Search

Artificial Intimacy is an AI art installation that explores what natural language processing (NLP) models in our everyday lives would feel like if they were to be personalized to match our own personalities and values. We explored the possibility of fine-tuning NLP models using personal social media data. Our selected data sources - Leslie Foster and Gorjeoux Moon - have offered their own social media data to fine-tune the models. We present a video capturing their conversations with their social media selves. The interactive portion of the installation invites the audience to engage with Foster's and Moon's chatbots and explore interactions with NLP models that are personalized in this way.

A multiplex network models different modes of interaction among same-type entities. In this article, we provide a taxonomy of community detection algorithms in multiplex networks. We characterize the different algorithms based on various properties and we discuss the type of communities detected by each method. We then provide an extensive experimental evaluation of the reviewed methods to answer three main questions: to what extent the evaluated methods are able to detect ground-truth communities, to what extent different methods produce similar community structures, and to what extent the evaluated methods are scalable. One goal of this survey is to help scholars and practitioners to choose the right methods for the data and the task at hand, while also emphasizing when such choice is problematic.

Longitudinal studies of gut microbiota following specific interventions are vital for understanding how they influence host health. However, robust longitudinal sampling of gut microbiota is a major challenge, which can be addressed using in vitro fermentors hosting complex microbial communities. Here, by employing 16S rRNA gene amplicon sequencing, we investigated the adaptation and succession of human fecal microbial communities in an automated multistage fermentor. We performed two independent experiments using different human donor fecal samples, one configured with two units of three colon compartments each studied for 22 days and another with one unit of two colon compartments studied for 31days. The fermentor maintained a trend of increasing microbial alpha diversity along colon compartments. Within each experiment, microbial compositions followed compartment-specific trajectories and reached independent stable configurations. While compositions were highly similar between replicate units, they were clearly separated between different experiments, showing that they maintained the individuality of fecal inoculum rather than converging on a fermentor-specific composition. While some fecal amplicon sequence variants (ASVs) were undetected in the fermentor, many ASVs undetected in the fecal samples flourished in vitro. These bloomer ASVs accounted for significant proportions of the population and included prominent healthassociated microbes such as Bacteroides fragilis and Akkermansia muciniphila. Turnover in community compositions is likely explained by feed composition and pH, suggesting that these communities can be easily modulated. Our results suggest that in vitro fermentors are promising tools to study complex microbial communities harboring important members of human gut microbiota.

This paper proposes to use MeetUp data to study the emergence and the evolution of the technological trend commonly known as Internet of Things (IoT). Starting from a manually selected sample of 220 European MeetUp groups we used MeetUp's APIs, to retrieve additional information about the events and participating users. The final dataset consists of 220 groups, 32967 members and 2386 events from 2011 until now (Jan 2019). The results suggest the presence of clearly identifiable European hubs for IoT development but a worldwide crowd of users. From a temporal perspective the MeetUp data shows how IoT exploded in 2015 and how it might have peaked in 2017. Within this period of time IoT has not been a “stable technology” but as evolved incorporating, within its area of “related topics” new and emerging technologies such as Cryptocurrency or cloud computing.

In the context of community detection in online social media, a lot of effort has been put into the definition of sophisticated network clustering algorithms and much less on the equally crucial process of obtaining high-quality input data. User-interaction data explicitly provided by social media platforms has largely been used as the main source of data because of its easy accessibility. However, this data does not capture a fundamental and much more frequent type of participatory behavior where users do not explicitly mention others but direct their messages to an invisible audience following a common hashtag. In the context of multiplex community detection, we show how to construct an additional data layer about user participation not relying on explicit interactions between users, and how this layer can be used to find different types of communities in the context of Twitter political communication.