|
|
 |
| |
|
|
|
Level zero: Emotional Intelligence in Artificial Intelligence- Exploring the Impact of Positive and Negative emotions on Machine Learning |
|
|
|
PP: 45-50 |
|
|
|
Author(s) |
|
|
|
Mahmood Saeed Mustafa Alalawi,
Nandita Sengupta,
Bassam Alhamad,
|
|
|
|
Abstract |
|
|
| The aim of this paper is to explore the pros and cons of machine learning in emotional intelligence. It uses positive emotions, presented in Barbara Fredrickson’s Theory ‘Broaden-and-Build.’ The said theory is the following: expand cognitive capacities and build psychological resources, enhance machine learning performance by broadening decision-making capabilities, improving learning rates in supervised learning, and aiding in pattern recognition in unsupervised learning environments [1], [2]. However, a concern was raised, negative emotions can narrow cognitive focus, which results in confused state of mind, poor decision-making, and reduced adaptability in AI systems. Research conducted by Schuller et al presented the concept of how emotionally charged data can lead to "poor learning". It would also minimize AI’s accuracy in interpretation of human emotions [3]. Hence, there is an urgent need to classify positive and negative emotions with high accuracy, removal of any negative emotion from the database. This will result in higher performance of AI system can be enhanced for decision making in the field of emotional intelligence. When creating an AI system, it is important to add a Balancing emotion where the system could function in an intelligent and effective way throughout complex, human-interactive environments. This Level Zero paper lays the groundwork for future research towards classifying positive and negative emotions efficiently which results in evaluating emotions in AI, aiming to optimize adaptability and performance in real-world applications. |
|
|
|
|
|
|
|
