跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 陳郁傑
Yu-Chieh Chen
論文名稱： 基於勝算比尋找影響在學經歷與畢業走向之未考慮因素
Exploring Out-Of-Context Factors Which Affect Learning Portfolio and Prospects After Graduation Based on Odds Ratio
指導教授： 蔡孟峰
口試委員:
學位類別： 碩士
Master
系所名稱： 資訊電機學院 - 資訊工程學系
Department of Computer Science & Information Engineering
論文出版年： 2023
畢業學年度： 111
語文別： 中文
論文頁數： 42
中文關鍵詞： 校務研究 、關聯規則探勘 、因果勝算比探勘 、未考慮因素
相關次數： 點閱：6  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 近幾年台灣各大專院校相繼成立校務研究辦公室，致力於探討學生議題及
    評估學校政策，透過科學化的決策支援系統輔助學校決策者改善學校經營方針與規劃未來發展，旨在促進學校和教育政策的發展，提高學校的教學品質、學生學習成果和學校整體運作效率。本校校務研究單位目前已蒐集、整理各單位的資料並進行整合，校務倉儲資料庫提供的實證資料涉及範疇極其多樣，教務、學務、總務以及人事等皆包含在內。
    在大數據分析的浪潮中，各個機構將面臨數量龐大的可分析議題，然而，
    對於選定何種類型資料針對特定議題進行分析仍然相當困難，有時候，分析結果可能出現錯誤或不如預期的情況，這些問題往往源於自身機構內資料面向的不完整性，由於資料種類的不足，可能導致錯誤的分析結果，因此，必須納入額外資料，重新進行分析，以獲得所期望的結果。本研究旨在探討在已有資料集中是否需要納入額外資料來進行分析，比較納入額外資料前後的準因果規則集，整理出三種情況進行評估：準因果規則強弱變化、更特定準因果規則以及更直接準因果規則，提供分析者一個以因果勝算比的角度考量是否應該納入額外資料的方法。本研究將以校務資料為例，以學生為主體，進行因果勝算比探勘，討論「給予學生經濟協助」與「學生擴大交友圈」兩大面向，在探究畢業走向相關議題時是否值得考慮進去。

    In recent years, many universities in Taiwan have established offices of institutional research one after another, dedicated to exploring student issues and
    evaluating school’s policies. These offices assist school administration in improving school management and future development planning by scientific decision support
    systems. The goal is to promote the development of schools and educational policies, enhance teaching quality, student learning outcomes and school operational efficiency.
    In the era of big data, institutions would face plenty of issues. However, it still difficult to determine which types of data should be used in specific issue analysis.
    Occasionally, analysis results may be wrong or unexpected. This situation often happen when the dimension of data within institution itself is not widely enough.
    Insufficient data types can lead to inaccurate analysis results, must incorporate additional data for reanalysis to obtain expected outcomes. This research aims to
    explore the need for incorporating additional data into existing datasets for analysis. We compare the odds ratio before and after incorporating additional data, evaluating
    three scenarios: changes in the strength of quasi causal rule, more specific quasi causal rules and core direct quasi causal rules. This provide analysts with a method to
    consider whether additional data should be incorporated from a causal odds ratio perspective. This research takes student data as an example, implements causal odds
    ratio mining with student data and discusses whether “providing financial assistance to student” and “expanding student’s social network” should be considered when
    exploring issues related to prospects after graduation.

