Personality Assessment — Machine Learning for Greater Human Understanding

Introduction

As humans, we are capable of failure. Beyond this, we are able to celebrate successes and learn from our failures. We are capable of investigating and interacting with the objects around us, and learn their function through trial and error. In this way, we learn that stoves are hot, water is wet, and electric sockets should not be tampered with. In this way, we learn from out mistakes.

Machine learning can described as technology’s way of not only learning from mistakes, but improving because of them (Nicholson, 2019). The algorithms that allow this to happen are known ‘neural networks’ and they’re designed to mimic human thought patterns. Currently, machine learning is being implemented across a multitude of fields, including but not limited to: advertising, media, manufacturing, healthcare, finances, and business. The capabilities of machine leaning are endless, but one field of machine leaning that deserves more study is the field of human-computer interaction and the way that machines can help assist us in understanding ourselves.

“With ever-expanding memory capacities, the computer is rapidly becoming a giant storage and processing facility for recording and retrieving ‘bits of life’” (Dijck, 2005). In modern times, the internet offers a wealth of information. As described by Bleidorn, “People’s social media accounts (e.g., Facebook, Twitter) offer a compelling source of rich and intimate digital information… Along with the digital data that is automatically stored, machine learning researchers can collect more traditional psychological assessment data, such as scores on personality questionnaires” (Bleidorn, 2018). By examining the data collected ethically though social media platforms, we can gain a better understanding of human nature.

Statement of Topic

Machines aid humans daily, but we must look beyond the conventional means of assistance to understand the true potential of machine-assisted personality assessments. When utilized correctly, machine learning and consensual data collection can allow sociology researchers to learn more about what it means to be human.

For this research assignment, I plan to examine what methods are currently being used for data extraction — known also as ‘data scraping’ — and what this data is currently being used for. Beyond this, I will examine what ethical data scraping for personality assessment could potentially be used for for the future of psychology and sociology.

Proposed Research

As stated by Harry Zhang, “Technological advances now enable a considerable amount of implicit user data to be captured when a user makes use of technology in their lives” (Zhang, 2010). To gain a better understanding of the potential benefits of machine-assisted learning, particularly in the case of ML personality assessments, I plan to examine several areas of technology. To start, I’d like to examine ways machine learning is currently being used when it comes to understanding humanity within social media. We most commonly see this in marketing with the rise of targeted advertisements. Digital trackers known as ‘cookies’ are able to track user data in order to send them accurate advertisements (Kurkovsky, S. & Harihar, K., 2006 ). After watching a video about roller skating, for example, a viewer may be recommended an advertisement for a local roller skating venue. This form of machine-learning is often viewed as unethical by the masses (Zhang, 2010), but this does not mean all forms of ML meant to assist in human behaviors should be seen this way. Rather than focus on the ethicality of machine learning, I will focus on strictly on factual evidence and data.

After looking at the ways machine learning is currently being used in regards to understanding, humanity, I plan to look into what is currently in development and what we can expect in the future.

Audience

This report is aimed at the consumer and the casual user of modern technology, social media in particular. This report will help users understand where machine learning technologies currently

are and what the progress of ML technology will mean for the future.

Sources

For outside consultation on this topic, I will look to two of my professors. Prof. Jody Newcomer, my communications professor, has a knowledge of human interaction through communication as well as digital communication. My mathematics professor, Prof. Joseph M. Miller, has knowledge in technology. I will also be able to speak with Prof. Jennifer Orlando, who currently teaches computer science at my campus.

Bleidorn, W., & Hopwood, C. J. (2019). Using Machine Learning to Advance Personality Assessment and Theory. Personality and Social Psychology Review, 23(2), 190–203. https://doi.org/10.1177/1088868318772990 This article informs a large portion of my report as it directly examines the ways in with machine learning is currently being adapted to examine human behaviors through social media.

Leising, D., & Zimmermann, J. (2011). An Integrative Conceptual Framework for Assessing Personality and Personality Pathology. Review of General Psychology, 15(4), 317–330. While not focused on technology, this article examines the current ways in which we assess personality through text, the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) in particular. Looking at ways personality can be examined without the aid of

machine learning will create a framework for comparison when looking at ML personality assessments.

Nicholson, C. (2019). Machine Learning Algorithms. Retrieved November 8, 2019, from https://skymind.ai/wiki/machine-learning-algorithms. This website gives basic, easily digestible definitions for highly technical topics. Since the topic of machine learning can be a difficult one to understand, this website was particularly useful to me since it breaks down complicated terminology and techniques.

Van Dijck, J. (2005). From shoebox to performative agent: the computer as … Retrieved November 8, 2019, from https://journals.sagepub.com/doi/10.1177/1461444805050765. Van Dijck does not examine machine learning in this article as much as he examines the concept of digital storage and human augmentation, which plays a role in the greater topic of social media.

Conclusion

When collected ethically, the data we collect about the masses can be used to improve machinery and used to better understand ourselves.