Facemark Recommendation based on Emotion, Communication, and Motion Type Estimation in Text
江村優花, Yohei Seki · 2012
Many users used facemarks in recent computer mediated communication environments such as e-mail, Twitter or facebook. Facemarks are useful to express the emotion or communication functions beyond texts. However, many users feel difficult to choose the right one from lots of candidates. We propose a method to recommend facemarks based on the estimation of emotions, communication, or motion types in texts written by users. We implemented emotion, communication, or motion type estimation system with k-NN, and the accuracy of estimation is 53.1% on macro average. We also estimated the relevance of recommended facemarks for user intention, and found that 66.7% of facemarks were recommended properly, which improved significantly over the recommendation results only from emotions categories. 1. は じ め に