查詢結果分析
來源資料
相關文獻
- 運用腦波識別專注狀態
- Identifying the Out-of-Control Variables of Multivariate Control Chart Using Ensemble SVM Classifiers
- 感應卡技術發展及市場概況--RFID應用範圍廣 接受度愈見高漲
- 最安全的人員確認方式--生物辨識系統
- 癲癇病人--腦波檢查的臨床意義
- Applying Hashing Search and Fuzzy Fault-Tolerant Algorithms for the Fast Recognition of Multi-Font Printed Chinese Characters
- The Calculation of Complexity in Normal and Apoplectic EEG Signals
- A Highly Reliable Character Recognition Mechanism
- Why a Statistics-Based Face Recognition System Should Base Its Recognition on the Pure Face Portion: A Probabilistic Decision-Based Proof
- 單一視角下三度空間工件的辨識
頁籤選單縮合
題名 | 運用腦波識別專注狀態= |
---|---|
作者 | 朱璿瑾; 江政祐; 劉寧漢; |
期刊 | 資訊科技國際期刊 |
出版日期 | 20131200 |
卷期 | 7:2 2013.12[民102.12] |
頁次 | 頁14-23 |
分類號 | 312.13 |
語文 | chi |
關鍵詞 | 腦波; 專注狀態; 辨識; 支援向量機; Electroencephalogram; Attention status; Recognition; Support vector machine; |
中文摘要 | 人們透過多元的方式獲取知識與經驗,但大部分的人在成長過程中,是經由就學吸收新知,而學習的過程中,專心(Attention)的意識對於學習的成效有一定程度的影響,因此,學生能不能專心於課堂學習之上,為其學習成功與否的依據,倘若老師能即時知道學生是否專心,將可以適時地提醒學生、改善學習情況。本研究以觀測腦波的方式,辨識出學生在課堂上的專心與非專心(Inattention)狀態。首先,運用科學儀器進行腦波偵測並予以紀錄,接著將收集到腦波數據值,利用人工的方式過濾無效數據,再結合Supper Vector Machine(支援向量機,SVM)分類器進行運算、分析,便能辨識出二類型腦波數據值(專心v.s.非專心)。在研究中正確的辨識率最高可達71.17%。 |
英文摘要 | Human got the knowledge in various ways, but mostly from school. In the process of learning, attention affects the learning result. In the other words, attention is a basis of learning well. If a teacher can find immediately that a student is not concentrative, then the teacher can alert the student to improve the learning attitude at the right moment. This research is expected to recognize the EEG signal as attention or inattention through EEG measure. First, we used the wireless apparatus to measure and record EEG signal. And all the useless data were filtered manually. Finally, we used Support Vector Machine (SVM) classifier to calculate and analyze the data. The classifier can recognize EEG data as two classes (attention or inattention). According to our experiment results, the attention recognition rate in this research is 71.17%. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。