Filtering by
- All Subjects: Social Media
Despite the importance of personal information, in many cases people do not reveal this information to the public. Predicting the hidden or missing information is a common response to this challenge. In this thesis, we address the problem of predicting user attributes and future or missing links using an egocentric approach. The current research proposes novel concepts and approaches to better understand social media users in twofold including, a) their attributes, preferences, and interests, and b) their future or missing connections and interactions. More specifically, the contributions of this dissertation are (1) proposing a framework to study social media users through their attributes and link information, (2) proposing a scalable algorithm to predict user preferences; and (3) proposing a novel approach to predict attributes and links with limited information. The proposed algorithms use an egocentric approach to improve the state of the art algorithms in two directions. First by improving the prediction accuracy, and second, by increasing the scalability of the algorithms.
This research examines the media literacy practices that news consumers use to inform their gatekeeping decisions. Using a constant comparative coding method, the author conducted a qualitative analysis of hundreds of discussion board posts from adult participants in a digital media literacy Massive Open Online Course (MOOC) to identify major themes and examine growth in participants’ sense of responsibility related to sharing news information, their feeling of empowerment to make informed decisions about the media messages they receive, and how the media literacy tools and techniques garnered from the MOOC have affected their daily media interactions. Findings emphasize the personal and contextual nature of media literacy, and that those factors must be addressed to ensure the success of a media literacy education program.
Adverse Drug Reaction (ADR) identification. Such methods employ a step of drug search followed by classification of the associated text as consisting an ADR or not. Although this method works efficiently for ADR classifications, if ADR evidence is present in users posts over time, drug mentions fail to capture such ADRs. It also fails to record additional user information which may provide an opportunity to perform an in-depth analysis for lifestyle habits and possible reasons for any medical problems.
Pre-market clinical trials for drugs generally do not include pregnant women, and so their effects on pregnancy outcomes are not discovered early. This thesis presents a thorough, alternative strategy for assessing the safety profiles of drugs during pregnancy by utilizing user timelines from social media. I explore the use of a variety of state-of-the-art social media mining techniques, including rule-based and machine learning techniques, to identify pregnant women, monitor their drug usage patterns, categorize their birth outcomes, and attempt to discover associations between drugs and bad birth outcomes.
The technique used models user timelines as longitudinal patient networks, which provide us with a variety of key information about pregnancy, drug usage, and post-
birth reactions. I evaluate the distinct parts of the pipeline separately, validating the usefulness of each step. The approach to use user timelines in this fashion has produced very encouraging results, and can be employed for a range of other important tasks where users/patients are required to be followed over time to derive population-based measures.
With a prison population that has grown to 1.4 million, an imprisonment rate of 419 per 100,000 U.S. residents, and a recidivism rate of 52.2% for males and 36.4% for females, the United States is facing a crisis. Currently, no sufficient measures have been taken by the United States to reduce recidivism. Attempts have been made, but they ultimately failed. Recently, however, there has been an increase in experimentation with the concept of teaching inmates basic computer skills to reduce recidivism. As labor becomes increasingly digitized, it becomes more difficult for inmates who spent a certain period away from technology to adapt and find employment. At the bare minimum, anybody entering the workforce must know how to use a computer and other technological appliances, even in the lowest-paid positions. By incorporating basic computer skills and coding educational programs within prisons, this issue can be addressed, since inmates would be better equipped to take on a more technologically advanced labor market.<br/>Additionally, thoroughly preparing inmates for employment is a necessity because it has been proven to reduce recidivism. Prisons typically have some work programs; however, these programs are typically outdated and prepare inmates for fields that may represent a difficult employment market moving forward. On the other hand, preparing inmates for tech-related fields of work is proving to be successful in the early stages of experimentation. A reason for this success is the growing demand. According to the U.S. Bureau of Labor Statistics, employment in computer and information technology occupations is projected to grow 11 percent between 2019 and 2029. This is noteworthy considering the national average for growth of all other jobs is only 4 percent. It also warrants the exploration of educating coders because software developers, in particular, have an expected growth rate of 22 percent between 2019 and 2029. <br/>Despite the security risks of giving inmates access to computers, the implementation of basic computer skills and coding in prisons should be explored further. Programs that give inmates access to a computing education already exist. The only issue with these programs is their scarcity. However, this is to no fault of their own, considering the complex nature and costs of running such a program. Accordingly, this leaves the opportunity for public universities to get involved. Public universities serve as perfect hosts because they are fully capable of leveraging the resources already available to them. Arizona State University, in particular, is a more than ideal candidate to spearhead such a program and serve as a model for other public universities to follow. Arizona State University (ASU) is already educating inmates in local Arizona prisons on subjects such as math and English through their PEP (Prison Education Programming) program.<br/>This thesis will focus on Arizona specifically and why this would benefit the state. It will also explain why Arizona State University is the perfect candidate to spearhead this kind of program. Additionally, it will also discuss why recidivism is detrimental and the reasons why formerly incarcerated individuals re-offend. Furthermore, it will also explore the current measures being taken in Arizona and their limitations. Finally, it will provide evidence for why programs like these tend to succeed and serve as a proposal to Arizona State University to create its own program using the provided framework in this thesis.