Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/152131
|
| Title: | 利用非監督式學習及基於條件熵之特徵選取探討華人青少年性格向度
Exploring personality dimensions of Chinese adolescents using unsupervised learning and feature selection based on conditional entropy |
| Authors: | 張煜均
Chang, Yu-Chun |
| Contributors: | 周珮婷
張育瑋
Chou, Pei-Ting
Chang, Yu-Wei
張煜均
Chang, Yu-Chun |
| Keywords: | 非監督式學習
條件熵
特徵選取
青少年性格量表
Unsupervised learning
Conditional entropy
Feature selection
Adolescent personality inventory |
| Date: | 2024 |
| Issue Date: | 2024-07-01 13:27:52 (UTC+8) |
| Abstract: | 在現今社會中，性格量表工具被廣泛應用於學校、職場和諮詢場所，用於評估個人的性格特徵和行為模式。日後也逐漸出現了針對不同對象及需求的量表，2008年台灣心理學者建立了青少年多向度性格量表，並採用了多種方式對於構念效度進行評估。在檢驗華人青少年多向度性格量表的構念效度時，發現測量相同概念的題目分散在不同的因素中，且傳統因素分析無法處理同時存在多個性格特徵的問題。
本研究旨在利用非監督式學習的分群演算法以及基於條件熵的特徵選取方法，探討華人青少年性格量表的結構及不同向度之間的交互情形。研究中，使用了階層式分群法對量表題目進行分群，分析單一向度的題目當中是否存在著多種性格特徵，並透過條件熵方法選取出對於七個向度最具代表性和重要性的題目。此外，還討論了向度組合之間是否存在方向不一致的題目。本研究的結果顯示，部分群體分群後的內部一致性高於因素分析結果，表明條件熵方法能改善部分量表向度的一致性，並在選取類別型資料的重要變數上優於隨機森林方法。同時，發現一些向度具有多種性格特徵，導致其他題目對目標向度的影響方向存在差異。透過釐清不同性格向度與量表題目之間的複雜關聯以及改善不合適的題目，能夠提升性格評估工具的準確性和解釋性，並為未來青少年性格量表的修訂提供新的方法和建議。
In contemporary society, personality inventory are widely used in schools, workplaces, and counseling settings to evaluate personality traits and behavior patterns. In 2008, Taiwanese psychologists developed the Multidimensional Personality Inventory for Chinese Adolescents and used various methods to evaluate its construct validity. However, items measuring the same concept were scattered across different factors, and traditional factor analysis could not address the issue of multiple coexisting personality traits.
This study explores the structure of the personality dimensions of Chinese adolescents and the interactions between different dimensions using unsupervised learning and feature selection based on conditional entropy. Hierarchical clustering was used to group the scale items, analyzing whether multiple personality traits exist within single-dimension items. Conditional entropy methods selected the most representative items for the seven dimensions and examined inconsistencies within dimension combinations. Results show that the internal consistency of some groups after clustering was higher than with factor analysis, indicating that the conditional entropy method can improve scale dimension consistency. It also outperforms the random forest method in selecting important variables for categorical data. Additionally, some dimensions contain multiple personality traits, causing directional differences in the impact of other items. By clarifying the complex relationships between personality dimensions and scale items and improving unsuitable items, the accuracy and interpretability of personality assessment tools can be enhanced. |
| Reference: | Ashton, M. C., & Lee, K. (2009). The HEXACO–60: A short measure of the major dimensions of personality. Journal of personality assessment, 91(4), 340-345.
Breiman, L. (1996). Bagging predictors. Machine learning, 24, 123-140.
Breiman, L. (2001). Random forests. Machine learning, 45, 5-32.
Chen, T. L., Chou, E. P., & Fushing, H. (2021). Categorical nature of major factor selection via information theoretic measurements. Entropy, 23(12), 1684.
Costa, P. T., & McCrae, R. R. (1992). Normal personality assessment in clinical practice: The NEO Personality Inventory. Psychological assessment, 4(1), 5.
Fushing, H., Chou, E. P., & Chen, T. L. (2023). Multiscale major factor selections for complex system data with structural dependency and heterogeneity. Physica A: Statistical Mechanics and its Applications, 630, 129227.
Goldberg, L. R. (1981). Language and individual differences: The search for universals in personality lexicons. Review of personality and social psychology, 2(1), 141-165.
Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4(1), 26–42.
Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. Personality psychology in Europe, 7(1), 7-28.
Johnson, S. C. (1967). Hierarchical clustering schemes. Psychometrika, 32(3), 241-254.
Lee, K., & Ashton, M. C. (2004). Psychometric properties of the HEXACO personality inventory. Multivariate behavioral research, 39(2), 329-358.
Liaw, A., & Wiener, M. (2002). Classification and regression by randomForest. R news, 2(3), 18-22.
Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes: Replicated factor structure in peer nomination personality ratings. The journal of abnormal and social psychology, 66(6), 574.
Quinlan, J. R. (1986). Induction of decision trees. Machine learning, 1, 81-106.
Shannon, C. E. (1948). A mathematical theory of communication. The Bell system technical journal, 27(3), 379-423.
Spearman, C. (1904). “General Intelligence,” Objectively Determined and Measured. The American Journal of Psychology, 15(2), 201–292.
Ward Jr, J. H. (1963). Hierarchical grouping to optimize an objective function. Journal of the American statistical association, 58(301), 236-244.
吳明隆（2006）。結構方程模式：SIMPLIS 的應用。五南圖書出版股份有限公司。
許功餘、鄭中平、董定儀、洪國洲、鍾苡文、徐達儒（2008）。華人性格特質七因素模式之結構不變性的檢驗。本土心理學研究，30，239-297。
許功餘（2017）。華人青少年多向度性格量表的構念效度之檢核(E10304)。取自中央研究院人文社會科學研究中心調查研究專題中心學術調查研究資料庫。https://doi.org/10.6141/TW-SRDA-E10304-1
許功餘（2018）。華人青少年性格向度的構念檢核：與北美性格向度之異與同。中華心理學刊，60，1–31。
楊國樞（1999）。中國人之基本性格向度、結構及效應的系統性研究（NSC 86-2143-H002-026）。行政院國家科學委員會。 |
| Description: | 碩士
國立政治大學
統計學系
111354015 |
| Source URI: | http://thesis.lib.nccu.edu.tw/record/#G0111354015 |
| Data Type: | thesis |
| Appears in Collections: | [統計學系] 學位論文
|
Files in This Item:
| File |
Description |
Size | Format | |
|---|
| 401501.pdf | | 2625Kb | Adobe PDF | 0 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
