Thanh Trung Phan | EPFL Graph Search

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Trust Indicators and Explainable AI: A Study on User Perceptions

Daniel Gatica-Perez, Nicolas Henchoz, Thanh Trung Phan, Delphine Ribes Lemay, Andreas Sonderegger

Nowadays, search engines, social media or news aggregators are the preferred services for news access. Aggregation is mostly based on artificial intelligence technologies raising a new challenge: Trust has been ranked as the most important factor for media business. This paper reports findings of a study evaluating the influence of manipulations of interface design and information provided in the context of eXplainable Artificial Intelligence (XAI) on user perception and in the context of news content aggregators. In an experimental online study, various layouts and scenarios have been developed, implemented and tested with 266 participants. Measures of trust, understanding and preference were recorded. Results showed no influence of the factors on trust. However, data indicates that the influence of the layout, for example implicit integration of media source through layout structuration has a significant effect on perceived importance to cite the source of a media. Moreover, the amount of information presented to explain the AI showed a negative influence on user understanding. This highlights the importance and difficulty of making XAI understandable for its users.

Springer2021

Learning Urban Nightlife Routines from Mobile Data

Daniel Gatica-Perez, Thanh Trung Phan

The use of smartphone sensing for public health studies is appealing to understand routines. We present an approach to learn nightlife routines in a smartphone sensing dataset volunteered by 184 young people (1586 weekend nights with location data captured between 8PM and 4AM.) Human activity is represented at two levels, namely as the types of places visited and as the areas of the city where those places are. Routines extracted with two topic models (Latent Dirichlet Allocation and Hierarchical Dirichlet Process) are semantically meaningful and represent different moments of the weekend night, depicting activities such as pub crawling. The inference capacity of the routine representation is demonstrated with two classification tasks of value for alcohol research (alcohol consumption throughout the night, and heavy alcohol consumption.) The results suggest that nightlife routine mining could be used as a complementary tool to traditional survey-based methods in public health studies, and also inform other institutional actors interested in understanding and supporting youth well-being.

ACM2020

Shooting shots: Estimating alcoholic drink sizes in real life using event-level reports and annotations of close-up pictures

Daniel Gatica-Perez, Thanh Trung Phan

Introduction and Aims Drinks consumed in real life are diverse, in terms of beverage type, container size and alcohol by volume. To date, most ecological momentary assessment studies have assessed drinking amounts with 'standard' drinks, although their event-level design allows for more advanced assessment schemes. The purpose of this empirical study is to compare participants' estimates of alcoholic drink characteristics, assessed using drink-specific questions, with estimates generated by annotators based on pictures of the same drinks. Design and Methods On weekend nights, 186 young adults took 1484 close-up pictures of their drinks using a custom-built smartphone application. Participants reported the beverage type, drink size and alcohol by volume. Annotators described the beverage type, container size and filling level. Correspondence between participants' and annotators' estimates was explored using descriptive statistics, difference tests and correlations. Results Annotators were unable to precisely identify the beverage types in most pictures of liqueurs, spirits and mixed drinks. Participants' drink size estimates converged with annotators' estimates of the container size for beer (41 cl corresponding to 16 g of pure alcohol) and mixed drinks (28 cl/35 g), and of the content size for wine (10 cl/9 g). However, annotators estimated larger sizes for liqueur/fortified wine (12 cl/14 g vs. 7 cl/9 g) and spirits (8 cl/26 g vs. 4 cl/10 g) than participants. Discussion and Conclusions Annotations of pictures should be considered as a complement to participants' reports rather than a substitute. Except for wine, real-life drinks vary largely and often exceed 10 g 'standard' drinks.

WILEY2020