| Project/Area Number |
17K12768
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Kansei informatics
|
|---|
| Research Institution | The University of Tokushima |
|---|
Principal Investigator |
ITO Shin-ichi 徳島大学, 大学院社会産業理工学研究部(理工学域), 助教 (90547655)
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000)
Fiscal Year 2019: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2018: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2017: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
|
|---|
| Keywords | 脳波 / 個人差 / 灰色理論 / 嗜好 / 意思検出 / BCI / 深層学習 / 性格 / サポートベクターマシン / 聴く意思 / 意思 / ノイズ除去 / 感性情報学 / 感性計測評価 / 独立成分分析 / 遺伝的アルゴリズム / 意思伝達 |
|---|
| Outline of Final Research Achievements |
In checking whether human understands contents of learning, Center cumulative frequency comparison (CCFC) method was used to judge EEG signals or EEG signals with artifact. Multistage independent components analysis (MICA) was proposed to remove artifact and noise signals. Multi-layer perceptron was used to judge whether human understands contents of learning. The experimental results showed 68.3% of recognition accuracy.
In detecting preference of listening to sounds, Gray associate degree was calculated to extract the features of EEG signals and remove the noise signals. Support vector machine (SVM) was used to detect the preference. The experimental results showed 88.27% of detection accuracy.
In human-wants detection during exposure to music, Convolutional neural networks (CNNs) was used to extract the features of EEG and remove the noise signals. The SVM was used to detect the human-wants. The experimental results showed 99.4% of recognition accuracy.
|
| Academic Significance and Societal Importance of the Research Achievements |
学習理解の有無の検出では68.3%の判別精度、聴取音に対する好みの音の検出では88.35の検出精度、聴取音楽に対する聴く意思の検出では99.4%の分類精度、を実現するに至った。これらの研究成果は、意思を司る前頭前野脳波からその意思を直接的に検出するため、訓練を必要としない意思伝達BCI の構築が可能になるという学術的意義をもつ。また、IoTでも使用可能な感性インタフェースの構築に役立つという社会的意義をもつ。とくに、介護・医療や教育現場などにおいて、真意を伝えるコミュニケーションの支援、などの新たなヒューマンインタフェースの構築など、幅広い分野での貢献が期待できる。
|
