Can AI understand the sophisticated human emotions related to revenge caused by jealousy?
What’s happening in this image?
The cheerleader falls for a guy on the football team, but he dumps her because he has feelings for another girl. At first the cheerleader is upset, but she then devises a plan to get revenge on the boy that broke not only her heart but her pride as well. She decides to tell everyone else in the school that she dumped him. She also tells him that the new girl he likes is flirting with someone else. This leaves her ex publicly embarrassed and upset. Successful in getting her revenge on the guy who hurt her pride, the cheerleader walks away with a smug smile on her face.
Basics of Revenge
People seek revenge for a number of reasons, such as when they’ve been hurt, betrayed, overly jealous, attacked, or are experiencing some sort of loss. People often desire revenge because they believe it will relieve some of the impact of the aforementioned; exacting revenge makes one feel better. The emotions underlying revenge, not only before, but also after, are varied. One often feels anger, envy, sadness and/or shame before seeking revenge. But in the moments right after exacting revenge, the same person may feel accomplished and even feel like they are better than the person they are getting revenge on.
M vs Microsoft Azure Analysis
M perfectly describes the complex emotional states that the cheerleader feels in that moment. M identifies negative "Contempt" as the girl's primary emotion. She does not accept the fact that she was rejected: “How dare he reject me?” She wants revenge because she feels that her ex and his new girl deserves this revenge that she will inflict on them.
This is visible in the tight-lipped smile she has on her face with her chin raised up, as well as in the hidden emotions from her face. M identifies "Disgust" as the secondary emotion. This could be for two reasons: (1) because of her strong dislike and hatred toward her ex and this new girl, or (2) she feels disgusted at herself for her evil actions and intentions. It is not wholly clear why M has identified disgust as the secondary emotion, but it’s presence is relatively strong.
Finally, M identifies "Sad" as the tertiary emotion. This is possibly because when a person commits an evil act, they can experience some degree of guilt or sadness related to their actions. M was able to capture not only the complexity of the human emotions related to evil-intentioned revenge caused by jealousy, but also understand it in the context of the end of a relationship.
Microsoft Azure, on the other hand, identifies the emotions as 54% neutral and 46% happy. Interestingly, the emotions captured by Microsoft Azure are not not even significant according to M’s analysis, and are among the lowest levels of emotion detected by M.
Thus, M detects the emotions based on this face as negative emotions. But Microsoft detects positive emotions based on the same face. Now it is up to you, our readers, to decide what the correct interpretation is.
Your LIKE means a lot to us!
Follow Project M on:
Company website: http://www.ipmdinc.com
Medium: Project M
Facebook: Project M
*As of October 1, 2019, the Project M team has devoted 70,000 hours to the AI platform, M. The sample input data we are using is the pure testing data that is completely new to M and Microsoft (assumed) and has never been used to train M, so this comparison is a fair trial between M and Microsoft. We appreciate Microsoft, the leader of the industry and emotional AI sector, for allowing the general public to test their testing site for identifying human emotions based on facial expressions.