Hi YouTube!

Abstract

Despite the evidence that social video conveys rich human personality information, research investigating the automatic prediction of personality impressions in vlogging has shown that, amongst the Big-Five traits, automatic nonverbal behavioral cues are useful to predict mainly the Extraversion trait. This finding, also reported in other conversational settings, indicates that personality information may be coded in other behavioral dimensions like the verbal channel, which has been less studied in multimodal interaction research. In this paper, we address the task of predicting personality impressions from vloggers based on what they say in their YouTube videos. First, we use manual transcripts of vlogs and verbal content analysis techniques to understand the ability of verbal content for the prediction of crowdsourced Big-Five personality impressions. Second, we explore the feasibility of a fully-automatic framework in which transcripts are obtained using automatic speech recognition (ASR). Our results show that the analysis of error-free verbal content is useful to predict four of the Big-Five traits, three of them better than using nonverbal cues, and that the errors caused by the ASR system decrease the performance significantly.

Automated video interview personality assessments: Reliability, validity, and generalizability investigations.

“…Further, nonverbal behavior (e.g., open posture, eye contact) influences interviewer judgments of all Big Five interviewee traits (DeGroot & Gooty, 2009). In terms of applying AVIs to assess Big Five personality traits, prior studies in computer science found that AVI-PAs can converge strongly with interviewer-reported traits when crossvalidated on holdout data drawn from the same sample (see Table 2 for a review of AVI-PA studies; Biel et al, 2013;Chen et al, 2016Chen et al, , 2018Nguyen et al, 2014;Nguyen & Gatica-Perez, 2016;Ponce-López et al, 2016). For example, Chen et al (2018) found that their machine learning models accurately classified interviewees as having high or low standing on interviewer-reported traits (macro F-1 score, or the harmonic mean of precision and recall, was approximately .80 for each trait).…”

Evoking presence in vlogging: A case study of U. K. beauty blogger Zoe Sugg

“…Additionally, verbal, audio-visual and facial elements expressed by vloggers connected to an audience's impression of a vlogger's personality (Biel, et al, 2013). Verbal and facial elements could affect social presence such as intimacy and immediacy, as concluded by Short, et al (1976), while audio-visual elements might affect physical presence, a sense of "being there".…”

Nonverbal Social Sensing: What Social Sensing Can and Cannot Do for the Study of Nonverbal Behavior From Video

“…For instance, to predict who gets hired for a job, the machine can first extract a set of specific NVB and then link it to the ground truth of hiring decisions. Another example is training a machine to predict social skills or personality (Biel et al, 2013;Muralidhar et al, 2018;Rasipuram and Jayagopi, 2018) or emotions (Ahn et al, 2010) based on previously extracted nonverbal cues. Again, the ground truth has to be measured (e.g., human coders assessing the personality of the people in the video or a self-report of their personality).…”

“…Concerning ethical algorithms, we suggested that social scientists, interested in the use of nonverbal social sensing systems, should be well-informed about policies related to artificial intelligence (AI). For instance, in Europe, a group of experts was commissioned to work on ethical guidelines for AI (Biel et al, 2013). The requirements for the so-called trustworthy AI are (1) human agency and oversight, (2) technical robustness and safety, (3) privacy data and governance, (4) transparency, (5) diversity, non-discrimination, and fairness, (6) environmental and societal wellbeing, and ( 7) accountability.…”