Personality Classification by Applying k-Means Clustering

Assem Talasbek, Azamat Serek, Meirambek Zhaparov, Seong-Moo Yoo, Yong-Kab Kim, Geun-Ho Jeong

2020

Closed access Metadata (CSL-JSON)
DOI 10.1109/icaiic48513.2020.9065244
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This paper describes personality classification by applying k-means clustering machine learning algorithm. Several previous studies have been attempted to predict personality type of human beings automatically by using various machine learning algorithms. However, only few of them have obtained good accuracy results. To classify person into personality types Jungian Type Inventory was used. Our testing results showed that k-means model has 107 inertia value, which is a good number for unsupervised learning model as an interim result. As the result, we divided the data into 16 clusters, which can be considered as personality types. We continue this research with trying to collect large data.