Intelligent Systems, Human-AI Collaboration

会議の名前
CHI 2022
Supporting Serendipitous Discovery and Balanced Analysis of Online Product Reviews with Interaction-Driven Metrics and Bias-Mitigating Suggestions
要旨

In this study, we investigate how supporting serendipitous discovery and analysis of online product reviews can encourage readers to explore reviews more comprehensively prior to making purchase decisions. We propose two interventions --- Exploration Metrics that can help readers understand and track their exploration patterns through visual indicators and a Bias Mitigation Model that intends to maximize knowledge discovery by suggesting sentiment and semantically diverse reviews. We designed, developed, and evaluated a text analytics system called Serendyze, where we integrated these interventions. We asked 100 crowd workers to use Serendyze to make purchase decisions based on product reviews. Our evaluation suggests that exploration metrics enabled readers to efficiently cover more reviews in a balanced way, and suggestions from the bias mitigation model influenced readers to make confident data-driven decisions. We discuss the role of user agency and trust in text-level analysis systems and their applicability in domains beyond review exploration.

著者
Mahmood Jasim
University of Massachusetts Amherst, Amherst, Massachusetts, United States
Christopher Collins
Ontario Tech University, Oshawa, Ontario, Canada
Ali Sarvghad
University of Massachusetts Amherst, Amherst, Massachusetts, United States
Narges Mahyar
University of Massachusetts Amherst, Amherst, Massachusetts, United States
論文URL

https://dl.acm.org/doi/abs/10.1145/3491102.3517649

動画
Shared Interest: Measuring Human-AI Alignment to Identify Recurring Patterns in Model Behavior
要旨

Saliency methods --- techniques to identify the importance of input features on a model's output --- are a common step in understanding neural network behavior. However, interpreting saliency requires tedious manual inspection to identify and aggregate patterns in model behavior, resulting in ad hoc or cherry-picked analysis. To address these concerns, we present Shared Interest: metrics for comparing model reasoning (via saliency) to human reasoning (via ground truth annotations). By providing quantitative descriptors, Shared Interest enables ranking, sorting, and aggregating inputs, thereby facilitating large-scale systematic analysis of model behavior. We use Shared Interest to identify eight recurring patterns in model behavior, such as cases where contextual features or a subset of ground truth features are most important to the model. Working with representative real-world users, we show how Shared Interest can be used to decide if a model is trustworthy, uncover issues missed in manual analyses, and enable interactive probing.

受賞
Honorable Mention
著者
Angie Boggust
Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
Benjamin Hoover
IBM Research AI, Cambridge, Massachusetts, United States
Arvind Satyanarayan
MIT, Cambridge, Massachusetts, United States
Hendrik Strobelt
IBM Research AI, Cambridge, Massachusetts, United States
論文URL

https://dl.acm.org/doi/abs/10.1145/3491102.3501965

動画
Q-Chef: The impact of surprise-eliciting systems on food-related decision-making
要旨

Choosing what to eat requires navigating a large volume of information and various competing factors. While recommendation systems are an effective approach to assist users with this culinary decision-making, they typically prioritise similarity to a query or user profile to give relevant results. This can expose users to an increasingly narrow band of phenomena, which could compromise dietary diversity, a factor in dietary quality. We designed Q-Chef, which combines a recipe recommendation system with a personalised model of surprise, and conducted a study to identify if surprise-eliciting recipes affect food decisions. Our study utilises a rigorous thematic analysis with over 40 participants to explore how computational models of surprise influence recipe choice. We also explored how these factors differed when people were presented with "surprising-yet-tasty" recipes, as opposed to just "tasty" recipes, and identified that being presented with surprising choices is more likely to elicit situational interest and prompt reflection on choices. We conclude with a set of suggestions for the design of future surprise-eliciting recipe systems.

著者
Kazjon Grace
University of Sydney, Sydney, Australia
Elanor Finch
The University of Sydney, Sydney, Australia
Natalia Gulbransen-Diaz
The University of Sydney, Sydney, Australia
Hamish Henderson
The University of Sydney, Sydney, Australia
論文URL

https://dl.acm.org/doi/abs/10.1145/3491102.3501862

動画
Scaling Creative Inspiration with Fine-Grained Functional Aspects of Product Ideas
要旨

Large repositories of products, patents and scientific papers offer an opportunity for building systems that scour millions of ideas and help users discover inspirations. However, idea descriptions are typically in the form of unstructured text, lacking key structure that is required for supporting creative innovation interactions. Prior work has explored idea representations that were either limited in expressivity, required significant manual effort from users, or dependent on curated knowledge bases with poor coverage. We explore a novel representation that automatically breaks up products into fine-grained functional aspects capturing the purposes and mechanisms of ideas, and use it to support important creative innovation interactions: functional search for ideas, and exploration of the design space around a focal problem by viewing related problem perspectives pooled from across many products. In user studies, our approach boosts the quality of creative search and inspirations, substantially outperforming strong baselines by 50-60%.

著者
Tom Hope
Allen Institute , Seattle , Washington, United States
Ronen Tamari
Hebrew University of Jerusalem, Jerusalem, Israel
Daniel Hershcovich
University of Copenhagen, Copenhagen, Hovedstaden, Denmark
Hyeonsu B. Kang
Carnegie Mellon University, Pittsburgh, Pennsylvania, United States
Joel Chan
University of Maryland, College Park, Maryland, United States
Aniket Kittur
Carnegie Mellon University, Pittsburgh, Pennsylvania, United States
Dafna Shahaf
The Hebrew University of Jerusalem, Jerusalem, Israel
論文URL

https://dl.acm.org/doi/abs/10.1145/3491102.3517434

動画
EmoGlass: an End-to-End AI-Enabled Wearable Platform for Enhancing Self-Awareness of Emotional Health
要旨

Often, emotional disorders are overlooked due to their lack of awareness, resulting in potential mental issues. Recent advances in sensing and inference technology provide a viable path to wearable facial-expression-based emotion recognition. However, most prior work has explored only laboratory settings and few platforms are geared towards end-users in everyday lives or provide personalized emotional suggestions to promote self-regulation. We present EmoGlass, an end-to-end wearable platform that consists of emotion detection glasses and an accompanying mobile application. Our single-camera-mounted glasses can detect seven facial expressions based on partial face images. We conducted a three-day out-of-lab study (N=15) to evaluate the performance of EmoGlass. We iterated on the design of the EmoGlass application for effective self-monitoring and awareness of users' daily emotional states. We report quantitative and qualitative findings, based on which we discuss design recommendations for future work on sensing and enhancing awareness of emotional health.

著者
Zihan Yan
Zhejiang University, Hangzhou, China
Yufei Wu
Zhejiang University, Hangzhou, China
Yang Zhang
University of California, Los Angeles, Los Angeles, California, United States
Xiang 'Anthony' Chen
UCLA, Los Angeles, California, United States
論文URL

https://dl.acm.org/doi/abs/10.1145/3491102.3501925

動画