Future-Generation Personality Prediction From Digital Footprints

Scope and Objective

Personality traits are generally referred to as relatively stable patterns of thoughts, feelings, and behaviours that have been associated with a wide range of important life outcomes and choices. Specifically, personality traits have repeatedly been related to the individual (e.g., well-being, psychopathology), interpersonal (e.g., relationship satisfaction), and social-institutional outcomes (e.g., occupational choices, job success). Hence, in the recent years, there has been a massive increase in the interest to develop models which use online data on human behaviour and preferences (i.e., digital footprints) to automatically predict an individuals’ personality traits.

Advances in consumer electronics (e.g., smartphones, wearables) and the subsequent development of mobile sensing methods have facilitated the collection of highly detailed, multi-dimensional data on behaviours and situations. Social media gives users the opportunity to build an online persona through posting of content such as text, images, links or through interaction with others. The way in which users present themselves is a type of behaviour usually determined by differences in demographic or psychological traits. The behavioural residue harvested from websites and online social media platforms is also another valuable source of data on behaviour linked to personality traits. Hence, automated personality prediction has important practical applications in diverse areas ranging from recommendation systems, computational advertising, marketing science, job screening to aiding in psychological counselling, intervention and therapy, enhanced human-computer interaction, etc. It is also interesting to see the benchmarking studies with regard to the sensitivity of the data, prediction performance of the models, and cost for businesses to securely store the data, models, react to GDPR (General Data Protection Regulation) requests, etc.

However, it should be noted that automated personality prediction is a controversial topic and serious concerns have been raised with regard to implications for individual privacy and the conception of informed consent. While the performance of these models is not high enough to allow for the precise distinction of people based on their traits, predictions can still be “right” on average and be utilized for digital mass persuasion and for personalization efforts. Focusing research on explainable models, rather than just using them as black-box personality predictors can help to bridge the seemingly distant fields of computational personality detection and personality research in psychological science.

The primary objective of this special issue is to bring together diverse, novel and impactful work on personality prediction in one place, thereby accelerating research in this field.

Topics of Interest

The topics of interest for this special issue include, but are not limited to:

• Personality prediction from multimodal and diverse input modalities (e.g., audio, video, text) along with new approaches effectively fusing features extracted from multiple sources (for e.g., using heterogeneous data collected from different devices)
• Deep learning-based approaches (e.g., CNNs, GANs for data augmentation, deep RL, etc.)
• Machine learning for automated personality prediction from user behaviour. For example:
•     ○ Social media interaction
•     ○ Author profiling based on writing
•     ○ Consumer device usage patterns (e.g., wearable devices, smartphones, etc.)
• Studies using interpretable/explainable machine learning
• Studies exploring the associations’ correlation between personality and behaviour and studies which show new associations with real-world behaviours in large samples. (from the viewpoint of broadening the scope to personality psychologists and stimulating the creation of new models)
• Predictive modelling
• Optimized models for deployment on edge devices (lightweight, low latency models)
• Novel datasets (e.g., computer gaming social data) for various personality measures (e.g., Big-Five, MBTI, etc.)
• Domain-specific personality prediction (e.g., Internet personality, brand personality)
• Privacy-preserving approaches for personality prediction (e.g., federated learning models)
• Methods for reliable personality labelling
• Unsupervised methods for personality prediction
• Novel methods in deploying personality models for large-scale use

Important Dates

Submission portal opens: November 15th 2020

Deadline for paper submission: February 15th, 2021

Reviewing: Continuous basis

Revision deadline: July 15th, 2021

Latest acceptance deadline for all papers: October 1st, 2021

Guest Editors

• Yash Mehta
  University College London, UK
  y.mehta@ucl.ac.uk

• Clemens Stachl
  Stanford University, USA
  stachl@stanford.edu
   
•  Joseph T Yun
  University of Illinois Urbana-Champaign, USA
  jtyun@illinois.edu
   
• Konstantin Markov
  University of Aizu, Japan
  markov@u-aizu.ac.jp
   
• Björn W. Schuller
  Imperial College London, UK
  bjoern.schuller@imperial.ac.uk

Evaluation Criterion

• Novelty of approach (how is it different from what exists already?)
• Technical soundness (e.g., rigorous model evaluation)
• Impact (how does it change our current state of affairs)
• Readability (is it clear what has been done)
• Reproducibility and open science: pre-registrations if confirmatory claims are being made (e.g., via osf.io), open data, materials, code as far as ethically possible.

