臺灣博碩士論文加值系統

鑑於數位科技發展迅速，網路及其他數位環境之性別暴力叢生，已然對於
隱私權與人身安全造成實害。基於性別差異而對他人進行的暴力行為，可能包 括性騷擾、性別歧視、性別誹謗等。由於網路的匿名性和即時性，暴力行為在 網路上的傳播速度快，影響範圍廣，對受害者的傷害也更大。在台灣， PTT 論 壇作為一個重要的網路社區，也不例外。這種暴力行為不僅對個人造成傷害， 也對社會的性別平等產生負面影響。目前對於網路性別霸凌的研究並不充足， 研究大多集中在實體空間的性別暴力，對於網路性別暴力的理解和認識還有待 提高。此外，由於網路性別暴力的匿名性和跨地域性，使得防治工作面臨著許 多挑戰，因此，我們需要更深入地研究網路性別暴力的現象，了解其產生的原 因、傳播的過程以及對個人和社會的影響，以便制定有效的防治策略和措施。 故本研究將透過 PTT 的特性以 boy-girl 版為主，研究工具運用 python 網路爬蟲、 文章分類及關鍵字計次。主要資料蒐集 2020 年 1 月至 2024 年 3 月之文章，時間 點以季為單位。研究結果以圖表和文字雲之視覺化資料呈現。這些資料按照季 度進行蒐集和分類，根據性別語言暴力相關關鍵字標記包含性別暴力言論的文 章，接著對關鍵字進行統計和去重計數分析。研究結果顯示，性別暴力言論在 2020 年尤其是夏季較為頻繁，隨後在 2022 年起出現下降趨勢。關鍵字使用頻率 存在季節性波動，反映了社會事件和政策變化對性別言論的影響。隨著社會對 性別平等問題的重視，性別暴力言論有所減少，但仍需持續努力進一步遏制這 種現象。

Given the rapid development of digital technology, the proliferation of gender- based violence in online and other digital environments has caused substantial harm to privacy and personal safety. Gender-based violence, which may include sexual harassment, gender discrimination, and gender defamation, is exacerbated by the anonymity and immediacy of the internet, leading to fast dissemination and wide- reaching impacts, thereby causing greater harm to victims. In Taiwan, the PTT forum, as an important online community, is no exception to these issues. This violence not only inflicts personal harm but also negatively affects societal gender equality. Currently, research on online gender bullying is insufficient, with most studies focusing on gender-based violence in physical spaces. There is a need for a deeper understanding and recognition of online gender-based violence. Additionally, the anonymity and cross-regional nature of online gender-based violence present many challenges for prevention and control. Therefore, it is essential to conduct in-depth studies on the phenomenon of online gender-based violence, understanding its causes, dissemination process, and impacts on individuals and society in order to develop effective prevention and control strategies. This study will focus on the Boy-Girl board of PTT due to its characteristics, using Python web scraping, article classification, and keyword counting as research tools. The primary data collection spans articles from January 2020 to March 2024, with the timeline divided into quarters. The research results will be presented using visual data in charts and word clouds. The data are collected and categorized quarterly, with articles containing gender-based violent language marked based on relevant keywords. Subsequently, keywords are statistically analyzed and deduplicated. The findings especially in the summer, but began to decline from 2022 onward. The frequency of keyword usage shows seasonal fluctuations, reflecting the influence of societal events and policy changes on gender discourse. As society places greater emphasis on gender equality issues, there has been a reduction in gender-based violent language, although continuous efforts are needed to further curb this phenomenon.

目錄
謝誌................................................................................................................................ I 摘要............................................................................................................................... II 英文摘要.................................................................................................................... III 目錄............................................................................................................................. IV 表目錄........................................................................................................................... V 圖目錄......................................................................................................................... VI 第壹章 緒論...................................................................................................................1
第一節 研究背景與動機.......................................................................................1
第二節 研究目的...................................................................................................2 第貳章 文獻探討...........................................................................................................4 第一節 網路性別暴力...........................................................................................4 第二節 社交媒體...................................................................................................5 第三節 法律規範...................................................................................................7 第四節 文字探勘...................................................................................................9 第參章 研究方法與設計.............................................................................................11 第一節 研究主題.................................................................................................11 第二節 研究方法.................................................................................................11 第肆章 研究結果與討論.............................................................................................14 第一節 性別暴力言論的時間變化趨勢.............................................................14 第二節 每季性別暴力關鍵字使用頻率.............................................................17 第三節 每季性別言論暴力關鍵字去重計數分析.............................................23 第伍章 結論與建議.....................................................................................................25第一節 結論.........................................................................................................25 第二節 建議.........................................................................................................25 參考文獻...................................................................................................................... 28 附錄.............................................................................................................................. 30





