Data Mining - Concepts and Techniques.

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Social Network Analysis

Emotion and Adaptation

This paper proposes a method for evaluating our emotion corpus and describes the result of evaluation. The emotion corpus was originally created in Japanese and English to be used for emotion recognition research and consisted of dialogue sentences including emotional expressions. To ensure the reliability and use the corpus as the standard of emotion evaluation, the objective evaluation is necessary because single person annotated tags on the corpus. For Japanese part of corpus evaluation, we conducted a questionnaire to the several examinees. The examinees were asked to tag arbitrarily emotion categories on the sentences in the corpus. The obtained sentence emotion tags were classified then the match rate of the initial tag and the tags annotated by the examinees. Reliability of each emotion tag annotated on the sentence was calculated and added as emotional weight on each sentence emotion tag. The 10-fold cross validation test was conducted using the weight added corpus to estimate sentence emotion and obtained recall of 0.62 on average.

An approach for evaluating emotion tagged corpus

Integration of Multiple Classifiers for Chinese Semantic Dependency Analysis

In this paper, we propose an approach on answer generation for cooking question answering system. We first review previous work of question analysis. Then, we give annotation scheme for knowledge database. Finally, we present the answer planning based approach for generate an exact answer in natural language. An evaluation has been conducted on natural language questions and the result shows that the system can satisfy user's demand.

Answer generation for Chinese cuisine QA system

https://www.jstage.jst.go.jp/article/ieejeiss/130/4/130_4_580/_pdf

A Practical Sightseeing Question Answering System Based on Integrated Knowledge-Base

The purpose of this research is to create an environmental recognition system, specifically on emotion expressions and human states, for a humanoid robot aiming for interpersonal services. Region Convolutional Neural Networks (R-CNN) are often used for detecting objects in the environment. We employ a Mask R-CNN model for detection of emotions and states of a target person from the robot’s field of view. The model was trained using various images of a human’s body in several emotional states. Experiments were conducted to validate the effectiveness of the model to detect the states of surrounding people from the robot’s camera. Although the set of human states assumed in the experiment was limited, the results of the experiments imply the potential of the proposed method to act as a basis of a recognition model for an intelligent humanoid robot for interpersonal services.

Fuji Ren

Papers

An approach for evaluating emotion tagged corpus

Integration of Multiple Classifiers for Chinese Semantic Dependency Analysis

Answer generation for Chinese cuisine QA system

A Practical Sightseeing Question Answering System Based on Integrated Knowledge-Base

Environment Recognition Using Robot Camera