Discernment of comfort/discomfort

A method of determining whether a pain corresponds to stress or unpleasantness, comprising:

• a) providing the computer implementing an unpleasantness determination classifier for determining whether a pain corresponds to stress or unpleasantness of an object, wherein the computer includes a processor and the unpleasantness determination classifier is generated by a machine learning process;
• and
• b) obtaining, by the processor, brainwave data or analysis data thereof from the object and applying the data to the unpleasantness determination classifier to determine unpleasantness of the object, wherein the machine learning process includes,
• c) applying the different stimuli to the object under the same environment to obtain each brainwave data or analysis data thereof by using the brainwave recording sensor, wherein the brainwave recording sensor has a plurality of electrodes including brainwave measurement electrodes configured to be placed on a subject's scalp,
• d) associating, by the processor, a difference in the brainwave data or analysis data thereof obtained under the same environment with a reaction of the object to the stimulation, and
• e) generating, by the processor, the unpleasantness determination classifier based on the association, wherein the generating the unpleasantness determination classifier includes,
• i) preprocessing the brainwave data by filtering to remove noises of eye blinking and myogenic potential and by normalizing the filtered brainwave data based on the brainwave data during a first predetermined period before applying the stimulation, and
• ii) generating the unpleasantness determination classifier by machine learning with feature selection based on first and second brainwave data, wherein the first brainwave data are the preprocessed brainwave data during a second predetermined period after applying a first level of the stimulation corresponding to unpleasantness, and the second brainwave data are the preprocessed brainwave data during the second predetermined period after applying a second level of the stimulation corresponding to pleasantness, wherein the associating the difference includes,
• extracting features by converting the brainwave data into the absolute value of amplitude during the second predetermined period and converting thereinto z values;
• and
• generating the unpleasantness determination classifier by using LASSO and Bayesian optimization method.

A system for determining stress or unpleasantness of an object, comprising:

• a brainwave recording sensor, wherein the brainwave recording sensor has a plurality of electrodes including brainwave measurement electrodes configured to be placed on a subject's scalp,
• a first computer including a first processor, the first processor configured to
• i) obtain, by using the brainwave recording sensor, brainwave data or analysis data thereof from the object and
• ii) perform, based on the obtained brainwave data or analysis data thereof, a process as an unpleasantness determination classifier for determining whether a pain corresponds to stress or unpleasantness of the object, the classifier generated by a machine learning process;
• and
• a second computer including a second processor, the second processor configured to execute the machine learning process, the machine learning process including,
• a) applying the different stimuli to the object under the same environment to obtain each brainwave data or analysis data thereof by using the brainwave recording sensor,
• b) associating, by the second processor, a difference in the brainwave data or analysis data thereof obtained under the same environment with a reaction of the object to the stimulation, and
• c) generating, by the second processor, the unpleasantness determination classifier based on the association, wherein the generating the unpleasantness determination classifier includes,
• i) preprocessing the brainwave data by filtering to remove noises of eye blinking and myogenic potential and by normalizing the filtered brainwave data based on the brainwave data during a first predetermined period before applying the stimulation, and
• ii) generating the unpleasantness determination classifier by machine learning with feature selection based on first and second brainwave data, wherein the first brainwave data are the preprocessed brainwave data during a second predetermined period after applying a first level of the stimulation corresponding to unpleasantness, and the second brainwave data are the preprocessed brainwave data during the second predetermined period after applying a second level of the stimulation corresponding to pleasantness, wherein the associating the difference includes,
• extracting features by converting the brainwave data into the mean value of amplitude and the frequency power during the second predetermined period;
• and
• generating, using Support Vector Machine Recursive Feature Elimination method, the unpleasantness determination classifier by ranking the features and eliminating the feature with the lowest contribution.

A system for determining stress or unpleasantness of an object, comprising:

• a brainwave recording sensor, wherein the brainwave recording sensor has a plurality of electrodes including brainwave measurement electrodes configured to be placed on a subject's scalp,
• a first computer including a first processor, the first processor configured to
• i) obtain, by using the brainwave recording sensor, brainwave data or analysis data thereof from the object and
• ii) perform, based on the obtained brainwave data or analysis data thereof, a process as an unpleasantness determination classifier for determining whether a pain corresponds to stress or unpleasantness of the object, the classifier generated by a machine learning process;
• and
• a second computer including a second processor, the second processor configured to execute the machine learning process, the machine learning process including,
• a) applying the different stimuli to the object under the same environment to obtain each brainwave data or analysis data thereof by using the brainwave recording sensor,
• b) associating, by the second processor, a difference in the brainwave data or analysis data thereof obtained under the same environment with a reaction of the object to the stimulation, and
• c) generating, by the second processor, the unpleasantness determination classifier based on the association, wherein the generating the unpleasantness determination classifier includes,
• i) preprocessing the brainwave data by filtering to remove noises of eye blinking and myogenic potential and by normalizing the filtered brainwave data based on the brainwave data during a first predetermined period before applying the stimulation, and
• ii) generating the unpleasantness determination classifier by machine learning with feature selection based on first and second brainwave data, wherein the first brainwave data are the preprocessed brainwave data during a second predetermined period after applying a first level of the stimulation corresponding to unpleasantness, and the second brainwave data are the preprocessed brainwave data during the second predetermined period after applying a second level of the stimulation corresponding to pleasantness, wherein the associating the difference includes,
• extracting features by converting the brainwave data into the absolute value of amplitude during the second predetermined period and converting thereinto z values;
• and
• generating the unpleasantness determination classifier by using LASSO and Bayesian optimization method.