A5083645580- Topic Modeling
- Natural Language Processing Techniques
- AI in Service Interactions
- Spatial Cognition and Navigation
- Augmented Reality Applications
- Mobile Crowdsensing and Crowdsourcing
- Software Engineering Research
- Misinformation and Its Impacts
- Cognitive Science and Mapping
- Speech and dialogue systems
- Ethics and Social Impacts of AI
- Artificial Intelligence in Healthcare and Education
- Multi-Agent Systems and Negotiation
- Computational and Text Analysis Methods
- Geographic Information Systems Studies
- Social Robot Interaction and HRI
- Geography Education and Pedagogy
- Explainable Artificial Intelligence (XAI)
- Expert finding and Q&A systems
- Innovative Human-Technology Interaction
- Personal Information Management and User Behavior
- Semantic Web and Ontologies
- Team Dynamics and Performance
- Human Motion and Animation
- Information Retrieval and Search Behavior
Johns Hopkins University
2023-2025
Microsoft (United States)
2023-2024
Research Canada
2023-2024
University of Illinois Urbana-Champaign
2017-2024
Purdue University West Lafayette
2021-2024
Microsoft Research Montréal (Canada)
2023
University of Arizona
2023
Microsoft Research (United Kingdom)
2023
American Society For Engineering Education
2021
University of Illinois System
2019
Qualitative analysis of textual contents unpacks rich and valuable information by assigning labels to the data. However, this process is often labor-intensive, particularly when working with large datasets. While recent AI-based tools demonstrate utility, researchers may not have readily available AI resources expertise, let alone be challenged limited generalizability those task-specific models. In study, we explored use language models (LLMs) in supporting deductive coding, a major...
Large language models (LLMs) powered conversational search systems have already been used by hundreds of millions people, and are believed to bring many benefits over conventional search. However, while decades research public discourse interrogated the risk in increasing selective exposure creating echo chambers—limiting diverse opinions leading opinion polarization, little is known about such a LLM-powered We conduct two experiments investigate: 1) whether how increases compared search; 2)...
The rise of increasingly more powerful chatbots offers a new way to collect information through conversational surveys, where chatbot asks open-ended questions, interprets user's free-text responses, and probes answers whenever needed. To investigate the effectiveness limitations such in conducting we conducted field study involving about 600 participants. In this with mostly half participants took typical online survey on Qualtrics other interacted an AI-powered complete survey. Our...
During a public health crisis like the COVID-19 pandemic, credible and easy-to-access information portal is highly desirable. It helps with disease prevention, planning, misinformation mitigation. However, creating such an challenging because 1) domain expertise required to identify curate intelligible content, 2) needs be updated promptly in response fast-changing environment, 3) should easily accessible by general public; which particularly difficult when most people do not have about...
Social media sites use different labels to help users find and select news feeds. For example, Blue Feed, Red a feed created by the Wall Street Journal, stance separate articles with opposing political ideologies people explore diverse opinions. To combat spread of fake news, Facebook has experimented putting credibility on readers decide whether content is trustworthy. systematically understand effects online selection consumption, we conducted controlled experiment study how these...
Content moderation systems for social media have had numerous issues of bias, in terms race, gender, and ability among many others. One proposal addressing such automated decision making is by designing contestability, whereby users can shape influence how decisions are made. In this study, we conduct a series participatory design workshops with participants from communities that experienced problems content the past. Together participants, explore idea contestability find users' designs...
Informed consent is a core cornerstone of ethics in human subject research. Through the informed process, participants learn about study procedure, benefits, risks, and more to make an decision. However, recent studies showed that current practices might lead uninformed decisions expose unknown especially online studies. Without researcher's presence guidance, must read lengthy form on their own with no answers questions. In this paper, we examined role AI-powered chatbot improving online....
The recent advancements in Large Language Models (LLMs) have significantly impacted numerous, and will impact more, real-world applications. However, these models also pose significant risks to individuals society. To mitigate issues guide future model development, responsible evaluation auditing of LLMs are essential. This workshop aims address the current "evaluation crisis" LLM research practice by bringing together HCI AI researchers practitioners rethink from a human-centered...
We are building an intelligent agent to help teaming efforts. In this paper, we investigate the real-world use of such understand students deeply and student team formation in a large university class involving about 200 40 teams. Specifically, interacted with each text-based conversation at beginning end class. show how was able elicit in-depth information from students, infer students' personality traits, reveal complex relationships between compositions results. also report on behavior...
In this paper, a target person tracking method based on the ultra-wideband (UWB) technology is proposed for implementation of human-tracking robots. For such robots, detection and calculation position orientation play an essential role in human-following function. A modified hyperbolic positioning algorithm devised to overcome challenge measurement errors. addition, virtual spring model implemented robot track person. One important advantage UWB-based compared with computer vision-based...
Interruptions, a fundamental component of human communication, can enhance the dynamism and effectiveness conversations, but only when effectively managed by all parties involved. Despite advancements in robotic systems, state-of-the-art systems still have limited capabilities handling user-initiated interruptions real-time. Prior research has primarily focused on post hoc analysis interruptions. To address this gap, we present system that detects manages them real-time based interrupter's...
Although psychometrics researchers have recently explored the use of large language models (LLMs) as proxies for human participants, LLMs often fail to generate heterogeneous data with human-like diversity, which diminishes their value in advancing social science research. To address these challenges, we potential theory-informed Personality Structured Interview (PSI) a tool simulating responses personality In this approach, simulation is grounded nuanced real-human interview transcripts...
Interview chatbots engage users in a text-based conversation to draw out their views and opinions. It is, however, challenging build effective interview that can handle user free-text responses open-ended questions deliver engaging experience. As the first step, we are investigating feasibility effectiveness of using publicly available, practical AI technologies chatbots. To demonstrate feasibility, built prototype scoped enable with subset active listening skills - abilities comprehend...
Advancements in large language models (LLMs) are poised to spark a proliferation of LLM-powered user experiences. In product teams, designers often tasked with crafting experiences that align needs. To involve and leverage their user-centered perspectives create effective responsible products, we introduce the practice designerly adaptation for engaging LLMs as an adaptable design material. We first identify key characteristics through formative study experienced designing products (N=12)....
Large language models (LLMs) powered conversational search systems have already been used by hundreds of millions people, and are believed to bring many benefits over conventional search. However, while decades research public discourse interrogated the risk in increasing selective exposure creating echo chambers -- limiting diverse opinions leading opinion polarization, little is known about such a LLM-powered We conduct two experiments investigate: 1) whether how increases compared search;...
Certain commercial video games, such as Portal 2 and Tetris, have been empirically shown to train spatial reasoning skills, a subset of cognitive skills essential for success in STEM disciplines. However, no research date has attempted understand which specific features these games tap into players' ability or how individual player differences interact with game features. This knowledge is crucially important first step towards understanding what makes effective why, especially...
Research has demonstrated that spatial visualization skills are crucial for success in Science, technology, engineering, and mathematics (STEM) disciplines. With an increasing number of students entering STEM disciplines, the question how to effectively train students» become very important. While a scalable existing solution is implement online workshops students, problem motivate participate these remains unsolved. In this study, we studied gamification as way first year engineering take...