We utilize a variety of methods from mining data on social media to conducting controlled experiments to interviewing users. Our work aims to understand the significance of people’s digital traces and to leverage this information for positive social good.
We are part of Jacobs Institute’s Connective Media hub at Cornell Tech. Some of our research is conducted within the Connected Experiences Lab. If you are interested in working with us, please get in touch by contacting us below!
3/2017 Matt Law received an Honor Mention for an NSF Fellowship. Nice work!
2/2017 Received Google Research Award to support our work on attention to online media
1/2017 Xiao Ma recognized as a Facebook PhD Fellowship Finalist. Congrats!
Associate Professor
PhD Student
PhD Student
PhD Student
PhD Student
As part of the AOL Connected Experiences Laboratory, we look at how data from mobile devices, sensors, as well as new cryptographic techniques and protocols can enable a socio-technical infrastructure to provide awareness, trust and meaningful connections between physically co-located individuals, including buildings, offices, and public spaces. Such infrastructure will empower people to make better connections and communication in their local communities, with long term impact on participation and democracy.
The goal of this project is to advance our understanding of the psychological mechanisms behind people's attention, as reflected through their interactions with digital content. In particular, we focus on the context of actions that people takeonline without any experimental intervention and examine how context affects behavior. We draw on theories from a wide range of fields to address questions that pertain to individual's attention to content, expectations for attention from others and the value in getting that attention. To that end, we harness machine learning methods as well as language and statistical modeling to analyze signals of human attention as they occurs naturally outside of lab settings.
Yiqing Hua, Mor Naaman, and Thomas Ristenpart. Characterizing Twitter Users who Engage in Adversarial Interactions Against Political Candidates. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (CHI 2020).
CHI 2020 [Paper]
Characterizing Twitter Users who Engage in Adversarial Interactions Against Political Candidates
Social media provides a critical communication platform for political figures, but also makes them easy targets for harassment. In this paper, we characterize users who adversarially interact with political figures on Twitter using mixed-method techniques. The analysis is based on a dataset of 400 thousand users’ 1.2 million replies to 756 candidates for the U.S. House of Representatives in the two months leading up to the 2018 midterm elections. We show that among moderately active users, adversarial activity is associated with decreased centrality in the social graph and increased attention to candidates from the opposing party. When compared to users who are similarly active, highly adversarial users tend to engage in fewer supportive interactions with their own party’s candidates and express negativity in their user profiles. Our results can inform the design of platform moderation mechanisms to support political figures countering online harassment.
Uzi Smadja, Max Grusky, Yoav Artzi, Mor Naaman. Understanding Reader Backtracking Behavior in Online News Articles. In Proceedings of the 2019 Web Conference (WWW 2019).
Rich engagement data can shed light on how people interact with online content and how such interactions may be determined by the content of the page. In this work, we investigate a specific type of interaction, backtracking, which refers to the action of scrolling back in a browser while reading an online news article. We leverage a dataset of close to 700K instances of more than 15K readers interacting with online news articles, in order to characterize and predict backtracking behavior. We first define different types of backtracking actions. We then show that “full” backtracks, where the readers eventually return to the spot at which they left the text, can be predicted by using features that were previously shown to relate to text readability. This finding highlights the relationship between backtracking and readability and suggests that backtracking could help assess readability of content at scale.
Xiao Ma, Justin Cheng, Shankar Iyer, Mor Naaman. When Do People Trust Their Social Groups? In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems (CHI 2019).
Trust facilitates cooperation and supports positive outcomes in social groups, including member satisfaction, information sharing, and task performance. Extensive prior research has examined individuals’ general propensity to trust, as well as the factors that contribute to their trust in specific groups. Here, we build on past work to present a comprehensive framework for predicting trust in groups. By surveying 6,383 Facebook Groups users about their trust attitudes and ex amining aggregated behavioral and demographic data for these individuals, we show that (1) an individual’s propensity to trust is associated with how they trust their groups, (2) smaller, closed, older, more exclusive, or more homogeneous groups are trusted more, and (3) a group’s overall friendship-network structure and an individual’s position within thats tructure can also predict trust. Last, we demonstrate how group trust predicts outcomes at both individual and group level such as the formation of new friendship ties.
Maurice Jakesch, Megan French, Xiao Ma, Jeff Hancock, Mor Naaman. AI-Mediated Communication: How the Perception that Profile Text was Written by AI Affects Trustworthiness. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems (CHI 2019).
