Adaptive Multi Model Architecture by Using Similarity Between Trained User and New User

Affective computing is a field of great interest in many applications, including video surveillance, behavior analysis, human robot interaction, life-logging. While accurately predicting these technologies (e.g. emotion estimation, confident for talking a topic) could have benefits for many fields, generic machine learning systems still yield low performance in some situation. We hypothesize that these sophisticated models suffer from individual differences of human's personality. Therefore, we proposed a multi characteristic model architecture which combine the personalized machine learning models and utilize each model 's prediction score in the inference. This architecture formed with reference to ensemble machine learning architecture. In this research, we presented the calculation method of weighted average in multi characteristic architecture by using similarity between a new sample and trained characteristics. We estimated confidence during communication as a human internal state. Empirical results demonstrate that using multi model trained each person 's information to account for individual differences provides improvements over traditional machine learning system and insight for dealing with various individual types.