| dc.contributor.advisor | 廖文宏 | zh_TW |
| dc.contributor.advisor | Liao, Wen Hung | en_US |
| dc.contributor.author (Authors) | 張晉文 | zh_TW |
| dc.contributor.author (Authors) | Chang, Chin Wen | en_US |
| dc.creator (作者) | 張晉文 | zh_TW |
| dc.creator (作者) | Chang, Chin Wen | en_US |
| dc.date (日期) | 2017 | en_US |
| dc.date.accessioned | 28-Aug-2017 12:05:34 (UTC+8) | - |
| dc.date.available | 28-Aug-2017 12:05:34 (UTC+8) | - |
| dc.date.issued (上傳時間) | 28-Aug-2017 12:05:34 (UTC+8) | - |
| dc.identifier (Other Identifiers) | G0102971021 | en_US |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/112266 | - |
| dc.description (描述) | 碩士 | zh_TW |
| dc.description (描述) | 國立政治大學 | zh_TW |
| dc.description (描述) | 資訊科學系碩士在職專班 | zh_TW |
| dc.description (描述) | 102971021 | zh_TW |
| dc.description.abstract (摘要) | 閱讀是吸收知識的途徑，不同的閱讀模式所帶來的閱讀成效也會不同。如何透過機器學習的方式，從凝視點找出閱讀行為的關聯性，將是本研究的目標。實驗選擇低成本眼動儀紀錄讀者閱讀過程中的眼動資料，採用dispersion-based演算法找出凝視點，以計算凝視點特徵，包含凝視時間、凝視距離、凝視位置以及凝視方向。本研究將閱讀模式分成五種類別，包含快讀、慢讀、精讀、跳讀與關鍵字識別，透過不同文章的呈現，引導30位測試者遵循其內容進行閱讀，藉此收集不同行為模式的眼動資料。實驗流程中所有的眼動資料會隨機被分成為兩份，依序建立不同維度的訓練資料，由交叉驗證的分類結果找出理想之特徵與維度。以每次挑選6位測試者的眼動數據為測試資料進行5次分類驗證，其平均正確率為78.24%、74.19%、93.75%、87.96%以及96.20%，均達到不錯的分類結果。 | zh_TW |
| dc.description.abstract (摘要) | Reading is one of the paths to acquire knowledge. The efficiency is different when different reading patterns are involved. It is the objective of this research to classify reading patterns from fixation data using machine learning techniques. In our experiment, a low-cost eye tracker is employed to record the eye movements during the reading process. A dispersion-based algorithm is implemented to identify fixation from the recorded data. Features pertaining to fixation including duration, path length, landing position and fixation direction are extracted for classification purposes.Five categories of reading pattern are defined and investigated in this study, namely, speed reading, slow reading, in-depth reading, skim-and-skip, and keyword spotting. We have recruited thirty subjects to participate in our experiment. The participants are instructed to read different articles using specific styles designated by the experimenter in order to assign label to the collected data. Feature selection is achieved by analyzing the predictive results of cross-validation from the training data obtained from all subjects. The average classification accuracies in five-fold cross-validation are 78.24%, 74.19%, 93.75%, 87.96% and 96.20% using the eye movements of the six randomly selected subjects as test data. | en_US |
| dc.description.tableofcontents | 第一章 緒論 1第一節 研究背景 1第二節 研究動機與目的 2第三節 論文架構 3第二章 相關研究 4第一節 眼球追蹤技術 42.1.1 瞳孔與角膜反射追蹤法 42.1.2 眼電圖法 52.1.3 搜尋線圈法 72.1.4 Purkinje 影像追蹤法 82.1.5 紅外線眼動圖法 8第二節 眼動資料分析指標 92.2.1 時間維度眼動指標 102.2.2 空間維度眼動指標 11第三節 隨機取樣一致性演算法 12第三章 器材評估與研究設計 15第一節 研究工具探討 153.1.1 軟硬體架構分析 153.1.1.1 儀器規格與架設方式 163.1.1.2 軟體與裝置相互性架構 173.1.1.3 校正程序過程 193.1.2 可用性評估 223.1.2.1 使用介面設計並制定比較規則 223.1.2.2 數據收集與計算程序 263.1.2.3 數據結果分析與裝置評估結果 27第二節 前期規劃 293.2.1 凝視點識別並紀錄凝視點特徵 293.2.2 製作文字內容，並決定適當的行距高度 32第三節 研究設計與實施方法 363.3.1 研究實施步驟 363.3.2 軟體流程設計 37第四章 研究過程與結果 39第一節 研究過程 394.1.1 眼動實驗應用程式 394.1.2 定義閱讀模式類別 404.1.3 依據不同的行為類別製作閱讀內容 414.1.4 評估不同條件下的眼動軌跡狀況 43第二節 資料分析程序 454.2.1 收集所需數據資料的過程規劃 464.2.2 凝視點特徵選取並對數據進行資料正規化 474.2.3 藉由凝視點特徵收集來建立訓練資料過程 48第三節 預測結果與討論 504.3.1 以不同維度資料進行訓練的結果 514.3.2 以交叉驗證方式進行資料維度調整測試 524.3.3 檢驗不同測試資料的分類結果 554.3.4 探究較難區分的閱讀模式所影響整體分類結果的程度 584.3.5 以預測分類機率值來檢驗測試資料的分類結果 61第五章 結論與未來研究方向 65第一節 結論 65第二節 未來研究方向 65參考文獻 67附錄A 70 | zh_TW |
| dc.format.extent | 5387547 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.source.uri (資料來源) | http://thesis.lib.nccu.edu.tw/record/#G0102971021 | en_US |
| dc.subject (關鍵詞) | 眼動資料 | zh_TW |
| dc.subject (關鍵詞) | 閱讀模式 | zh_TW |
| dc.subject (關鍵詞) | 眼動儀 | zh_TW |