CHI 2019 [Paper]
AI-Mediated Communication: How the Perception that Profile Text was Written by AI Affects Trustworthinesss
We are entering an era of AI-Mediated Communication (AI-MC) where interpersonal communication is not only mediated by technology, but is optimized, augmented, or generated by artificial intelligence. Our study takes a first look at the potential impact of AI-MC on online self-presentation. In three experiments we test whether people find Airbnb hosts less trustworthy if they believe their profiles have been written by AI. We observe a new phenomenon that we term the Replicant Effect: Only when participants thought they saw a mixed set of AI- and human-written profiles, they mistrusted hosts whose profiles were labeled as or suspected to be written by AI. Our findings have implications for the design of systems that involve AI technologies in online self-presentation and chart a direction for future work that may upend or augment key aspects of Computer-Mediated Communication theory.
Max Grusky, Mor Naaman, Yoav Artzi NEWSROOM: A Dataset of 1.3 Million Summaries with Diverse Extractive Strategies In Proceedings, North American Chapter of the Association for Computational Linguistics: Human Language Technologies. (NAACL 2018), June 2018, New Orleans, Louisiana.
NAACL 2018 [Paper]
NEWSROOM: A Dataset of 1.3 Million Summaries with Diverse Extractive Strategies
We present NEWSROOM, a summarization dataset of 1.3 million articles and summaries written by authors and editors in newsrooms of 38 major news publications. Extracted from search and social media metadata between 1998 and 2017, these high-quality summaries demonstrate high diversity of summarization styles. In particular, the summaries combine abstractive and extractive strategies, borrowing words and phrases from articles at varying rates. We analyze the extraction strategies used in NEWSROOM summaries against other datasets to quantify the diversity and difficulty of our new data, and train existing methods on the data to evaluate its utility and challenges. The dataset is available online at summari.es
Matthew V. Law, Mor Naaman, and Nicola Dell. ShareBox: Designing A Physical System to Support Resource Exchange in Local Communities. In Proceedings, Designing Interactive Systems. (DIS 2018), June 2018, Hong Kong.
DIS 2018 [Paper]
ShareBox: Designing A Physical System to Support Resource Exchange in Local Communities.
ShareBox is a technology probe that facilitates local borrowing and lending of physical items using a chatbot and smart lockbox. We describe the design of ShareBox and evaluate a set of design guidelines, including the physical-virtual nature of the system, anonymity and asynchronous exchange, through three short deployments. Our findings, based on semi-structured interviews with community members, highlight the potential and remaining challenges in our design.
Xiao Ma, Megan Cackett, Leslie Park, Eric Chien, and Mor Naaman. Web-Based VR Experiments Powered by the Crowd. In Proceedings, International World Wide Web Conference. (WWW 2018), April 2018, Perth, Australia
We build on the increasing availability of Virtual Reality (VR) devices and Web technologies to conduct behavioral experiments in VR using crowdsourcing techniques. A new recruiting and validation method allows us to create a panel of eligible experiment participants recruited from Amazon Mechanical Turk. Using this panel, we ran three different crowdsourced VR experiments, each reproducing one of three VR illusions: place illusion, embodiment illusion, and plausibility illusion. Our experience and worker feedback on these experiments show that conducting Web-based VR experiments using crowdsourcing is already feasible, though some challenges---including scale---remain. Such crowdsourced VR experiments on the Web have the potential to finally support replicable VR experiments with diverse populations at a low cost.
Emily Sun and Mor Naaman. A Multi-site Investigation of Community Awareness Through Passive Location Sharing. In Proceedings, ACM International Conference on Human Factors in Computing Systems. (CHI 2018), April 2018, Montreal, Canada
CHI 2018 [Paper]
A Multi-site Investigation of Community Awareness Through Passive Location Sharing
We aim to understand if and how passive location sharing might be socially beneficial for communities. We conducted a deployment of MoveMeant (see Sun et al. CHI 2017) and 45 interviews with users of the system and community leaders across three different communities. We discuss the issues that communities face and the potential positive and negative repercussions of surfacing passive location information.
Xiao Ma, Trishala Neeraj, and Mor Naaman. A Computational Approach to Perceived Trustworthiness of Airbnb Host Profiles. In Proceedings, International AAAI Conference on Web and Social Media. (ICWSM 2017), May 2017, Montreal, Canada
ICWSM 2017 [Poster]
A Computational Approach to Perceived Trustworthiness of Airbnb Host Profiles
Building on our previous work (see below by Ma et al. CSCW 2017), we developed a novel computational framework to predict the perceived trustworthiness of host profile texts in the context of online lodging marketplaces. We developed a dataset of 4,180 Airbnb host profiles annotated with perceived trustworthiness, and also provide insights into the linguistic factors that contribute to higher and lower perceived trustworthiness for profiles of different lengths.