Submission Guidelines

The submission portal will be accepting submissions to our special issue from November 15, 2020 to February 15, 2021. Please select the article type “VSI: Personality Prediction“, while submitting your manuscript. The guide for authors and the link to submit your manuscript is available on the Journal’s homepage. Each manuscript can have no more than 14 pages (plus one page after revision) in double-column format, including all its content (e.g., figures, references, appendix, and so on). All submissions deemed suitable to be sent for peer review will be reviewed by at least two independent reviewers. Whenever suitable, we will aim to have at least one reviewer from the ML community and one from core personality research in psychology. Once the manuscript is accepted, it will go into production and will be simultaneously published in the current regular issue and pulled into the online Special Issue. Articles from this Special Issue will appear in different regular issues of the journal, though they will be clearly marked and branded as Special Issue articles. Again, please ensure you read the guide for authors before writing your manuscript to prevent desk rejection.

Note: Submissions are reviewed on a continuous basis and published online as soon as accepted.

Editors’ bios

● Yash Mehta is currently a researcher at the Gatsby Computational Neuroscience Unit at University College London. One of his most representative works in personality detection is the literature survey, ‘Recent Trends in Deep Learning-Based Personality Detection’. In addition to machine learning-based personality detection, he is also very interested in biologically plausible learning rules to train neural networks and Neural Architecture Search (NAS). He will soon be joining Frank Hutter’s famous AutoML group at the University of Freiburg, putting his brain cells to work on fundamental and applied NAS research. Previously, he was a software developer at Amazon and like all coders of course, enjoys coffee and pretending he’s smart by searching for stuff on Stack Overflow.

● Clemens Stachl is a postdoctoral scholar at the Communication Department at Stanford University. His work is focused at the prediction, description, and explanation of psychological phenomena (i.e., personality traits) from digital behavioral footprints and at the collection of these data with mobile sensing techniques. Another aspect of his work deals with the personalization of user interfaces and digital systems with regard to individual differences (i.e., personality traits).

● Joseph T. Yun is a Research Assistant Professor of Accountancy and Director of Data Science Research Services at the Gies College of Business at the University of Illinois Urbana Champaign. He also has affiliations with the Charles H. Sandage Department of Advertising, the National Center for Supercomputing Applications, Illinois Informatics, and the Grainger Engineering Center for Health Informatics. Yun’s research is primarily focused on novel data science algorithms, user-centric analytics systems, and societal considerations of AI-based advertising and marketing (e.g. privacy, ethics). His research includes both the development of personality detection models from social media data as well as innovation in hosting of those models for the purposes of reproducible research. Yun is also the principal investigator of the Social Media Macroscope, which is an open research environment for social media analytics (www.socialmediamacroscope.com). Recently, Dr. Yun founded a start-up company in Research Park, Marketing AI Partners (www.marketingaipartners.com), that focuses on bringing AI and data science innovation to the marketing industry.

● Konstantin Markov is a full Professor of Information Science at the University of Aizu, Japan. He has been a regular visiting Professor at the Institute of Statistical Mathematics, Tokyo, Japan. Prior to becoming a faculty member of the University of Aizu, he was a senior research scientist at the Spoken Language Communication Lab of the Advanced Telecommunication Research (ATR) Institute International, Kyoto, Japan. His research is focused mainly on statistical signal processing, machine and deep learning, and AI in general with applications to audio, natural language and image processing.

● Björn W. Schuller received his diploma, doctoral degree, habilitation, and Adjunct Teaching Professor in Machine Intelligence and Signal Processing all in EE/IT from TUM in Munich/Germany. He is Full Professor of Artificial Intelligence and the Head of GLAM at Imperial College London/UK, Full Professor and Chair of Embedded Intelligence for Health Care and Wellbeing at the University of Augsburg/Germany, co-founding CEO and current CSO of audEERING – an Audio Intelligence company based near Munich and in Berlin/Germany, and permanent Visiting Professor at HIT/China amongst other Professorships and Affiliations. Previous stays include Full Professor at the University of Passau/Germany, and Researcher at Joanneum Research in Graz/Austria, and the CNRS-LIMSI in Orsay/France. He is a Fellow of the IEEE and Golden Core Awardee of the IEEE Computer Society, Fellow of the ISCA, Fellow of the BCS, President-Emeritus of the AAAC, and Senior Member of the ACM. He (co-)authored 900+ publications (30k+ citations, h-index=83), is Field Chief Editor of Frontiers in Digital Health and was Editor in Chief of the IEEE Transactions on Affective Computing, served for 17 journals as Associate Editor, Guest Edited 26 Special Issues amongst manifold further commitments and service to the community. His 30+ awards include having been honoured as one of 40 extraordinary scientists under the age of 40 by the WEF in 2015. He served as Coordinator/PI in 15+ European Projects, is an ERC Starting Grantee, and consultant of companies such as Barclays, GN, Huawei, or Samsung.