• Digital Humanities and Cultural Heritage
• Publicationsclose
• Research dataclose
• Research softwareclose
• 2017-2021close
• VTechWorks close

Cybergrooming refers to the crime of establishing personal close relationships with potential victims, commonly teens, for the purpose of sexual exploitation or abuse via online social media platforms. Cybergrooming has been recognized as a serious social problem. However, there have been insufficient programs to provide proactive prevention to protect the youth users from cybergrooming. In this thesis, we present a generative chatbot framework, called SERI (Stop cybERgroomIng), that can generate simulated conversations between a perpetrator chatbot and a potential victim chatbot. To realize the simulation of authentic conversations in the context of cybergrooming, we take deep reinforcement learning (DRL)-based dialogue generation to simulate the authentic conversations between a perpetrator and a potential victim. The design and development of the SERI are motivated to provide a safe and authentic chatting environment to enhance the youth's precautionary awareness and sensitivity of cybergrooming while any unnecessary ethical issues (e.g., the potential misuse of the SERI) are removed or minimized. We developed the SERI as a preliminary platform that the perpetrator chatbot can be deployed in social media environments to interact with human users (i.e., youth) and observe the conversations that the youth users respond to strangers or acquaintances when they are asked for private or sensitive information by the perpetrator. We evaluated the quality of conversations generated by the SERI based on open-source, referenced, and unreferenced metrics as well as human evaluation. The evaluation results show that the SERI can generate authentic conversations between two chatbots compared to the original conversations from the used datasets in perplexity and MaUde scores. Master of Science Cybergrooming refers to the crime of building personal close relationships with potential victims, especially youth users such as children and teenagers, for the purpose of sexual exploitation or abuse via online social media platforms. Cybergrooming has been recognized as a serious social problem. However, there have been insufficient methods to provide proactive protection for the youth users from cybergrooming. In this thesis, we present a generative chatbot framework, called SERI (Stop cybERgroomIng), that can generate simulated authentic conversations between a perpetrator chatbot and a potential victim chatbot by applying advanced natural language generation models. The design and development of the SERI are motivated to ensure a safe and authentic environment to strengthen the youth's precautionary awareness and sensitivity of cybergrooming while any unnecessary ethical issues (e.g., the potential misuse of the SERI) are removed or minimized. We used different metrics and methods to evaluate the quality of conversations generated by the SERI. The evaluation results show that the SERI can generate authentic conversations between two chatbots compared to the original conversations from the used datasets.

Historically, research on language acquisition among multicompetent speakers has devoted significant attention to the area of morphosyntax, compared to other domains of language (Ortega 2009, p [...] Published version

Risk Sentiment Analysis proposes a solution for companies to evaluate user submitted reports to determine workplace safety. The web application develops a pipeline for managers to submit employee CSVs to be processed through a custom sentiment analysis model in AWS. Alongside that, it has graphical presentation capabilities to easily interpret queried data. Finally, the web application supports a feedback loop of re-training the model to improve its accuracy with a page dedicated for human review. RiskPredictionSentimentAnalysisReport.pdf: Contains our project report. RiskPredictionSentimentAnalysisReportDocs.zip: Contains the assets for our LaTeX document for future edits. RiskPredictionSentimentAnalysisPresentation.pdf: Contains the slides for our final presentation. RiskPredictionSentimentAnalysisPresentationDoc.pptx: Contains the powerpoint slides of our final presentation for future edits.

Abstract: As the gap between strategic commitments and budgetary constraints continues to grow, defence organisations have introduced performance management initiatives to support decision-making and to improve governance. However, introducing managerial practices in public organisations, including defence, proves to be challenging. As performance management initiatives within defence suffer from an implementation gap, strategic benefits are not being harnessed. In our study, we first exploit the results of a systematic literature review to better anchor the encountered challenges within the literature. We then apply thematic analysis to a unique dataset from twelve NATO countries to propose a new defence-specific performance management framework for the strategic level. As the framework preserves the benefits of existing initiatives while mitigating most recorded challenges, it is proposed as a new guide for designing and assessing defence performance management efforts. Thereby, professionals and scholars are provided with a powerful instrument to address the implementation gap. Moreover, the theoretical and empirical lens adopted facilitates alignment between performance management initiatives, defence policy, defence strategy, and strategic objectives. Notably, policy goals and strategic “ends” are clearly connected to critical processes and resources. Thereby, the proposed framework better supports discussions with key defence stakeholders pertaining to the gap between commitments and constraints.