    摘要 ii ABSTRACT iii 誌謝 v 目錄 vi 表目錄 viii 圖目錄 ix 一、 緒論 1 1-1 研究背景與動機 1 1-2 研究目的 2 二、 文獻探討 4 2-1 校務研究 4 2-2 關聯規則探勘(Association Rule Mining) 4 2-2-1 頻繁樣式探勘(Frequent Pattern Mining) 5 2-2-2 支持度(Support) 5 2-2-3 信賴度(Confidence) 6 2-3 因果勝算比探勘(Causal Odds Ratio Mining) 6 2-4 隊列研究(Cohort Study)[12] 7 2-5 勝算比(Odds Ratio, OR) 8 三、 研究方法 10 3-1 系統流程與架構 10 3-2 資料前處理 10 3-3 準因果規則定義與驗證 11 3-3-1 準因果規則候選 11 3-3-2 準因果規則驗證 11 3-4 分析擴增前後因果勝算比差異 13 3-4-1 已有準因果規則之強弱變化 13 3-4-2 更特定的準因果規則 14 3-4-3 更直接的準因果規則 15 四、 實驗 18 4-1 實驗規格與環境 18 4-2 實驗資料集 18 4-3 準因果規則比對與分析 18 4-3-1 擴增社團維度 18 4-3-2 擴增校內工讀維度 20 4-3-3 擴增獎學金維度 22 4-4 規則、情況(Case)數目與運算時間 23 4-5 隨機抽樣 24 4-5-1 100筆 24 4-5-2 200筆 25 4-5-3 400筆 25 4-6 隨機抽樣耗時與驗證 25 五、 結論與討論 27 六、 參考文獻 29

    [1] J. W. Tukey. Exploratory Data Analysis. Addison-Wesley, 1977.
    [2] J. L. &. M. J. R. Saupe. “The nature and role of institutional research: Memo to a college or university.” Tallahassee, FL: Association for Institutional Research. 1970.
    [3] J. L. Saupe. “The functions of institutional research.”, 2nd Edition, Tallahassece,FL: Association for Institutional Research. 1990.
    [4] 李政翰：〈我國推動大學校務研究之策略〉，2015年9月，取自https://ods.tmu.edu.tw/upload_file/tmudc/811/15875498111.pdf
    [5] 臺灣校務研究專業協會，取自https://tair.tw/
    [6] J. Han and M. Kamber. Data Mining: Concepts and Techniques. 2000.
    [7] R. Agrawal, T. Imielinski, and A. Swami. “Mining association rules between sets of items in large databases.” SIGMOD, Vol 22, June 1993, pp. 207–216.
    [8] R. Agrawal and R. Srikant. “Fast algorithms for mining association rules.” VLDB’94, September 1994, pp. 487-499.
    [9] J. Han, J. Pei and Y. Yin. 2000. “Mining Frequent Patterns Without Candidate Generation.” SIGMOD, Vol 29, June 2000, pp. 1–12.
    [10] F. Yusuf, S. Cheng, S. Ganapati and G. Narasimhan. “Causal Inference Methods and their Challenges: The Case of 311 Data.” DG.O’21, June 2021, pp. 49-59.
    [11] E Kummerfeld, J Ramsey. “Causal Clustering for 1-Factor Measurement Models.” KDD’16, August 2016, pp. 1655–1664.
    [12] J. Li, et al. “From observational studies to causal rule mining.” TIST, Vol 7, November 2015, pp 1-27.
    [13] J. W. Song and K. C. Chung. “Observational studies: Cohort and case-control studies. ” Plastic and Reconstructive Surgery, December 2010, pp. 2234–2242.
    [14] A.M Euser, et al. “Cohort studies: prospective versus retrospective.” Nephron Clinical Practice, 2009, pp. c214-c217.
    [15] J. M. Bland and D. G. Altman. “The odds ratio.” BMJ, Vol 320, May 2000, pp. 1468.
    [16] 行政院主計總處，取自https://earnings.dgbas.gov.tw/experience_sub_01.aspx
    [17] R. W. Floyd. “Algorithm 97: Shortest Path.” Communications of the ACM, Vol 5, June 1962, pp. 345.
    [18] R. Sedgewick. Algorithms in C, Part 5: Graph Algorithms. 2001.
    [19] P. O. Johnson and J. Neyman. “Tests of certain linear hypotheses and their application to some educational problems.” Statistical Research Memoirs, 1936, pp. 57-93.
    [20] Tegan George. “What Is a Cohort Study? | Definition & Examples.” February 2023. From https://www.scribbr.com/methodology/cohort-study/

    QR CODE
    :::