-
Machine Learning Can Be Used to Model the Human Brain
Japanese Researchers have proposed a content recommendation system that states that our subconscious mind largely controls our brain activity. While we all believe that we know ourselves better, a hypothesis does state that our subconscious mind has control over our mind and the researchers have proposed this to be true.
Such a system makes use of the user’s brain signals, acquired using an MRI scan for instance when exposed to particular content. By exploring various users and content, it builds up a general model of brain activity.
The team also found a solution for maintaining a high inference performance and at the same time limiting the information cost. The selection process quantified the contribution of a questionnaire item by attributing thereto an “importance score”. And then retained only those which had top importance scores for the inference.
In order to effectively validate the scheme, the performance accuracy was evaluated by the team using a brain model obtained experimentally and a profile based on real information. It was noted that the team achieved the same level of interference accuracy of the brain model as the case by employing 209 questionnaires by using only 15-20 topmost items. They suggested that only the top 10% of the questionnaire items could infer the brain model.
Well, it does seem with the research, that the Machine Learning technique could be used to know more about ourselves than we do!
This is a summary of an article published on TechXplore. Click here to read the full article.
Sorry, there were no replies found.
Log in to reply.