Community resilience (CR) has been studied as an indicator to measure how well a given community copes with a given disaster and provides policy directions on what aspects of the community should be improved with high priority. Although the impact of the COVID-19 has been serious all over the world and every aspect of our daily life, some countries have handled this disaster better than others. In this thesis, I aim to assess the effect of various news and Tweets collected during the COVID-19 pandemic on community functionality and resilience. First, we measure the community resilience (CR) in five different countries using Tweeter data and investigated how each country shows different trends of the CR, which is measured based on real or fake Tweets. We use Tweets generated in Australia (AUS), Singapore (SG), Republic of Korea (ROK), the United Kingdom (UK), and the United States (US) for Mar.-Nov. 2020 and measured the CR of each country and associated attributes for analyzing the overall trends. In the next step, we scrap and manually clean 4,952 full-text news articles from Jan. 2020 to Jun. 2021 and classify them into real, mixed, and fake news by fact-checking. Then we retrieve Tweets from 42,877,312 Tweets IDs from the same period and classify them into real, mixed, and fake Tweets using machine learning classifiers. We compare CR measured from news articles and Tweets based on three categories, namely, real, mixed, and fake. Based on the news articles and Tweets collected, we quantify CR based on two key factors, community wellbeing and resource distribution. We evaluate community wellbeing by assessing mental wellbeing and physical wellbeing while evaluating resource distribution by assessing economic resilience, infrastructural resilience, institutional resilience, and community capital. Based on the estimates of these two factors, we quantify CR from both news articles and Tweets and analyze the extent to which CR measured from the news articles can reflect the actual state of CR measured from Tweets. M.S. The COVID-19 pandemic has severely harmed every aspect of our daily lives, resulting in a slew of social problems. It is critical to accurately assess the current state of community functionality and resilience under this pandemic to recover from it successfully. To accomplish this, various types of social sensing techniques, such as Tweeting and publicly released news, have been employed to understand individuals’ and communities’ thoughts, behaviors, and attitudes during the COVID-19 pandemic. However, some portions of the released news are fake and can easily mislead the community to respond improperly to disasters like COVID-19. In this thesis, I aim to assess the effect of various news and Tweets collected during the COVID-19 pandemic on community functionality and resilience. First, we measure the community resilience (CR) in five different countries, i.e., Australia (AUS), Singapore (SG), Republic of Korea (ROK), the United Kingdom (UK), and the United States (US), for Mar.-Nov. 2020 and measured the CR of each country and associated attributes for analyzing the overall trends. In the next step, we compare CR measured from news articles and Tweets based on three categories, namely, real, mixed, and fake. We quantify CR based on two key factors, community wellbeing and resource distribution. We evaluate community wellbeing by assessing mental wellbeing and physical wellbeing while evaluating resource distribution by assessing economic resilience, infrastructural resilience, institutional resilience, and community capital.

AbstractThis article develops a typology of historical and archival gaps—physical, historiographical, and epistemological—to consider how non-existent sources are central to understanding colonial law and governance. It does so by examining the institutional and archival history of a court known as the Chaudrie in the French colony of Pondichéry in India in the eighteenth century, and integrating problems that are specific to the study of legal history—questions pertaining to jurisdiction, codification, evidence, and sovereignty—with issues all historians face regarding power and the making of archives. Under French rule, Pondichéry was home to multiple judicial institutions, administered by officials of the French East Indies Company. These included the Chaudrie court, which existed at least from 1700 to 1827 as a forum where French judges were meant to dispense justice according to local Tamil modes of dispute resolution. However, records of this court prior to 1766 have not survived. By drawing on both contemporaneous mentions of the Chaudrie and later accounts of its workings, this study centers missing or phantom sources, severed from the body of the archive by political, judicial, and bureaucratic decisions. It argues that the Chaudrie was a court where jurisdiction was decoupled from sovereignty, and this was the reason it did not generate a state-managed and preserved archive of court records for itself until the 1760s. The Chaudrie’s early history makes visible a relationship between law and its archive that is paralleled by approaches to colonial governance in early modern French Empire.

