (39%) or treatment with an AED (6%) before EMS arrival. Consumer research indicates the public’s motivation to learn CPR/AED skills and help in emergencies increases with understanding of CA and the impact of bystander action. Two tested messages resonate most with the public: CPR/AED use can double or triple the chance of survival, and CA can happen to a loved one. Hypothesis: An inclusive national movement of organizations using unified, co-branded and tested messaging in a coordinated, multi-channel social media campaign could improve targeted reach and, over time, public understanding of CA and the importance of bystander action. Creating a gold standard for educational content on CA could improve message retention and motivation to act over time. Methods: We created a movement through consistent messaging in a user-friendly toolkit. We built a landing page (CallPushShock.org) and co-sponsored a social media campaign on CA and the importance of bystander action. We launched the campaign during CPR-AED Awareness Week (June 2018). We invited other organizations to join for SCA Awareness Month (Oct 2018), providing co-branded social media assets (Facebook/Twitter posts, videos, infographic) and posting schedules to all partners. Social media analytics were used to measure the reach of the campaign. Results: The campaign launched in June, attracting 20 partners by October. Combined co-sponsor results for June were: News release: 548,464 headline impressions (PR Web); hashtag (#callpushshock) impressions: 243,345 (Keyhole); FB reach: 86,134; Twitter impressions: 25,107; YouTube views: 351; e-news opens: 4,226 (MailChimp, Constant Contact); landing page: 1,291 users (Google). [ Oct 2018 and June 2019 results in development, reflecting extended partner reach.] Conclusion: A social media campaign, leveraging consumer-tested messaging used by multiple organizations in a unified, consistent movement can be effective in improving targeted reach and educational outcomes. A sustained effort is needed to determine campaign impact in improving public understanding of CA and bystander action over time.

Download Full-text

Comparison of Deep Transfer Learning Techniques in Human Skin Burns Discrimination

10.20944/preprints202003.0204.v1 ◽

2020 ◽

Author(s):

Aliyu Abubakar ◽

Mohammed Ajuji ◽

Ibrahim Usman Yahya

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Learning Strategies ◽

Transfer Learning ◽

Standard Technique ◽

Fine Tuning ◽

Machine Learning Techniques ◽

Support Vector ◽

Learning Models ◽

Learning Techniques

While visual assessment is the standard technique for burn evaluation, computer-aided diagnosis is increasingly sought due to high number of incidences globally. Patients are increasingly facing challenges which are not limited to shortage of experienced clinicians, lack of accessibility to healthcare facilities, and high diagnostic cost. Certain number of studies were proposed in discriminating burn and healthy skin using machine learning leaving a huge and important gap unaddressed; whether burns and related skin injuries can be effectively discriminated using machine learning techniques. Therefore, we specifically use pre-trained deep learning models due to deficient dataset to train a new model from scratch. Experiments were extensively conducted using three state-of-the-art pre-trained deep learning models that includes ResNet50, ResNet101 and ResNet152 for image patterns extraction via two transfer learning strategies: fine-tuning approach where dense and classification layers were modified and trained with features extracted by base layers, and in the second approach support vector machine (SVM) was used to replace top-layers of the pre-trained models, trained using off-the-shelf features from the base layers. Our proposed approach records near perfect classification accuracy of approximately 99.9%.

Download Full-text

Predicting Depression Using Social Media Posts

Journal of Soft Computing and Data Mining ◽

10.30880/jscdm.2021.02.02.004 ◽

2021 ◽

Vol 2 (2) ◽

Author(s):

Fahem Abu Bakar ◽

◽

Nazri Mohd Nawi ◽

Abdulkareem A. Hezam ◽

◽

...

Keyword(s):

Neural Network ◽

Mental Health ◽

Machine Learning ◽

Social Media ◽

Support Vector Machine ◽

Neural Network Model ◽

Support Vector ◽

Learning Models ◽

The Neural Network ◽

Machine Learning Models

The use of Social Network Sites (SNS) is on the rise these days, particularly among the younger generations. Users can communicate their interests, feelings, and everyday routines thanks to the availability of social media sites. Many studies show that properly utilizing user-generated content (UGC) can aid in determining people's mental health status. The use of the UGC could aid in the prediction of mental health, particularly depression, where it is a significant medical condition that impairs one's ability to work, learn, eat, sleep, and enjoy life. However, all information about a person's mood and negativism can be gathered from their SNS user profile. Therefore, this study utilizes SNS as a data source by using machine learning models to screen and identify users in categorizing users based on their mental health. The performance of three machine learning models is evaluated to classify the UGC: Decision Forest, Neural Network, and Support Vector Machine (SVM). The results show that the accuracy and recall result of the Neural Network model is the same as the Support Vector Machine (SVM) model, which is 78.27% and 0.042, but Neural Network performs better in the average precision value. This proves that the Neural Network model is the best model for making predictions to determine the level of depression by using social media posts.

Download Full-text

Utilizing Blockchain Technology in Social Media Bot Identification

10.36227/techrxiv.12049374 ◽

2020 ◽

Author(s):

Shreya Reddy ◽

Lisa Ewen ◽

Pankti Patel ◽

Prerak Patel ◽

Ankit Kundal ◽

...

