A5025175160- Mobile Crowdsensing and Crowdsourcing
- Personal Information Management and User Behavior
- Data Visualization and Analytics
- Digital Marketing and Social Media
- Hate Speech and Cyberbullying Detection
- Team Dynamics and Performance
- Ethics and Social Impacts of AI
- AI in Service Interactions
- Topic Modeling
- Innovative Human-Technology Interaction
- Human-Automation Interaction and Safety
- Sensor Technology and Measurement Systems
- Open Source Software Innovations
- Knowledge Management and Sharing
- Artificial Intelligence in Healthcare and Education
- Diverse Aspects of Tourism Research
- Anomaly Detection Techniques and Applications
- Innovative Approaches in Technology and Social Development
- Marketing and Advertising Strategies
- Social Media and Politics
- Information Systems Theories and Implementation
- Virtual Reality Applications and Impacts
- Design Education and Practice
- Persona Design and Applications
- Context-Aware Activity Recognition Systems
All India Institute of Medical Sciences Raipur
2024-2025
Google (United States)
2018-2024
Sriram Chandra Bhanja Medical College Hospital
2024
Samsung (India)
2021-2022
Cornell University
2012-2020
RWTH Aachen University
2010
Tata Consultancy Services (India)
2006
Artificial intelligence (AI) presents new challenges for the user experience (UX) of products and services. Recently, practitioner-facing resources design guidelines have become available to ease some these challenges. However, little research has investigated if how are used, they impact practice. In this paper, we industry practitioners use People + AI Guidebook. We conducted interviews with 31 (i.e., designers, product managers) understand human-AI when designing AI-enabled products. Our...
Machine learning models are commonly used to detect toxicity in online conversations. These trained on datasets annotated by human raters. We explore how raters' self-described identities impact they annotate comments. first define the concept of Specialized Rater Pools: rater pools formed based identities, rather than at random. three such for this study - specialized raters from U.S. who identify as African American, LGBTQ, and those neither. Each these same set comments, which contains...
Recently, the AI/ML research community has indicated an urgent need to establish Responsible AI (RAI) values and practices as part of lifecycle. Several organizations communities are responding this call by sharing RAI guidelines. However, there gaps in awareness, deliberation, execution such for multi-disciplinary ML practitioners. This work contributes discussion unpacking co-production challenges faced practitioners they align their values. We interviewed 23 individuals, across 10...
The usual presentation of foreign body (FB) aspiration into the tracheobronchial tree is a choking sensation followed by recurrent respiratory symptoms. Without definitive history, diagnosis FB can be challenging. Here, we present case 40-year-old male who presented with episodes blood-stained sputum over past six years. Contrast-enhanced computed tomography chest and serological investigations suggested allergic bronchopulmonary aspergillosis. Video bronchoscopy was performed for hemoptysis...
Responsible design of AI systems is a shared goal across HCI and communities. (RAI) tools have been developed to support practitioners identify, assess, mitigate ethical issues during development. These take many forms (e.g., playbooks, software toolkits, documentation protocols). However, research suggests that use RAI shaped by organizational contexts, raising questions about how effective such are in practice. To better understand -- might be evaluated, we conducted qualitative analysis...
Large language model (LLM) prompting is a promising new approach for users to create and customize their own chatbots. However, current methods steering chatbot's outputs, such as prompt engineering fine-tuning, do not support in converting natural feedback on the model's outputs changes or model. In this work, we explore how enable interactively refine through feedback, by helping them convert into set of principles (i.e. constitution) that dictate behavior. From formative study, (1) found...
Synchronous colocated collaborative sensemaking requires that analysts share their information and insights with each other. The challenge is to know when the right time what without disrupting present state of analysis. This crucial in ad-hoc sessions mobile devices because small screen space limits display. To address these tensions, we propose evaluate RAMPARTS - a spatially aware system for crime analysis aims support faster sharing, clue-finding, We compare an interactive tabletop...
Online harassment is a major societal challenge that impacts multiple communities. Some members of community, like female journalists and activists, bear significantly higher since their profession requires easy accessibility, transparency about identity, involves highlighting stories injustice. Through multi-phased qualitative research study involving focus group interviews with 27 we mapped the journey target who goes through harassment. We introduce PMCR framework, as way to on needs for...
Online social spaces provide much needed connection and belonging—particularly in a context of continued lack global mobility due to the ongoing Covid-19 pandemic climate crisis. However, norms online can create environments which toxic behaviour is normalized, tolerated or even celebrated. This occur without consequence, leaving its members vulnerable hate, harassment, abuse. A vast majority adults have experienced toxicity harm more prevalent for marginalized minoritized groups, who are...