Max Grusky, Jeiran Jahani, Josh Schwartz, Dan Valente, Yoav Artzi, and Mor Naaman. Modeling Sub-Document Attention Using Viewport Time. In Proceedings, ACM International Conference on Human Factors in Computing Systems. (CHI 2017), May 2017, Denver, USA
Website measures of engagement captured from millions of users, such as in-page scrolling and viewport position, can provide deeper understanding of attention than possible with simpler measures, such as dwell time. Using data from 1.2M news reading sessions, we examine and evaluate three increasingly sophisticated models of sub-document attention computed from viewport time, the time a page component is visible on the user display. Our approach supports refined large-scale measurement of user engagement at a level previously available only from lab-based eye-tracking studies.
Emily Sun, Ross McLachlan, Mor Naaman. MoveMeant: Anonymously Building Community Through Shared Location Histories. In Proceedings, ACM International Conference on Human Factors in Computing Systems. (CHI 2017), May 2017, Denver, USA
CHI 2017 [Note]
MoveMeant: Anonymously Building Community Through Shared Location Histories
We describe the design and implementation of MoveMeant, a system aimed to increase local community awareness through shared location traces. We report findings from interviews with residents in the Bronx, New York City who participated in a deployment of MoveMeant over a 6-week period.
Xiao Ma, Nazanin Andalibi, Louise Barkhuus, Mor Naaman. “People Are Either Too Fake or Too Real”: Opportunities and Challenges in Tie-Based Anonymity. In Proceedings, ACM International Conference on Human Factors in Computing Systems. (CHI 2017), May 2017, Denver, USA
CHI 2017 [Paper]
“People Are Either Too Fake or Too Real”: Opportunities and Challenges in Tie-Based Anonymity
We interviewed users of two anonymous social applications, Secret and Mimi, both of which allowed people to share information anonymously with friends. Our findings show that although users feel more comfortable sharing information on these “tie-based” anonymous applications, they are still concerned about being identified, and at the same time, engage in sometimes elaborate attempts to uncover the identities of others.
Nir Grinberg, Shankar Kalyanaraman, Lada Adamic, and Mor Naaman. Understanding Feedback Expectation on Facebook. In Proceedings, ACM International Conference on Computer-Supported Cooperative Work. (CSCW 2017), February 2017, Portland, USA
In this study, we quantitatively examined the factors contributing to feedback expectations and find that fulfilling expectations is linked to connectedness, an important ingredient for well-being. By conducting two large surveys on Facebook.com we find that people report higher expectations on posts they evaluated as more important, and to a lesser extent more personal. Expectations varied across people and friendships, most notably by recency of communication, geographical proximity, and the type of relationship (e.g. family, co-worker). The study provides a conceptual framework for thinking about feedback expectations in social media settings and a computational model for utilizing expectations in the design of social systems.
Xiao Ma, Jeff Hancock, Kenneth Lim Mingjie, Mor Naaman. Self-Disclosure and Perceived Trustworthiness of Airbnb Host Profiles. In Proceedings, ACM International Conference on Computer-Supported Cooperative Work. (CSCW 2017), February 2017, Portland, USA
CSCW 2017 [Paper]
Self-Disclosure and Perceived Trustworthiness of Airbnb Host Profiles
Best Paper Honorable Mention
We conducted the first large-scale mixed-method analysis of Airbnb host profiles, by categorizing the types of information hosts share in their profiles, and assessing how trustworthy their profiles are perceived to be. We found that hosts do not always follow the prompts by Airbnb website, but instead disclose information that reduces the uncertainty of anticipated future interaction. The language of hospitality, i.e., making direct promises to take care of guests, was found to be the most effective in establishing perceived trustworthiness.
Emily Sun, Ross McLachlan, Mor Naaman. TAMIES: A Study and Model of Adoption in P2P Resource Sharing and Indirect Exchange Systems. In Proceedings, ACM International Conference on Computer-Supported Cooperative Work. (CSCW 2017), February 2017, Portland, USA
CSCW 2017 [Paper]
TAMIES: A Study and Model of Adoption in P2P Resource Sharing and Indirect Exchange Systems
Peer-to-peer indirect exchange services, such as Peerby and NeighborGoods, do not seem to have been as widely adopted as direct exchange systems, such as Uber and AirBnb. Building upon the results of interviews with 37 residents of New York City, a survey with 195 respondents, previous technology acceptance models, critical mass theory, and prior research on peer economies, we propose a technology acceptance model for indirect exchange systems that includes generalized trust and ease of coordination.
Xiao Ma, Emily Sun, Mor Naaman. What Happens in happn: The Warranting Power of Location History in Online Dating. In Proceedings, ACM International Conference on Computer-Supported Cooperative Work. (CSCW 2017), February 2017, Portland, USA
CSCW 2017 [Paper]
What Happens in happn: The Warranting Power of Location History in Online Dating
We interviewed users of a novel mobile dating application, happn, which shows users the number of times they crossed path with potential matches. We show that users assigned significant meaning to the minimal cues available from location history information. At the same time, users have concerns about security and recognition by known others as a result of sharing personal location history.