Keyword(s):

Machine Learning ◽

Social Media ◽

Gold Standard ◽

The Internet ◽

Learning Models ◽

Current Time ◽

Machine Learning Methods ◽

Blockchain Technology ◽

Modern Age ◽

Machine Learning Models

As bots become more prevalent and smarter in the modern age of the internet, it becomes ever more important that they be identified and removed. Recent research has dictated that machine learning methods are accurate and the gold standard of bot identification on social media. Unfortunately, machine learning models do not come without their negative aspects such as lengthy training times, difficult feature selection, and overwhelming pre-processing tasks. To overcome these difficulties, we are proposing a blockchain framework for bot identification. At the current time, it is unknown how this method will perform, but it serves to prove the existence of an overwhelming gap of research under this area.

Download Full-text

Monitoring the Foliar Nutrients Status of Mango Using Spectroscopy-Based Spectral Indices and PLSR-Combined Machine Learning Models

Remote Sensing ◽

10.3390/rs13040641 ◽

2021 ◽

Vol 13 (4) ◽

pp. 641

Author(s):

Gopal Ramdas Mahajan ◽

Bappa Das ◽

Dayesh Murgaokar ◽

Ittai Herrmann ◽

Katja Berger ◽

...

Keyword(s):

Machine Learning ◽

Remote Sensing ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Support Vector ◽

Spectral Indices ◽

Learning Models ◽

Leaf Nutrients ◽

Machine Learning Models

Conventional methods of plant nutrient estimation for nutrient management need a huge number of leaf or tissue samples and extensive chemical analysis, which is time-consuming and expensive. Remote sensing is a viable tool to estimate the plant’s nutritional status to determine the appropriate amounts of fertilizer inputs. The aim of the study was to use remote sensing to characterize the foliar nutrient status of mango through the development of spectral indices, multivariate analysis, chemometrics, and machine learning modeling of the spectral data. A spectral database within the 350–1050 nm wavelength range of the leaf samples and leaf nutrients were analyzed for the development of spectral indices and multivariate model development. The normalized difference and ratio spectral indices and multivariate models–partial least square regression (PLSR), principal component regression, and support vector regression (SVR) were ineffective in predicting any of the leaf nutrients. An approach of using PLSR-combined machine learning models was found to be the best to predict most of the nutrients. Based on the independent validation performance and summed ranks, the best performing models were cubist (R2 ≥ 0.91, the ratio of performance to deviation (RPD) ≥ 3.3, and the ratio of performance to interquartile distance (RPIQ) ≥ 3.71) for nitrogen, phosphorus, potassium, and zinc, SVR (R2 ≥ 0.88, RPD ≥ 2.73, RPIQ ≥ 3.31) for calcium, iron, copper, boron, and elastic net (R2 ≥ 0.95, RPD ≥ 4.47, RPIQ ≥ 6.11) for magnesium and sulfur. The results of the study revealed the potential of using hyperspectral remote sensing data for non-destructive estimation of mango leaf macro- and micro-nutrients. The developed approach is suggested to be employed within operational retrieval workflows for precision management of mango orchard nutrients.

Download Full-text

Transfer Learning to CCSD(T): Accurate Anharmonic Frequencies from Machine Learning Models

Journal of Chemical Theory and Computation ◽

10.1021/acs.jctc.1c00249 ◽

2021 ◽

Author(s):

Silvan Käser ◽

Eric D. Boittier ◽

Meenu Upadhyay ◽

Markus Meuwly

Keyword(s):

Machine Learning ◽

Transfer Learning ◽

Learning Models ◽

Anharmonic Frequencies ◽

Machine Learning Models

Download Full-text

Machine learning models to identify low adherence to influenza vaccination among Korean adults with cardiovascular disease

BMC Cardiovascular Disorders ◽

10.1186/s12872-021-01925-7 ◽

2021 ◽

Vol 21 (1) ◽

Author(s):

Moojung Kim ◽

Young Jae Kim ◽

Sung Jin Park ◽

Kwang Gi Kim ◽

Pyung Chun Oh ◽

...

Keyword(s):

Machine Learning ◽

Cardiovascular Disease ◽

Influenza Vaccination ◽

Machine Learning Techniques ◽

Gradient Boosting ◽

Support Vector ◽

Age Group ◽

Learning Models ◽

Extreme Gradient Boosting ◽

Machine Learning Models

Abstract Background Annual influenza vaccination is an important public health measure to prevent influenza infections and is strongly recommended for cardiovascular disease (CVD) patients, especially in the current coronavirus disease 2019 (COVID-19) pandemic. The aim of this study is to develop a machine learning model to identify Korean adult CVD patients with low adherence to influenza vaccination Methods Adults with CVD (n = 815) from a nationally representative dataset of the Fifth Korea National Health and Nutrition Examination Survey (KNHANES V) were analyzed. Among these adults, 500 (61.4%) had answered "yes" to whether they had received seasonal influenza vaccinations in the past 12 months. The classification process was performed using the logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGB) machine learning techniques. Because the Ministry of Health and Welfare in Korea offers free influenza immunization for the elderly, separate models were developed for the < 65 and ≥ 65 age groups. Results The accuracy of machine learning models using 16 variables as predictors of low influenza vaccination adherence was compared; for the ≥ 65 age group, XGB (84.7%) and RF (84.7%) have the best accuracies, followed by LR (82.7%) and SVM (77.6%). For the < 65 age group, SVM has the best accuracy (68.4%), followed by RF (64.9%), LR (63.2%), and XGB (61.4%). Conclusions The machine leaning models show comparable performance in classifying adult CVD patients with low adherence to influenza vaccination.

Download Full-text