Racial Bias in Customer Service: Evidence from Twitter

Detecting and reporting systemic racial bias is an essential step toward the eradication of racial discrimination in our society. Doing so not only requires society members to voice and share their anecdotal experiences, but also relies on researchers to document systematic statistical evidence of racial bias. This paper documents the first large-scale evidence of business-to-customer racial bias on digital platforms on which the perpetrators are individual employees who act on behalf of a company and the victims are customers. This is in contrast to existing studies of racial bias on digital platforms that focus on peer-to-peer marketplaces in which both the perpetrators and the victims are individuals acting independently and on their own behalf. By analyzing more than 57,000 social media customer complaints to U.S. airlines and leveraging a variety of analytics techniques, including text mining and facial recognition, we present quantitative evidence that African American customers are less likely to receive a response when they complain than otherwise similar White customers. Furthermore, our deep learning–based falsification test shows that the bias is absent without the race-revealing visual cue. This study offers a practical yet powerful recommendation for companies: conceal all customer profile pictures from their employees while delivering social media customer service.

Morphometric analysis of pre and post operative facial angles following rhinoplasty

Facial aesthetic surgery requires a thorough preoperative analysis of face and identification of problems to frame a comprehensive surgical plan. The conceptions of an attractive face must be analyzed beforehand. Powell and Humphrey had defined facial angles and had formulated range which would be perceived as attractive. To compare the change in pre- and post-operative nasolabial, nasofrontal and nasofacial angles in patients undergoing Rhinoplasty. A prospective observational study was conducted from October 2014 to October 2019 in ENT department of a tertiary care hospital. Twenty-one patients undergoing Rhinoplasty and fulfilling the inclusion criteria were considered as study population. Their pre- and post-operative profile photographs were taken for morphometric analysis. The nasolabial, nasofrontal and nasofacial angles were measured and compared, before and one month after the surgery in profile pictures. A significant change in the nasolabial and nasofacial angles were seen following Rhinoplasty. The mean nasofrontal angle showed an increase following Rhinoplasty but was not found significant. Rhinoplasty results in significant change in nasofacial and nasolabial angles and in expert hands, is expected to restore the various facial angles to their acceptable range.

Profile Pictures in the Digital World: Self-Photographs Predict Better Life Satisfaction

Profiles pictures as online identities represent an extension of the user's self in the digital world. Changes in self-representation are responsible for reduced well-being in individuals in the offline world. However, whether profile picture selection predicts the well-being of internet users is unknown. To address this question, we tested the relationship between the type of profile picture (e.g., self-photographs or other pictures) used on social media and the life satisfaction of internet users, accounting for gender and personality traits that have been thought to relate to the selection of profile pictures. The results showed that individuals using self-photographs as profiles reported a higher level of life satisfaction compared to individuals using other pictures as profiles. This effect was influenced by gender, openness, and extraversion. Hierarchical regression and moderation analyses revealed that openness and profile type interacted to predict life satisfaction in women, while openness and profile picture independently predicted life satisfaction in men. Furthermore, extraversion directly predicted life satisfaction in both men and women. These results indicate that the consistency between one's online and offline self-representation may characterize internet users' well-being, with potential implications for digital wellness.

