You can also feel the mood when typing on your computer.

Computers have better performance in many aspects related to mathematics than humans, and some computing tasks that people cannot do can be left to computers. However, on the other hand, computers are far from comparable to humans-EQ. However, scientists are helping computers achieve this evolution. Through a program, computers will judge their emotions by recognizing the expression of human typing, this laid the foundation for development to catch up with human computers in all aspects. Scientists call this concept a "singularity ".

In a new study, the investigator asked a group of subjects to input a sample article and analyzed how they typed keys and features. The investigator tried to ask the computer to tell which of the emotions of these subjects were Joy, fear, anger, sorrow, disgust, shame, and guilt. As a result, among these emotions, the most accurate computer resolution is happy (87%) and angry (81% ).

"If we can build a system that allows computers to deal with people with 'emotics'-that is, to detect users' emotions, and adjust your behavior based on these emotions-people will feel more friendly and efficient when using computers." Researchers from the Islamic University of Technology in Bangladesh said. Kamrul Hasan from the systems and software lab and his colleagues posted the findings on Behavior & Information Technology.

The researchers mentioned that this emotional recognition system is suitable for online teaching and other aspects, A "high emotional intelligence" teaching system can change the interface, teaching style, and teaching content of individual students' emotional states.

In the first part of the study, 25 subjects aged 15-40 entered two paragraphs from Louis Caro's famous novel Alice in Wonderland, then I chose one of the following emotions that best describe my mood when typing: Joy, fear, anger, sorrow, disgust, shame, guilt, no feeling, or boredom. The last two are alternative options. If the subject finds that the first seven emotions are not consistent with their mood, they will be selected from these two options.

In the second part of the study, the investigator used software to collect text from the subject, and the subject was asked to submit his current mental state every 30 minutes. Options include: joy, fear, anger, sorrow, disgust, shame, guilt, or "None of above ". In this part of the experiment, the investigator did not specify the sample input by the subject, but instead collected the typing information for their daily use of computers. The investigator used a special background software to record each of the subject's buttons and recorded the time when the subject pressed and opened the buttons.

Then, the investigator extracted 19 key information from the data as analysis samples, including the subject's typing speed within five seconds, and the duration from pressing a key to releasing it, and so on. Next, the investigator used a standard database of words related to the seven emotions to analyze the sample input by the subject.

Experiment process. The investigator filters feature data from the source data of the two experiments and establishes a model through the software developed by the investigator. This model is used together with the word database to analyze the emotions of the input. Source: Research Paper

Jeon myunghoon, Assistant Professor of Applied Cognitive Science at the University of Michigan (Michigan scientific University), who was not involved in the study, commented, this new emotional Detection System "does not seem like a breakthrough ". "However, researchers are meaningful and promising in their efforts to integrate existing methods ."

However, he pointed out that the method used by the Institute to detect emotions from words has certain limitations. For example, unlike speech recognition technology or facial expression detection equipment, the typing task in the study was required by the subject. Therefore, if a person is really angry or sad, they may be unwilling or unable to complete the typing tasks required by the investigator because of these emotions.

However, he said, but the system is still an effective tool for online consulting. For example, in areas that are very popular with online consulting, a psychiatrist may be able to assess his or her mental state without communicating with the patient via voice. (Edit: Calo)

You can also feel the mood when typing on your computer.