Accepted version

Background: The vaccination uptake rates of the human papillomavirus (HPV) vaccine remain low despite the fact that the effectiveness of HPV vaccines has been established for more than a decade. Vaccine hesitancy is in part due to false information about HPV vaccines on social media. Combating false HPV vaccine information is a reasonable step to addressing vaccine hesitancy. Objective: Given the substantial harm of false HPV vaccine information, there is an urgent need to identify false social media messages before it goes viral. The goal of the study is to develop a systematic and generalizable approach to identifying false HPV vaccine information on social media. Methods: This study used machine learning and natural language processing to develop a series of classification models and causality mining methods to identify and examine true and false HPV vaccine-related information on Twitter. Results: We found that the convolutional neural network model outperformed all other models in identifying tweets containing false HPV vaccine-related information (F score=91.95). We also developed completely unsupervised causality mining models to identify HPV vaccine candidate effects for capturing risk perceptions of HPV vaccines. Furthermore, we found that false information contained mostly loss-framed messages focusing on the potential risk of vaccines covering a variety of topics using more diverse vocabulary, while true information contained both gain- and loss-framed messages focusing on the effectiveness of vaccines covering fewer topics using relatively limited vocabulary. Conclusions: Our research demonstrated the feasibility and effectiveness of using predictive models to identify false HPV vaccine information and its risk perceptions on social media. Technology Services and the Cancer Center at Illinois at the University of Illinois at Urbana-Champaign; National Institutes of HealthUnited States Department of Health & Human ServicesNational Institutes of Health (NIH) - USA [1R21CA237483] Published version We thank Chieh-Li (Julian) Chin and Wen-Yuh Su for their assistance in the data collection and preparation for this study. We also thank the Technology Services and the Cancer Center at Illinois at the University of Illinois at Urbana-Champaign for their support of our study. This study was supported by a grant from the National Institutes of Health (#1R21CA237483). Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the National Institutes of Health.

During the unprecedented global pandemic of Covid-19, the general public is suffering from inaccurate Covid-19 related information including outdated information and fake news. The most used media: TV, social media, newspaper, and radio are incompetent in providing certitude and flash updates that people are seeking. In order to cope with this challenge, several public data resources that are dedicated to providing Covid-19 information were born. They rallied with experts from different fields to provide authoritative and up-to-date pandemic updates. However, the general public cannot still make complete use of such resources since the learning curve is too steep, especially for the aged and under-aged users. To address this problem, in this Thesis, we propose a question answering system that can be interacted with using simple natural language-based sentences. While building this system, we investigate qualified public data resources and from the data content they are providing, and we collect a set of frequently asked questions for Covid-19 tracking. We further build a dedicated dataset named CovidQA for evaluating the performance of the question answering system with different models. Based on the new dataset, we assess multiple machine learning-based models that are built for retrieving relevant information from databases, and then propose two empirical models which utilize the pre-defined templates to generate SQL queries. In our experiments, we demonstrate both quantitative and qualitative results and provide a comprehensive comparison between different types of methods. The results show that the proposed template-based methods are simple but effective in building question answering systems for specific domain problems. Master of Science During the unprecedented global pandemic of Covid-19, the general public is suffering from inaccurate Covid-19 related information including outdated information and fake news. The most used media: TV, social media, newspaper, and radio are incompetent in providing certitude and flash updates that people are seeking. In order to cope with this challenge, several public data resources that are dedicated to providing Covid-19 information were born. They rallied with experts from different fields to provide authoritative and up-to-date pandemic updates. However, there is room for improvement in terms of user experience. To address this problem, in this Thesis, we propose a system that can be interacted with using natural questions. While building this system, we evaluate and choose six qualified public data providers as the data sources. We further build a testing dataset for evaluating the performance of the system. We assess two Artificial Intelligence-powered models for the system, and then propose two rule-based models for the researched problem. In our experiments, we provide a comprehensive comparison between different types of methods. The results show that the proposed rule-based methods are simple but effective in building such systems.