Ali Almadan’s research while affiliated with University of North Carolina at Charlotte and other places

Public opinion provides policy makers with meaningful information on how the public feels towards a certain issue. Stance detection in social media is the problem of automatically determining the standpoint expressed in a specific tweet towards a target of interest such as a topic or an issue. One such application of stance detection is to gauge public opinion from Twitter data as an alternative to traditional methods such as surveys and polls. In this paper, we define user-based stance and claim that the aggregation of user-based stance is more aligned with the goal of gauging public opinion than the aggregation of tweet-based stance. Our analysis shows that tweet-based stance aggregation leads to significantly different results than user-based stance aggregation, and the effect size varies per stance class. This paper provides the basis and argument for user-based stance to measure public opinion from Twitter data.KeywordsData analysisPublic opinionStance detectionStance analysis

Stance detection is the task of detecting the standpoint of a user towards a target of interest, such as a controversial topic. Stance detection has various applications such as surveying and polling the public as an alternative to traditional instruments to measure public opinion. One of the implications of using stance detection on Twitter data to measure public opinion is the prevalence of social bots that can impact the measured public opinion. In this paper, we propose a user-based stance analysis to mitigate the impact of social bots on measuring public opinion from stance detection in Twitter. In contrast to a tweet-based stance analysis, the user-based stance analysis shows a minimal impact of social bots on measured public opinion for all stance classes: favor, against, and neutral.KeywordsPublic opinionStance detectionStance analysisSocial bots

People turn to social media to express their opinion towards different topics and issues. This makes social media a valuable resource for mining public opinion. Stance detection is an approach to analyzing social media users’ content to determine public opinion. In this paper, we present a replicable methodology for coding tweets for stance detection towards the COVID-19 vaccination. The methodology includes a codebook for coding the stance towards COVID-19 vaccination and 2 approaches to sampling Twitter data for manual coding: keywords and hashtags. The codebook provides a template for other researchers to code social media data for stance towards vaccination. Our analysis of the results from 2 sampling approaches shows that sampling with hashtags leads to high inter-coder agreement. We analyze the stance and compare it with the results from sentiment analysis on the same dataset to highlight the distinction of our methodology for stance analysis when compared to sentiment analysis towards vaccination. The major contributions of this paper are: a replicable methodology for annotating stance towards COVID-19 vaccination with a codebook and dataset, and a comparison of sampling Twitter data with keywords and hashtags on the inter-coder agreement along with the resulting distribution of stance.KeywordsCOVID-19VaccinationPublic opinionStance

This paper presents an AI-based co-creative system in which the interaction model focuses on emotional feedback, that is, the decisions about the creative contribution from the AI agent is based on the emotion detected in the human co-creator. In human-human collaboration, gestures, verbal communications, and emotional responses are among the general communication strategies used to shape the interactions between the collaborators and negotiate the contributions. Emotional feedback allows human collaborators to passively communicate their experience and their perception of the process without distracting the flow of the task. In human-human co-creative collaboration, participants interact and contribute to the task based on their perception of the collaboration over time. In designing human-AI co-creative collaboration, we address two challenges: (1) perceiving the user’s cognitive state to determine the dynamics of collaboration, such as whether the system should lead, follow, or wait, and (2) deciding what the agent should contribute to the artifact. This paper presents a model of an AI agent that addresses these challenges and the results of our study of participants that interact with the co-creative agent.