Profiling the Self in Mobile Online Dating Apps: a Serial Picture Analysis

Profiles in the widely used phenomenon of mobile online dating applications are characteristically reduced to condensed information mostly containing one or a few pictures. Thus, these picture(s) play a significant role for the decision-making processes and success, supposedly holding vital meaning for the subjects. While profile pictures in social media are omnipresent and some research has already focused on these pictures, especially selfies, there has been little attention with regards to the actual self-presentation when mobile online dating. In this paper, we show the results of a reconstructive serial analysis of 524 mobile online dating profile pictures investigating how subjects present themselves in the context of a mobile online dating app. This context is highly specific and characterized by continuous and dichotomous judgments by (unknown) others, unseen competition, and permanent validation of the self. Despite the conceivable multitude of possible self-presentations, our analysis led to eight clear types of self-presentation. Contemplating on subject’s good reasons for presenting the self as one of many and not as varied and unique when mobile online dating, we refer to the discourse of the private self (Gergen, The saturated self: Dilemmas of identity in contemporary life, Basic Books, New York, 1991; Rose, Governing the soul: Shaping of the private self, Free Association Books, London, 2006) and to (Holzkamp, 1983. Grundlagen der Psychologie. Frankfurt a.M.: Campus.) concept of restrictive and generalized agency in a context of socially constituted norms.

Coming to Grips with Age Prediction on Imbalanced Multimodal Community Question Answering Data

For almost every online service, it is fundamental to understand patterns, differences and trends revealed by age demographic analysis—for example, take the discovery of malicious activity, including identity theft, violation of community guidelines and fake profiles. In the particular case of platforms such as Facebook, Twitter and Yahoo! Answers, user demographics have impacts on their revenues and user experience; demographics assist in ensuring that the needs of each cohort are fulfilled via personalizing and contextualizing content. Despite the fact that technology has been made more accessible, thereby becoming evermore prevalent in both personal and professional lives alike, older people continue to trail Gen Z and Millennials in its adoption. This trailing brings about an under-representation that has a harmful influence on the demographic analysis and on supervised machine learning models. To that end, this paper pioneers attempts at examining this and other major challenges facing three distinct modalities when dealing with community question answering (cQA) platforms (i.e., texts, images and metadata). As for textual inputs, we propose an age-batched greedy curriculum learning (AGCL) approach to lessen the effects of their inherent class imbalances. When built on top of FastText shallow neural networks, AGCL achieved an increase of ca. 4% in macro-F1-score with respect to baseline systems (i.e., off-the-shelf deep neural networks). With regard to metadata, our experiments show that random forest classifiers significantly improve their performance when individuals close to generational borders are excluded (up to 20% more accuracy); and by experimenting with neural network-based visual classifiers, we discovered that images are the most challenging modality for age prediction. In fact, it is hard for a visual inspection to connect profile pictures with age cohorts, and there are considerable differences in their group distributions with respect to meta-data and textual inputs. All in all, we envisage that our findings will be highly relevant as guidelines for constructing assorted multimodal supervised models for automatic age recognition across cQA platforms.

Relationship of natural occlusal plane with different anatomical landmarks

Objectives: To evaluate the parallelism of natural maxillary occlusal plane with inter-pupillary line and ala-tragus line among dentate subjects. To evaluate anatomic relationship of natural mandibular occlusal plane with retromolar pad among dentate subjects Methodology: Front and profile photographs of 109 students of Fatima Jinnah Dental College with age ranging from 20 to 28 years were taken while holding the camper’s plane against the maxillary occlusal plane. The study was completed in six months. Photographs were imported in software AutoCAD(2017). An interpupillary line was drawn and angle with Camper’s plane was measured. On both profile pictures, line were drawn from base of the ala to the three different points on tragus (superior, middle and inferior). The angle between ala-tragus line and camper’s plane were measured. Intra-orally, height of the mandibular occlusal plane in relation to the retromolar pad was evaluated using a stainless steel scale. Data was analyzed using SPSS Version 23. Student’s paired sample t test and Pearson’s correlation coefficient were applied. Results: Horizontal parallelism of occlusal plane with inter-pupillary line was observed (mean angle 1.17 ±1.27). The angle between the occlusal plane and the inferior ala-tragus line was 4.25 and 4.50 degrees on right and left side, respectively. Intraorally, mandibular occlusal plane coincided with inferior (44%) and middle third (44%) of the retromolar pad. Conclusions: Inter-pupillary line and retromolar pad area (middle & inferior third) should be used as a guide in determination of plane of occlusion. Continuous...