Case: Analysis of odor sensor output

Analysis experiment utilizing AI! Conducting visualization and principal component analysis on the output data from the odor sensor.

We would like to introduce a case study of output analysis from customers who are evaluating and considering odor sensors. This sensor is composed of 20 types of elements, and by inputting samples that are the source of odors, the reactions of each element are output as data. In this case, samples of three types of vinegar and three types of fabric softener were prepared by the customer, with 50 measurements for each (the total number of data points is 6×50×20=6,000). An analytical experiment utilizing AI was conducted to determine whether these data could identify the samples. [Case Overview] - Visualization and principal component analysis were performed on the output data from the odor sensor. - A machine learning model was constructed to predict the groups and classes of vinegar and fabric softeners. - Moving averages were also utilized for evaluation to improve classification performance. - In group classification, many results had an accuracy rate of over 0.9, indicating high precision. *For more details, please refer to the PDF document or feel free to contact us.

Contract AnalysisOther analysesContract Analysis