表目錄
表 1 性別語言暴力詞庫..............................................................................................11 表 2 2020 年 1 月至 2024 年 3 月文章數量的統計 ................................................... 15 表 3 2020 年前 10 名關鍵字文字雲 ........................................................................... 18 表 4 2021 年前 10 名關鍵字文字雲 ........................................................................... 19 表5 2022年前10名關鍵字文字雲..........................................................................20 表6 2023年前10名關鍵字文字雲..........................................................................21 表7 2024年Q1前10名關鍵字文字雲...................................................................22



圖目錄
圖 1 數位科技正反面影響............................................................................................5 圖 2 PTT 論壇分類看板示意圖.................................................................................... 6 圖 3 PTT 論壇文章排列示意圖.................................................................................... 7 圖 4 臺灣新竹地方法院 112 年度易字第 64 號刑事判決-1 ...................................... 7 圖 5 臺灣新竹地方法院 112 年度易字第 64 號刑事判決-2 ...................................... 8 圖 6 性騷擾防治法第 7 條 ............................................................................................ 9 圖 7 自然語言處理......................................................................................................10 圖 8 資料探勘 vs 文字探勘 ........................................................................................ 10 圖 9 PTT 網誌規則圖.................................................................................................. 11 圖 10 圖 10 去重計數 python 程式碼......................................................................... 13 圖 11 2020 年 1 月至 2024 年 3 月文章數量趨勢圖 ................................................. 16 圖 12 2020 年 1 月至 2024 年 3 月文章去重計數帶有關鍵字的詞頻趨勢圖 ......... 24

中文文獻 王珮玲、陳怡青、方念萱(2023)。數位性別暴力盛行率研究初探。臺大社會工
作學刊，第四十七期，149 頁-200 頁。 江淑惠、張嘉耀和黃文鼎(2022)青少年社群媒體使用行為影響人際關係、學業
成就、從眾行為之研究，醒吾學報，65 2022.01，頁 75-104。 行政院性別平等會(2021)。數位/網路性別暴力之定義、類型及其內涵說明。取
自 https://gec.ey.gov.tw/Page/ED8994F4EF5AD73E/2ab74b7e-0bdb-4067-
b43a-4a3cfc9e2a1e
行政院性別平等會(2021)。《性別平等教育法》修正案。取自
https://www.moe.gov.tw/News_Content.aspx?n=9E7AC85F1954DDA8&s=D71
495741C3D7B2A 台灣網路講堂(2021)。社群網路中的性別歧視與性別暴力探討。取自
https://www.twsig.tw/20210526-2/ 鳴人堂(2021)。數位性暴力在台灣無法可管?民團訴求專法防止挖面事件再發
生(洪敏隆)。取自 https://opinion.udn.com/opinion/story/120626/5899111 余貞誼(2016)。「我說妳是妳就是」:從 PTT「母豬教」的仇女行動談網路性
霸凌的性別階層。取自 https://news.readmoo.com/2019/07/23/misogyny-
2/#fn20 簡淑如、吳孟芸(2018)。網路強制分級之隱憂:網路內容管制政策之探討與建
議。廣播與電視，第二十八期，頁 161-頁 194。 今日新聞(2022)。不忍了!鄭家純精選 80 位網友提告:不要以為諧音罵人沒
事。取自 https://www.nownews.com/news/6011514 司法院判決書系統。臺灣新竹地方法院 112 年度易字第 64 號刑事判決。取自
https://judgment.judicial.gov.tw/FJUD/data.aspx?ty=JD&id=SCDM,112%2c%e
6%98%93%2c64%2c20230113%2c1 全國法規資料庫。性騷擾防治法第 7 條。取自
https://law.moj.gov.tw/LawClass/LawAll.aspx?PCode=D0050074


28



陳世榮(2015)。社會科學研究中的文字探勘應用:以文意為基礎的文件分類及 其問題。人文及社會科學集刊，第二十七卷第四期，頁 683-頁 718。 楊妮蓁(2019)。應用 PTT 論壇文字探勘與情感分析探討政府政策進行之研究:
以勞基法為例。
我與資種 | ITSEED 資訊種子官方部落格。輕鬆理解文字探勘(上) — 從機器
學習到資料探勘。取自 https://reurl.cc/09xzxl 葉國暉、林宥均、李林發、楊冠彥(2021)，性別歧視語意情緒分析與智慧諮商
系統。
蔡淳如(2023)。COVID-19 疫情期間增長的數位/網路性別暴力。取自
https://reurl.cc/LWpova
英文文獻
Grimmer J, &Stewart BM(2013). Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts. Political Analysis.
Herring, S. C., & Martinson, A(2004). Assessing gender authenticity in computer- mediated language use: Evidence from an identity game. Journal of Language and Social Psychology
Peter Turney. Thumbs(2002) Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews. Proceedings of the Association for Computational Linguistics.
Johnson, M.P. (2020). A typology of domestic violence: Intimate terrorism, violent resistance, and situational couple violence. Northeastern University Press.