Minsu Park, Mor Naaman, Jonah Berger. A Data-driven Study of View Duration on YouTube. In Proceedings, ACM International Conference on Web Logs and Social Media. (ICWSM 2016), May 2016, Cologne, Germany
Using two distinct YouTube datasets, we examine whether and how indicators of collective preferences and reactions are associated with view duration of videos.
Xiao Ma, Jeff Hancock, Mor Naaman. Anonymity, Intimacy and Self-Disclosure in Social Media. In Proceedings, ACM Conference on Human Factors in Computing Systems. (CHI 2016), May 2015, San Jose, USA
We conducted an experiment to study the relationship between content intimacy and self-disclosure in social media, and how anonymity and audience type (social ties vs. people nearby) moderate that relationship.
Nir Grinberg, P. Alex Dow, Lada A. Adamic, and Mor Naaman. Changes in Engagement Before and After Posting to Facebook. In Proceedings, ACM Conference on Human Factors in Computing Systems. (CHI 2016), May 2015, San Jose, USA
CHI 2016 [Paper]
Changes in Engagement Before and After Posting to Facebook
Best Paper Honorable Mention
The study identified a unique pattern of engagement that accompanies posting on Facebook. Using observational data analysis we show that after posting content, people visit the site more often, are more attentive to content from friends, and even interact more with friends content.
Raz Schwarz, Mor Naaman, Rannie Teodoro. Editorial Algorithms: Using Social Media to Discover and Report Local News. In Proceedings, ACM International Conference on Web Logs and Social Media. (ICWSM 2015), May 2015, Cambridge, England
We discuss CityBeat, a large-screen visualization that builds on machine learning techniques to expose hyper-local events in New York City from social media data. We deployed it with some national and local media and describe the gap between the journalistic needs and what our algorithm could provide.
Minsu Park, Ingmar Weber, Mor Naaman, Sarah Vieweg. Understanding Musical Diversity via Online Social Media. In Proceedings, ACM International Conference on Web Logs and Social Media. (ICWSM 2015), May 2015, Cambridge, England
Using data from users of Last.fm and Twitter, we design and evaluate a novel measure for computing diversity of musical tastes, and investigate its relationships with socioeconomic status and personal traits such as openness and degree of interest in music.
David Flatow, Mor Naaman, Ke Eddie Xie, Yana Volkovich, Yaron Kanza. On the Accuracy of Hyper-local Geotagging of Social Media Content. In Proceedings, the ACM International Conference on Web Search and Data Mining (WSDM 2015). March 2015, Shanghai, China.
Can we use geotagged social media to identify phrases that correspond to hyper-local geographic features? Does the geographic spread of such phrases differ between services or device types? (the answer to both is yes).
Funda Kivran-Swaine, Jeremy Ting, Jed Richards Brubaker, Rannie Teodoro, Mor Naaman.Understanding Loneliness in Social Awareness Streams: Expressions and Responses. Eighth International AAAI Conference on Weblogs and Social Media. (ICWSM 2014)
ICWSM 2014 [Paper]
Understanding Loneliness in Social Awareness Streams: Expressions and Responses
Content analysis of tweets expressing loneliness exposes some key patterns in how people talk about loneliness on Twitter, as well as when such posts are likely to receive a response from friends and followers.
Rannie Teodoro, Pinar Ozturk, Mor Naaman, Winter Mason, and Janne Lindqvist. The Motivations and Experiences of the On-demand Mobile Workforce. In Proceedings, ACM Conference on Computer Supported Cooperative Work. (CSCW 2014), February 2014
We performed semi-structured interviews with workers of on-demand mobile workforce applications that match users with physical world tasks to expose their motivations, practices, and experiences.
Jessa Lingel, Mor Naaman and danah boyd. City, Self, Network: Transnational Migrants and Online Identity Work. In Proceedings, ACM Conference on Computer Supported Cooperative Work. (CSCW 2014), February 2014
What are the challenges new immigrants face in online self-presentation and use of Facebook? We did a qualitative study to find out.
Grinberg, Nir, Mor Naaman, Blake Shaw, and Gilad Lotan. "Extracting Diurnal Patterns of Real World Activity from Social Media." In ICWSM. 2013.
We developed methods to identify how people talk about real-world activities on social media. By examining activities such as nightlife, food or shopping we peek at the fundamental rhythm of human behavior at a city level and observe how it was disrupted during Hurricane Sandy.
We're located at Cornell Tech. Come by and say hi!
2 W Loop Rd
Built by Hannah Xue and Huai-Che Lu. Designed by Emily Sun.
