This course focuses on sensory information sensing including physiological measurement, sensory test, measurement by sensors and sensor fusion. Moreover, students will learn multivariate data analysis such as principal component analysis, discrimination analysis, regression analysis, cluster analysis and then learn machine learning such as MultiLayer Perceptron, Self-Organizing Map, and Support Vector Machine. They will learn sensory information sensing based upon the knowledge above using several case studies and examples.
Nowadays it is required to match a machine with human. The harmony between human and a machine is becoming more and more important since we have aging society. Since sensory information sensing of a human is indispensable for making a human-friendly machine, this course provides the measurement and analysis methods of sensory information sensing.
At the end of this course, students will be able to
1) understand the principles of obtaining physiological data.
2) perform sensory test.
3) understand principles of sensors for sensory information sensing.
4) analyze physiological data, sensory-test data and sensor data using multivariate analysis and machine learning.
Sensory information, physiological data, sensory test, multivariate analysis, sensor, machine learning
|✔ Specialist skills||Intercultural skills||Communication skills||Critical thinking skills||✔ Practical and/or problem-solving skills|
The class reviews the contents of previous class and then provides the contents for each class. As many examples and case studies as possible are introduced.
|Course schedule||Required learning|
|Class 1||Introduction of sensory data measurement||Explain importance of sensory-information measurement and its introduction using several examples.|
|Class 2||Electrocardiogram measurement||Explain electrocardiogram measurement, the most popular physiological measurement, and describe the information obtained from ECG.|
|Class 3||Electroencephalogram measuremant and Near Infra- Red Spectoroscopy||Explain electroencephalogram measuremant and Near Infra- Red Spectoroscopy, a physiological measurement method becoming popular recently.|
|Class 4||Classification of sensory test||Explain categories of sensory test such as those of analysis and preference types.|
|Class 5||Presentation method of sensory information||Explain physical method and procedure of sensory-information presentation.|
|Class 6||Data analysis method of sensory test||Explain statistical method of dealing with data from sensory test.|
|Class 7||Sensors required for measuring sensory information (Vision, audition and haptics)||Explain image sensor, microphone and haptic sensorrequirred for sensory information sensing.|
|Class 8||Sensors required for measuring sensory information (Olfaction and gustation)||Explain odor and taste sensors required for sensory information sensing.|
|Class 9||Sensor fusion||Explain information integration among various types of sensors.|
|Class 10||Classification of muitivariate data analysis||Explain supervised and unsupervised methods in multivariate analysis.|
|Class 11||Principal component analysis and discrimination analysis||Explain the most popular methods such as principal component analysis and discrimination analysis in multivariate analysis.|
|Class 12||Regression analysis and cluster analysis||Explain linear and nonlinear regression analysis and cluster analysis based upon various criteria.|
|Class 13||Machine learning (MultiLayer Perceptron, Self-Organizing Map, Support Vector Machine, Deep learning)||Explain methods of machine learning for analyzing sensory data.|
|Class 14||Sensory information sensing based on sensor data measurement||Describe sensory information sensing using artificial sensors. Several case studies are introduced.|
To enhance effective learning, students are encouraged to spend approximately 100 minutes preparing for class and another 100 minutes reviewing class content afterwards (including assignments) for each class.
They should do so by referring to textbooks and other course material.
Distribute course materials during class.
R.Oduda, P.E.Hart and D.G.Stork, Pattern classification, Wiley-interscience, 2001 (ISBN 0-471-05669-3)
T.Nakamoto Ed., Essentials of machine olfaction and taste, Wiley, 2016.
Evaluation of achievement by final examination (70%) and quizzes (30%).
Students must have successfully completed Probability and Statics (ICT. M202) or have equivalent knowledge.
Takamichi Nakamoto e-mail: nakamoto[at]nt.pi.titech.ac.jp, Ex.5017
Contact by e-mail or phone in advance is required to schedule an appointment.