My colleagues and I recently published the paper, Theory-Guided Multiclass Text Classification in Online Academic Discussions, in the prestigious journal, Journal of Computer Information Systems. The paper presents on machine learning approaches to enhance our understanding of idea generation in asynchronous online discussions. Below is the tag cloud, abstract and full citation of the research.

Abstract

Machine learning (ML) and deep learning (DL) provide significant opportunities to enhance our understanding of idea generation in asynchronous online discussions (AODs). Drawing on the interaction analysis model (IAM) as our theoretical framework, we built one baseline ML and three DL systems to automate message classification when assessing collaborative knowledge construction depth in academic AODs. The viability of these systems was demonstrated via four offerings of a traditional online course. We achieved 79% as the highest overall accuracy score across all phases of the IAM. To the best of our knowledge, this study is the first to classify AOD messages across all IAM phases. We contribute to the theory by updating the IAM to better explain how to promote deeper interactions in AODs. Additionally, we provide a methodological blueprint for future research where classifying text is crucial.

Reference

E. Eryilmaz, B. Thoms, Z. Ahmed. “Theory-Guided Multiclass Text Classification in Online Academic Discussions,” Journal of Computer Information Systems, 1–12, 2024. (Impact Factor 3.91)

Read More

Read More

My colleagues and I recently published the paper, Classifying Vaccine Misinformation in Online Social Media Videos using Natural Language Processing and Machine Learning, which was presented at HICSS 57 in Oahu, Hawaii. The paper presents on machine learning approaches to detecting misinformation in social media videos related to vaccines. Below is the tag cloud, abstract and full citation of the research.

Abstract

The spread of information through online social media videos is one of the most popular ways to share and obtain information, while at the same time the spread of misinformation across these same social spaces has become a significant concern affecting human well-being. Being able to detect this misinformation before it spreads is becoming more and more desirable for many social media platforms. This research focuses on exploring the accuracy of detecting misinformation across two social media platforms, YouTube and BitChute. This involves the classification of video data into two types: genuine information or misinformation.  More specifically, this research generates additional metadata embedded within online videos related to the COVID-19 vaccination. Using natural language processing (NLP) we extract medical subject headings (MeSH) terms from video transcripts and classify videos using four machine learning techniques including naïve Bayes, random forest, support vector machine, and logistic regression. Implementation of each classifier is presented, and the accuracy of each technique is compared and discussed. 

Reference

S. Schmidt, B. Thoms, E. Eryilmaz, J. Isaacs, “Classifying Vaccine Misinformation in Online Social Media Videos using Natural Language Processing and Machine Learning,” Proceedings of Hawaiian International Conference on System Sciences (HICSS 57), January 2-6, 2024.

Read More

Recent news has raised the alarm for how internet companies manage user data. The biggest concern, at least according to U.S. lawmakers is TikTock, which has been criticized for its extensive data collection of user data, including personal information, location data, device information, and browsing history. Lawmakers have attempted to prove that foreign Influence and TikTok’s Chinese ownership (by ByteDance) pose a national security threat by the Chinese government. Whether true or not, TikTock is just one company in an ever-growing number of internet companies that exploit user data.

The HTTP Cookie has been around since the heyday of the Dotcom Era and remains the simple, yet core technology that supports how user data is tracked across websites. Specifically, cookies are critical for:

1. Personalization: Websites use cookies to remember your preferences and provide personalized content and experiences.
2. Analytics: Cookies track user behavior on websites, such as pages visited, time spent on each page, and links clicked. This data helps website owners analyze and improve their site’s performance and user experience.
3. Advertising: Third-party cookies are often used by advertisers to track users across different websites and deliver targeted advertisements based on their interests and browsing history.
4. Authentication: Cookies can store login credentials and session information, allowing users to stay logged in to websites and access restricted content without having to re-enter their credentials each time.

Today, internet users are provided with ‘more’ transparency in how websites use these cookies. An interesting article on Wired.com (https://www.wired.com/story/cookie-pop-up-ad-tech-partner-top-websites/) discusses this phenomenon in elaborate detail. In California, for example, websites are required to ask permission to collect and share cookie data with ‘partners’, providing more information to internet users. And we all know that more information is always better, right? Maybe yes, maybe no. Obviously there are many different types of internet users. Some users might prefer to micromanage cookie access, while others might prefer a broader solution where a single policy could apply to a larger swath of websites. In any case, this is a gentle reminder to pay attention to those pop-up notifications, because you are being tracked and the page you land on is just the first stop in a vast network of data sharing ‘partners’.

Read More

Hot off the presses, How to Build a Health eProfile: Empowering Yourself with Personal Health Data, aims to help individuals unlock the power of their personal health information. Through the creation of a comprehensive Health eProfile, individuals can gain valuable insights into their health and make informed decisions about their lifestyle and healthcare choices.

The book is now available for purchase on Amazon and Kindle Stores: https://www.amazon.com/dp/B0CSXHVJBX?ref_=cm_sw_r_cp_ud_dp_GHVV0HRMK0RX5QXE9HPM

In this book, readers will learn:

• How to gather and store health information in a secure and easily accessible location.
• The importance of data quality and how to ensure the accuracy and reliability of health data.
• How to utilize a Health eProfile to make informed decisions about personal health and lifestyle.
• The best practices for sharing a Health eProfile with trusted friends, family, and healthcare providers.
• Essential considerations for building a successful Health eProfile, including privacy and security, data storage, and access.

Read More

Most of us are familiar with Artificial Intelligence, aka, AI. If you are new to the game, check out this Wikipedia link.

Eloquently defined by ChatGPT, AI is a field of computer science and technology that focuses on creating systems and machines capable of performing tasks that typically require human intelligence. These tasks can include understanding natural language, recognizing patterns, solving problems, making decisions, and learning from experience. AI systems aim to simulate human-like cognitive functions and processes, albeit often in a more specialized and narrow context. AI can be categorized into several subfields, including: machine learning, deep learning, natural language processing, computer vision, robotics, expert systems, reinforcement learning, and AI planning. Some systems, like ChatGPT incorporate many of these fields, but what makes ChatGPT so effective is its combination of deep learning, natural language processing (NLP), and neural network architecture, which is built atop a tremendous amount of data across a large distribution of powerful servers.

How we decide to incorporate AI into our daily lives will largely dictate how effective AI can will be. Used in the field of healthcare, AI shows promising potential in drug discovery and treatment plans based on an individual’s unique genetic composition. On the opposite end this spectrum, AI can be used for maleficent purposes, including robotic warfare or social profiling. AI’s impact on humanity can also be more nuanced. Take for example AI’s ability to generate ‘art’ or write music. How society addresses the many challenges of AI remains to be seen, but for the time being, AI is continues to happen all around us, whether we are aware or not. Either way, for now, let’s give Weird Al some love:

Read More