Carol Hargreaves

Dr. Carol Hargreaves

Department of Statistics and Data Science, National University of Singapore, Singapore, Central Singapore, Singapore
orcid
Associate Professor

Research Focus & Metrics

Academic Career Length

24Years (2003-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 October 2021 in Science of The Total Environment
Verified
Journal Article
Closed access
YJYuting Ju
Yuting Ju
Carol HargreavesCarol Hargreaves
Carol Hargreaves

8 October 2022 in International Journal of Information Management Data Insights
Verified
Journal Article
Open access
RGReyes Michaela Denise Gonzales
Reyes Michaela Denise Gonzales
Carol HargreavesCarol Hargreaves
Carol Hargreaves

11 January 2021 in Journal of Experimental & Theoretical Artificial Intelligence
Verified
Journal Article
Closed access
CWChengyuan Wu
Chengyuan Wu
Carol HargreavesCarol Hargreaves
Carol Hargreaves
We develop a method for analyzing multivariate time series using topological data analysis (TDA) methods. The proposed methodology involves converting the multivariate time series to point cloud data, calculating Wasserstein distances between the persistence diagrams and using the k -nearest neighbours algorithm ( k -NN) for supervised machine learning. Two methods (symmetry-breaking and anchor points) are also introduced to enable TDA to better analyze data with heterogeneous features that are sensitive to translation, rotation or choice of coordinates. We apply our methods to room occupancy detection based on 5 time-dependent variables (temperature, humidity, light, CO2 and humidity ratio). Experimental results show that topological methods are effective in predicting room occupancy during a time window. We also apply our methods to an Activity Recognition dataset and obtained good results.

1 September 2012
Verified
Conference Paper
Closed access
Carol HargreavesCarol Hargreaves
Carol Hargreaves
YHYi Hao
Yi Hao
With the easy access to share information and data, many investors worldwide are interested in predicting stock prices. The prediction of stock prices using data mining techniques applied to technical variables has been widely researched but not much research to date has been done in applying data mining techniques to both technical and fundamental information. This paper is based on a personal approach to stock selection, using both technical and fundamental information. In this paper we construct a framework that enables us to make class predictions about industrial stock companies' financial performances. In order to have a systemized approach for the selection of stocks and a high likelihood of the performance of the stock price increasing, a Data Mining Approach is applied. A trading strategy is also designed and the performance of the stocks evaluated. Our two goals are to validate our stock selection methodology and to determine whether our trading strategy allows us to outperform the Australian market. Simulation results show that our selected stock portfolios outperform the Australian All-Ordinaries Index. Our findings justify the use of data mining techniques for classification and prediction purposes. Further, in conclusion, we can safely say that our stock selection and trading strategy outperformed the Australian Ordinary index.

1 May 2013 in Journal of Emerging Technologies in Web Intelligence
Verified
Journal Article
Closed access
PDF file
Carol HargreavesCarol Hargreaves
Carol Hargreaves
YHYi Hao
Yi Hao
With an easy access to share information and data nowadays, many investors worldwide are interested in predicting stock prices. The prediction of stock prices using data mining techniques applied to technical variables has been widely researched but not much research to date has been done in applying data mining techniques to both technical and fundamental information. This paper is based on a personal approach to stock selection, using both technical and fundamental information. In this paper we construct a framework that enables us to make class predictions about industrial stock performances. In order to have a systemized approach for the selection of stocks and a high likelihood of the performance of the stock price increasing, several analytical techniques are applied. A trading strategy is also designed and the performance of the stocks evaluated. Our two goals are to validate our stock selection methodology and to determine whether our trading strategy allows us to outperform the Australian market. Simulation results show that our selected stock portfolios outperform the Australian All-Ordinaries Index. Our findings justify the use of analytics for classification and prediction purposes. Further, in conclusion, we can safely say that our stock selection and trading strategy outperformed the Australian Ordinary index.

19 July 2022 in International Journal of Information Management Data Insights
Verified
Journal Article
Open access
YWYu Wang
Yu Wang
Carol HargreavesCarol Hargreaves
Carol Hargreaves
In the fight against COVID-19, the immediate and accurate screening of infected patients is of great significance. Chest X-Ray (CXR) and Computed Tomography (CT) screening play an important role in the diagnosis of COVID-19. Studies showed that changes occur in Chest Radiological images before the beginning of COVID-19 symptoms for some patients, and the symptoms of COVID-19 and other lung diseases can be similar in their very early stages. Further, it is crucial to effectively distinguish COVID-19 patients from healthy people, and patients with other lung diseases as soon as possible, otherwise inaccurate diagnosis may expose more people to the coronavirus. Many researchers have developed end-to-end deep learning techniques for the classification of COVID-19 patients without manual feature engineering. In this paper, we review the different deep learning techniques that have been used to analyze Chest X-Ray and Computed Tomography scans to classify patients with Covid-19. In addition, we also summarize the common public datasets, challenges, limitations, and possible future work. This review paper is extremely valuable as it confirms that (1) Deep Learning models are effective in classifying chest X-Ray images provided the training dataset is sufficiently large. (2) Data augmentation and generative adversarial networks (GANs) solve the small training dataset problem. (3) Transfer learning methods greatly enhanced the extraction and selection of features that were important for chest image classification. (4) Hyperparameter tuning was valuable for increasing the deep learning model accuracies to generally more than 97%. Our review study helps new researchers identify the gaps and opportunities for further or new research.