Chuck Bednar for redOrbit.com – Your Universe Online
A new Facebook application could help bring e-learning closer to an actual classroom environment by monitoring the emotional states of students and notifying teachers when those individuals are having a rough day.
The application is known as SentBuk, and the Autonomous University of Madrid (AUM) researchers behind it say that it uses algorithms to analyze social media messages, then discerns a person’s emotional state. When transmitted to online instructors, it could give them the same type of information other teachers can obtain by looking at their students’ faces.
“SentBuk is an application external to Facebook which, with the user’s permission, analyses the messages he/she publishes and calculates his/her emotional state,” explained Álvaro Ortigosa, Director of the UAM National Centre of Excellence in Cybersecurity. “The tool is based on two algorithms: the first calculates the emotional load of each message and classifies it as positive, negative or neutral. The second deduces emotional state by comparing it with the emotional load of recent messages.”
Ortigosa, one of the authors of a Computers in Human Behavior paper discussing the tool, added that it “utilizes a natural language analysis technique to recognize significant words with emotional load. It also uses an automatic, machine-learning-type classification system. Based on a large bank of sentences classified by humans, the application has been trained to learn to reproduce human judgment. The emotional load assigned to each sentence arises from a combination of both calculations.”
He and his fellow UAM scientists believe that the application could be used in online learning courses to create more adaptive programs, which would allow students to complete tasks at the most appropriate times. It would recommend avoiding particular assignments when it detects that a person is in a negative state of mind, replacing them with more motivational content instead.
Such an analysis would enable teachers to gather feedback similar to those who can simply look at their students’ faces while working in a traditional classroom – something that typically is not possible with online schooling. He added that it could also be used to remotely monitor people who are ill, or could be adapted by companies to gauge customer satisfaction.
In its most basic form, though, Ortigosa said that the application will alert professors when it detects that several students are in a negative frame of mine. While it does not determine the cause of the distress, since students of an e-learning course have little to no relation to each other outside of the virtual classroom, the data could serve as a warning sign to the instructor.
“Although there may be many reasons for the emotional state, the hypothesis is that these negative emotions should be uniformly distributed across time,” he said, noting that “if at any given moment a negative emotional peak is detected in a representative sample of the students, it is highly probable that such emotional variation is due to some situation relating to the course, and thus the tool will send a warning message to the teacher.”
In addition to Ortigosa, José M. Martín and Rosa M. Carro of the AUM Department of Computer Science were also involved in the research, which is described as one part of a larger project attempting to infer characteristics such as personality and emotional load of social media users.