| dc.subject (關鍵詞) | 交叉驗證 | zh_TW |
| dc.subject (關鍵詞) | Eye movement | en_US |
| dc.subject (關鍵詞) | Reading pattern | en_US |
| dc.subject (關鍵詞) | Eye tracker | en_US |
| dc.subject (關鍵詞) | Cross-validation | en_US |
| dc.title (題名) | 基於眼動軌跡之閱讀模式分析 | zh_TW |
| dc.title (題名) | Classification of reading patterns based on gaze information | en_US |
| dc.type (資料類型) | thesis | en_US |
| dc.relation.reference (參考文獻) | [1] 王凱平. "移動式眼動儀之實作與視線軌跡分析." 政治大學資訊科學學系學位論文 (2008): 1-97.[2] 劉洪瑞, 邱文信, and 劉貞勇. 眼動儀在運動研究之應用.屏東教大體育15 (2012)[3] Rayner, Keith, et al. "Eye movements as reflections of comprehension processes in reading." Scientific studies of reading 10.3 (2006): 241-255[4] 黃孟隆, 唐大崙, 李執中, 林故廷. 眼動儀於瞳孔測謊之初探.犯罪偵查與鑑識科學研討會報告論文，2004.[5] 施懿芳. "行動廣告版面設計對眼球運動與美感情緒影響之研究."交通大學傳播研究所學位論文 (2013): 1-122.[6] Morimoto, Carlos Hitoshi, et al. "Pupil detection and tracking using multiple light sources." Image and vision computing 18.4 (2000): 331-335.[7] Tracking, Tobii Eye. "An Introduction to eye tracking and Tobii eye-trackers, White Paper." (2010).[8] Martínez, José A., et al. "Multimodal system based on electrooculography and voice recognition to control a robot arm." International Journal of Advanced Robotic Systems 10.7 (2013): 283.[9] Kim, Myoung Ro, and Gilwon Yoon. "Control signal from EOG analysis and its application." World Academy of Science, Engineering and Technology, International Journal of Electrical, Electronic Science and Engineering 7.10 (2013): 830-834.[10] Scleral Search Coils from:http://www.audiologyonline.com/articles/ics-impulse-revolutionizing-vestibular-assessment-12003[11] Popelka, Stanislav, et al. "Advanced Map Optimalization Based on Eye-Tracking." (2012).[12] Hansen, Dan Witzner, and Qiang Ji. "In the eye of the beholder: A survey of models for eyes and gaze." IEEE transactions on pattern analysis and machine intelligence 32.3 (2010): 478-500.[13] Methods of measuring eye movements, Electrooculography from:https://www.liverpool.ac.uk/~pcknox/teaching/Eymovs/emeth.htm[14] Eye monitors, Infra-Red Oculography from:http://www.cabiatl.com/mricro/eyemon/index.html[15] Rayner, Keith. "Eye movements in reading and information processing: 20 years of research." Psychological bulletin 124.3 (1998): 372. [16] Fischler, Martin A., and Robert C. Bolles. "Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography." Communications of the ACM 24.6 (1981): 381-395.[17] Tobii/developer zone, “Sentry Versus the EyeX” from:http://developer.tobii.com/community/forums/topic/sentry-versus-the-eyex/[18] Tobii/developer zone, “Fixing Sampling/refresh Rate” from:http://developer.tobii.com/community/forums/topic/fixing-samplingrefresh-rate[19] The Eye Tribe, “Developers Guide” from:https://s3.eu-central-1.amazonaws.com/theeyetribe.com/theeyetribe.com/dev/dev/index.html[20] Tobii Tech, “Developer`s Guide:, Tobii EyeX SDK for .NET. p5, September. 2015[21] Microsoft, “Visual Studio Community” from:https://www.visualstudio.com/zh-hant/vs/community/?rr=https%3A%2F%2Fwww.google.com.tw%2F[22] The Eye Tribe, “Getting Started” from:https://s3.eu-central-1.amazonaws.com/theeyetribe.com/theeyetribe.com/dev/start/index.html#setup[23] Martinez-Conde, Susana, Stephen L. Macknik, and David H. Hubel. "The role of fixational eye movements in visual perception." Nature Reviews Neuroscience 5.3 (2004): 229-240.[24] 韓承靜, and 蔡介立. "眼球軌跡記錄—科學學習研究的明日之星."科學教育310008): 2-11.[25] Aga Bojko, EYE TRACKING THE USER EXPERIENCE A Practical Guide To Research, Rosenfeld Media, 2013.[26] Smallpdf from: https://smallpdf.com/zh-TW/pdf-to-jpg[27] Chang, Chih-Chung, and Chih-Jen Lin. "LIBSVM: a library for support vector machines." ACM Transactions on Intelligent Systems and Technology (TIST) 2.3 (2011): 27. Software available at https://www.csie.ntu.edu.tw/~cjlin/libsvm/ | zh_TW |