Sensemaking tasks are difficult to accomplish with limited time and attentional resources because analysts faced a constant stream of new information. While this information is often important, the timing interruptions may detract from analyst's work. In an ideal world, there would be no interruptions. But that not case in real world sensemaking tasks. So, study, we explore value based on state arousal as detected by Electrodermal activity derived form galvanic skin response (EDA)....
Collaborative sensemaking requires that analysts share their information and insights with each other, but this process of sharing runs the risks prematurely focusing investigation on specific suspects. To address tension, we propose test an interface for collaborative crime analysis aims to make more aware processes. We compare our translucence a standard without special features in controlled laboratory study. found significantly improved clue finding solving performance, rated lower...
Many sophisticated tools have been developed to help analysts detect patterns in large datasets, but the value of these tools' individual features is rarely tested. In an experiment which participants played detectives solving homicides, we tested utility a visualization data links and notepad for collecting organizing annotations. The significantly improved participants' ability solve crime whereas did not. Having both available provided no benefit over having just visualization. results...
When crime analysts collaborate to solve cases, they need share insights in order connect the clues, identify a pattern, and attribute right culprit. We designed collaborative analysis tool explore value of implicitly sharing notes, without requiring explicitly push information or request it from each other. In an experiment, pairs remote individuals played role solving set serial killer crimes with both partners having some, but not all, relevant clues. implicit notes was available,...
Metaverse is the new buzz word in field of technology, particularly media and industry. Studies virtual spaces have been going on academia for decades, industry has creating such as commercial products. However, specific term "metaverse" used sparingly, when used, definitions inconsistent. In an effort to understand how researchers are using this term, we conducted a systematic literature review reviewed articles that defined metaverse. We categorized main characteristics being properties...
Wiki-like or crowdsourcing models of collaboration can provide a number benefits to academic work. These techniques may engage expertise from different disciplines, and potentially increase productivity. This paper presents model massively distributed collaborative authorship papers. model, developed by collective thirty authors, identifies key tools that would be necessary useful the writing process. The process collaboratively this was used discover, negotiate, document issues in authored...
With the rise of open data in last two decades, more datasets are online and people using them for projects research. But how do find datasets? We present first user study Google Dataset Search, a dataset-discovery tool that uses Web crawl ecosystem to datasets. Search contains superset other tools â total 45M from 13K sources. found addresses previously identified need: search engine across entire Web, including tools. However, introduced new challenges due its approach: building mental...
Our ability to build autonomous agents that leverage Generative AI continues increase by the day. As builders and users of such it is unclear what parameters we need align on before start performing tasks our behalf. To discover these parameters, ran a qualitative empirical research study about designing can negotiate during fictional yet relatable task selling camera online. We found for an agent perform successfully, humans/users over 6 dimensions: 1) Knowledge Schema Alignment 2) Autonomy...
Pleural effusion is due to the pathological accumulation of pleural fluid in space, 25%-30% which may remain undiagnosed despite combination biochemical, microbiological, and tests closed biopsy. Medical thoracoscopy help physicians diagnose such cases. We aimed study diagnostic yield medical patients with exudative assess safety profile thoracoscopy.
In this paper, we study the relevance of Hall and Hofstede's works to web design beyond traditional domain areas like e-commerce, advertising. Existing theories explain how may be affected by cultural differences, explore those differences can seen in government website across Brazil, Russia, India, China, US. We describe our findings confirming that exist, more so between China US than rest, point out where fail results, particular for Russia India finally, focus on
SAVANT is a web-based tool that enables information and knowledge sharing between remote partners through explicit implicit communication to help them collaboratively analyze make sense of distributed data. SAVANT's provides an opportunity leverage partners' insights reduce cognitive tunneling, facilitates discussion. Both techniques assist collaborative sensemaking processes.
Many free mobile applications are supported by advertising. Ads can greatly affect user perceptions and behavior. In apps, ads often follow a “native” format: they designed to conform in both format style the actual content context of application. Clicking on ad leads users second destination, outside hosting app, where unified experience provided native within app is not necessarily reflected landing page arrives at. Little known about whether how this type impacting experience. paper, we...
Lack of proper English pronunciations is a major problem for immigrant population in developed countries like U.S. This poses various problems, including barrier to entry into mainstream society. paper presents research study that explores the use speech technologies merged with activity-based and arcade-based games do pronunciation feedback Hispanic children within A 3-month long California was used investigate analyze effectiveness computer aided through games. In addition quantitative...
My research aims to design systems for complex sensemaking by remotely located non-expert collaborators (crowds), solve computationally hard problems like crimes.