Project/Area Number |
16K16116
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Multi-year Fund |
Research Field |
Intelligent informatics
|
Research Institution | Yamaguchi University |
Principal Investigator |
Mabu Shingo 山口大学, 大学院創成科学研究科, 准教授 (70434321)
|
Project Period (FY) |
2016-04-01 – 2019-03-31
|
Project Status |
Completed (Fiscal Year 2018)
|
Budget Amount *help |
¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000)
Fiscal Year 2018: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
|
Keywords | 機械学習 / ニューラルネットワーク / 進化論的計算手法 / データマイニング / パターン認識 / 識別器 / 特徴抽出 / 人工知能 / ソフトコンピューティング |
Outline of Final Research Achievements |
This research aimed to develop AI systems that can extract and combine useful features for data classification with high accuracy. First, based on data mining methods that extract useful knowledge from data, ensemble learning algorithms that select and use extracted knowledge appropriately according to the necessity were proposed, and unsupervised and semi-supervised learning algorithms that can build classifiers with a small number of labeled data were proposed. Second, some feature extraction methods that extract useful features for data classification using deep learning were proposed. Finally, the proposed methods were applied to opacity classification of medical images (chest CT images), normal and abnormal sounds classification of the lung, and disaster area classification of satellite images, then the effectiveness of the proposed methods were clarified.
|
Academic Significance and Societal Importance of the Research Achievements |
大量のデータが利用可能となった高度情報社会において,データから知識を発見し,人間と同等以上の判断能力を有する人工知能を開発することは今後の社会の発展のために重要である.本研究では,データマイニング手法や深層学習を用い,医用データにおける疾患の識別や,災害発生時に人工衛星画像から迅速に被災地域を発見できるシステムの構築を例に,識別に有用な特徴抽出法や,問題に適した識別器の構成を複数提案し,その性能を明らかにしている.
|