New Program Can Identify Emotion from Typing Style

January 5, 2015

Unlocking Word Meanings
 今日の単語・フレーズ

Read the following words/expressions found in today’s article. 

1. estimate /ˈɛs təˌmeɪt/ (v.) – to provide a calculation that is almost accurate based on a given data
Example: The weather reporter estimated that tomorrow will be sunny based on today’s temperature.

2. keystroke /ˈkiˌstroʊk/ (n.) – the act of using the keyboard by pushing its keys
Example: She can do 50 keystrokes in just five minutes.

3. spontaneous /spɒnˈteɪ ni əs/ (adj.) – doing something instinctively without being aware of it
Example: People often do a spontaneous frown when they are trying to concentrate on something.

4. open new doors /ˈoʊ pən nu, nyu dɔrz, doʊrz/ (idiom) – to introduce possibilities
Example: Discovering new planets opens new doors to finding other life forms in space.

5. oriented /ˈɔr i ən tɪd, iˌɛn tɪd/ (adj.) – being adjusted to the needs and/or functions of something
Example: The cartoon show that talks about science is both scientifically oriented and kid-oriented.


Article
 ニュース記事

Read the text below.
A study developed a new computer program that can estimate human emotion through a person’s style of typing.

The study was conducted by researchers from the Islamic University of Technology in Bangladesh. Their goal was to use a technological process that can detect emotion from a person’s computer keystrokes. To do this, the researchers developed a program using various computer formulas.

This program analyzes the characteristics of a person’s keystroke. These characteristics include speed, rhythm of typing pattern, and pauses between pressing and releasing a key. The researchers then examined the program’s data alongside the questionnaires filled out by the participants to identify any of the seven basic emotions, some of which are anger, joy, and sadness.

The program was able to accurately detect two emotions. Based on results, the program correctly identified “joy” 87 percent of the time and “anger” 81 percent of the time. Sadness was the least recognized, with only 60 percent accuracy. This was because participants were less willing to indicate their true emotions to the study’s process when in a negative mood, the researchers explained.

Also, the whole process itself was limited to reading the participants’ typing style and textual statements. It did not consider spontaneous emotion indicators like facial expressions.

Regardless of its limitations, the study’s program may open new doors to building technological machines that are emotionally-oriented and, therefore, more user-friendly. The researchers stated that their program can be a useful tool in developing better online education systems. For instance, content and teaching methods can be made adjustable to the students’ emotional state.

Viewpoint Discussion
 ディスカッションテーマ

Enjoy a discussion with your tutor.  

Discussion A

·         Would you want to use this program for yourself? Why or why not?
·         Other than as a teaching tool, how else can this program be useful? Please explain your answer.

Discussion B

·         In your opinion, can computers completely measure human emotions accurately? Why or why not?
·         Why do you think people would want to create technology that can read human emotions? Please explain your answer.

January 5, 2015