Emotional Analysis Beyond the Usual Suspects

Emotion detection technology uses mainly two techniques:

  • Computer vision: AI’s analysis of digital images and videos to precisely identify facial expressions.
  • Machine learning algorithms: AI self-learning model used to analyze and interpret the emotional content of those facial features.

Its value comes from expanding its reach and going beyond its earliest database analysis. By showing thousands of emotional images to this algorithm, it can recognize other faces outside of its original dataset and create an emotional map for any human face.

The applications of this technology are endless. Companies are using it to help drivers have a better driving experience, test people’s experience while playing video games, help doctors assess the well-being of their patients, and use it to fight crime.

Criminal behavior under scrutiny

Our behavior has a strong unconscious component where emotions are in the driving seat. Emotions can elicit the most wonderful actions but also the most horrific behavior. Emotional-driven criminal activity oscillates between different levels of intensity. 

According to criminal psychologists Andrews and Bonta (2010), there are 9 levels of negative emotions engaged in criminal behavior:

  1. Bothered
  2. Annoyed
  3. Indignation
  4. Frustration
  5. Infuriated
  6. Hostile
  7. Wrath
  8. Fury
  9. Rage

Only after it reaches the sixth level (hostile) or above, a person will be consumed by emotion and lose control over their behavior. It reaches a point where the offender can’t correctly assess the consequences of their actions and restrain themselves without a third party involved.

People surpass a limit that makes them unlikely to snap out of it on their own. If no one is there to intervene, the consequences can be catastrophic. This is what they call “blinded by rage”.

A group of Chinese researchers wanted to see if they could use emotion recognition paired with smart cameras across cities to analyze people’s emotions without direct contact with them. By combining big data on emotions with crime data researchers are laying the groundwork to be able to predict the probability of crime types inside cities. Their study will give the police information about spots that are more probable of crime activity and stay one step ahead of lawbreakers. 

Just as emotional AI could be used to help recruiters in their selection process, this tech can also be used to help police and detectives during their interviews. By looking at the witnesses' and suspects’ reactions, emotional analysis can aid officers to interpret the answers to the key interview questions. Behavioral analysis interviews, which focus on both verbal and non-verbal cues, are already common in police practice. Thus, emotional recognition software can be a complement to the practices already being used by the police.

The EU funded a project called iBorderCTRL to create a faster and more thorough check system for third-country nationals at the borders using non-verbal micro-gestures. The experimental runs showed an accuracy of more than 70% on truthful and deceptive statements, resulting in higher accuracy than the average of expert officials and non-expert people. Deploying emotional AI systems, such as this deception detection software, can be the next frontier. Even the father of emotion detection, Paul Ekman, used to help train the CIA, FBI, Customs and Border Protection and the TSA back in the day.

Inside the mind of serial killers

The video allows us to see Dahmer’s emotional reactions to his different statements. 

When he says that his motivation to kill is to keep his victims nearby, facial recognition helps us see beyond the words. We see the aftertaste of his claims. In this case, higher levels of happiness while thinking about his deepest desires. And you’re able to do the same with the rest of the content. 

Facial recognition software can analyze images or videos of killers in order to identify patterns in their facial expressions or body language. This could include identifying specific expressions or gestures that are associated with different emotions or identifying changes in the killer's emotional state over time.

It is important to note that while Emotional AI can provide valuable insights in criminal investigations, the results should be used in conjunction with other forms of evidence and analysis, and not rely solely on it.

The global market for emotion detection 

With the widespread of remote work thanks to the coronavirus, the business of non-invasive assessment of people’s emotions like facial expressions has been booming. According to Markets and Markets, the market size of this industry is expected to grow from $23.6 billion in 2022, to $43.3 billion in 2027, at an impressive CAGR of 12.9%.

Currently, major growth for emotional AI across the world comes from the car industry (e.g. creating a wholesome driving experience), service providers (e.g. socially intelligent artificial agents), and the metaverse (e.g. boosting the VR headset experience).

Is your company ready to adopt this technology?

Let's Talk
Contact Us