A Counseling System of Multiple Intelligence Theory Combined With kNN Classification Algorithm

Abstract

Choosing the right career is always a big issue, an important concern for everyone. To have a job, which is suitable for you, firstly you must look at yourself, called the self, and you should be aware of what the self is then you can promote the strength of your own self and avoid your weakness. To help discover more about yourself, during researching and studying, we come up with the idea that we would propose a career counseling system based on Howard Gardner's theory. The system uses the theory of multiple intelligences (Abenti & Daradoumis, 2020) which is combined with the K-nearest neighbors (KNN) (Tang, Ying; Tang, Ying; Hare, Ryan; Wang, Fei-Yue;, 2020) algorithm to assist people and to give out a suitable suggestion about career path for them. We use the results of the eight intelligences retrieved from the KNN classification algorithm to give users the consulting for their career paths. This system is built with a dataset based on 56 multiple-choice questions. These include 48 multiple choice questions based on Howard Gardner's theory of multiple intelligences (Bravo, Leonardo Emiro Contreras; Molano, Jose Ignacio Rodriguez; Trujillo, Edwin Rivas, 2020), (businessballs, 2017) and 8 multiple choice questions which are the labels of the classifier. We divided the dataset into 8 subsets corresponding with 8 Intelligences defined by Howard Gardner with the collected dataset. In each subset, we build the KNN classifier model using KNN classification algorithm. This processing of 8 subsets come out with the results accuracy for the 8 Intelligences: linguistic intelligence (80.95%), logical-mathematical intelligence (82.14%), musical intelligence (96.43%), bodily-kinesthetic intelligence (82.14%), spatial-visual intelligence (82.14%), interpersonal intelligence (89.29%), intrapersonal intelligence (88.1%), existential intelligence (78.57%). With the outcome of 8 models, we have tested with 5 students and compared them to their actual intelligences. The comparison results tell us about the valuable potential in career path of the proposed counselling system, the advantages of this combination between Multiple Intelligence and KNN classifier.

The level of perceived efficacy from teachers to access AI-based teaching applications

“…The ability of the AI education application and learning mechanism is based on the stages of diagnosis, prediction, guidance, and prevention of 'precision education' (du Boulay, 2016;Fahimirad & Kotamjani, 2018;Guilherme, 2019). Nqoc and Van (2021) propose an AI system platform that detects students' sexual orientation and the best learning style through AI, which can use Bloom's thinking levels and Gardener's categories of multiple intelligence, and the school can detect students' aptitudes and the best learning methods to help guide them in their learning (Mozer et al, 2019). Teachers could use AI education applications to combine information and communications technology (ICT) and AI technologies, to construct a teaching platform, to provide virtual or real teaching situations, and to improve students' abilities (Nakata, 2019;Simonov et al, 2019).…”