Publications

Co-Designing an AI Chatbot to Improve Patient Experience in the Hospital: A human-centered design case study of a collaboration between a hospital, a university, and ChatGPT

Wang X., Abubaker S., Babalola G. & Tulk Jesso S. (2024). "Co-Designing an AI Chatbot to Improve Patient Experience in the Hospital: A human-centered design case study of a collaboration between a hospital, a university, and ChatGPT." Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems

Healthcare; Patient Experience; Human-centered Design; Natural Language Processing; ChatGPT; Large Language Model; Data Mining; Qualitative Analysis

Constructing an Anti-Asian Hate Indicator for Pandemic-related Comments from Mainstream Media YouTube Channels

Wang X., Chen X., Bolian L. & Zhao P. (2022). "Constructing an Anti-Asian Hate Indicator for Pandemic-related Comments from Mainstream Media YouTube Channels." International Journal of Society Systems Science in Publishing

Anti-Asian hate indicator; YouTube comments; Mainstream media channels; BERT embedding; Support vector machine; Random forest; LSTM; CNN; YouTube Data API

Abstract: Anti-Asian racism, linked to COVID-19, has become a serious social problem in the United States and all over the world and even led to hate crime and violence. Even though the current anti-Asian hate study focuses anti-Asian hate classification using machine learning and sentiment analysis toward tweets, this study provides a novel pandemic-news-related anti-Asian hate indicator to depict the anti-Asian hate shift of YouTube mainstream media commentary section. A new dataset for daily hate signal generation, which contains over 1 million YouTube comments, has been generated in this study. To train the classifier, 3,759 comments are sampled and manually labelled as hate and non-hate. In the model selection among machine learning and deep learning algorithms, a CNN model is selected as the best one with a 95% accuracy and a 0.99 AUC score, which can classify 1,433,246 comments.

Classifying COVID-19-related hate Twitter users using deep neural networks with sentiment-based features and geopolitical factors

Zhao, P., Chen, X., & Wang, X. (2021). "Classifying COVID-19-related hate Twitter users using deep neural networks with sentiment-based features and geopolitical factors." International Journal of Society Systems Science 13(2), 125-139. doi:10.1504/IJSSS.2021.116373

Pandemic-related-hate; GIS; Sentiment analysis; Anti-Asian Hate tweets; Deep neural network algorithm

Abstract: Anti-Asian hate tweets caused by COVID-19 pandemic is an ongoing social problem in the USA and around the world. Although existing studies have been done by using a text classifier, little is known on how deep learning works with public sentiments of political opinions and geographical diversities. This paper provides a new method to classify the pandemic-related anti-Asian hater on Twitter. A novel dataset for tracking pandemic-related Twitter users, which contains more than 10 million tweets, is created in this study. Target users are annotated by identifying their sentiments towards the US elections with their geolocations. The empirical result indicates that the political sentiments and the county-level election results make significant contributions to the model building. By training a DNN model, over 190,000 Twitter users are classified as hate or non-hate with a 61% accuracy and a 0.63 AUC score.

Fake news and misinformation detection on headlines of COVID-19 using deep learning algorithms

Wang, X., Zhao, P., & Chen, X. (2020). "Fake news and misinformation detection on headlines of COVID-19 using deep learning algorithms." International Journal of Data Science 5(4), 316-33. doi:10.1504/IJDS.2020.115873

COVID-19; Fake News Detection; Mainstream media credibility; Long-term memory; Converlutional neural network; Deep belief networks; Cloud computing

Abstract:This paper proposed a deep learning algorithm system to fulfil fake news and misinformation detection on COVID-19 related headlines. Long short-term memory (LSTM), convolutional neural network (CNN) and Deep belief networks (DBNs) are performed in order to determine the optimal algorithm. Based on the model performance measures, such as accuracy, AUC score, and F1 score, this study figures out the optimal models, which are CNN and LSTM with an accuracy of up to 94%, for the COVID-19 fake news detection. Finally, this paper provides an algorithm-based ranking method for mainstream media credibilities. The result indicates that mainstream media channels in the US are reliable for reporting the COVID-19 related news and information.

Satellite Image Recognition for Smart Ships Using A Convolutional Neural Networks Algorithm

Xiao H., Wang X., & Zhao, P. (2019). "Satellite Image Recognition for Smart Ships Using A Convolutional Neural Networks Algorithm." International Journal of Decision Science 10(2): 85–91.

Image feature extraction from JSON object; CNN algorithm; Python Tensorflow

Abstract:In recent years, along with the development of artificial intelligence technologies and related technical products, the evolution of smart ship has accelerated. Smart ship has become the main development direction of ship industry in the future. In this paper, we proposed a CNN model to recognize ships in bay and sea area. Data sets of Ships in Satellite Imagery data and Airbus data were employed for model training and testing and features are pixel data of images and used in the classification problem. We used labels either “ship” or “no-ship” as our dependant variable to train the CNN model. Finally we get high accuracy of 98.125% for ship satellite images recognition. Through the performance metrics, including precision, recall and F1-score, we proved the reliability of this CNN model. Moreover, our CNN model is able to identify real bay and sea satellite images as well. The results make a great contribution for the development of smart ship and carve out the possibilities for fully automated operation of ship and ports.

The Relationship between Bitcoin and Stock Market

Wang, X., Chen, X., & Zhao, P. (2020). The relationship between Bitcoin and stock market.; International Journal of Operations Research and Information Systems 11(2): 22-35. doi:10.4018/IJORIS.2020040102

Blockchain technique; Cryptocurrecy; VAR model; Impulse response analysis; Price return; Sliding window technique; Yahoo API

Abstract:This article analyzes the relationship between Bitcoin and the stock market by using a vector autoregressive model. To enhance the impulse response signal, the Sliding Window technique is applied. Study results show the relationship between Bitcoin and the stock market. First, the S&P 500 has a relatively significant effect on Bitcoin, while the influence caused by the S&P 500 is weak. In addition, after involving the Sliding Window technique, the effects caused by the standard deviation of the S&P 500 and the mean of the Dow Jones are remarkably strong on the mean of Bitcoin and the standard deviation of the S&P 500 has a comparatively significant effect on the standard deviation of Bitcoin as well. Generally, the S&P 500 and the Dow Jones indexes have an advantageous effect on Bitcoin. Financial investment can be made based on this model and conclusion.

Geospatial and Temporal Data Analysis on the New York City Taxi Trip

Subramaniam V., Srungarapu G., Matijosaitiene I., Supe M., Agarwal A., Zhao P., Wang X., Kwartler E. and Jaume S. (2016). "Geospatial and Temporal Data Analysis on the New York City Taxi Trip." International Journal of Data Analysis and Information Systems 8(2), 63-73.

Study on Scheduling Algorithm Based on Capacity Constraints of Maintenance Plan

Wang X. (2015). "Study on Scheduling Algorithm Based on Capacity Constraints of Maintenance Plan." Industrial Engineering Practice ISSN 2304-5337.