Researchers in Bangladesh developed software that can accurately recognize a user’s emotional state up to 87 percent of the time, depending on the emotion.

Through techniques known as keystroke dynamics and text pattern analysis, the software can discern between emotions such as joy, fear, anger, sadness, disgust, shame and guilt. Keystroke dynamics relates to the relative timing of each keystroke, while text pattern analysis is concerned with which words a user chooses while typing. Researchers created a database of the most commonly used words and sentences associated with each of the seven target emotions.