本文選題：網(wǎng)絡(luò)輿情 + 情感傾向��； 參考：《湘潭大學(xué)》2017年碩士論文

【摘要】：隨著移動(dòng)互聯(lián)網(wǎng)的快速發(fā)展,社交網(wǎng)絡(luò)已經(jīng)成為用戶獲取信息、表達(dá)意見、交流看法的重要平臺(tái)。熱點(diǎn)事件一旦發(fā)生后,網(wǎng)絡(luò)用戶可以通過文本、圖片、小視頻等方式表達(dá)自己對(duì)某個(gè)社會(huì)事件的態(tài)度、認(rèn)知、意見和情感等主觀性信息。信息經(jīng)過轉(zhuǎn)發(fā)、評(píng)論和點(diǎn)贊等方式進(jìn)行傳播,同時(shí)若用戶在轉(zhuǎn)發(fā)與評(píng)論信息時(shí)加入個(gè)人主觀性情感,從而促進(jìn)了事件的演化。近年來,網(wǎng)絡(luò)群體性事件數(shù)量急劇上升,在網(wǎng)民中引起了巨大的輿論反響,當(dāng)突發(fā)事件爆發(fā)時(shí)若不對(duì)不良情感進(jìn)行控制和引導(dǎo),輿論則很容易極端化,甚至危及社會(huì)安全與穩(wěn)定。因此,有必要面向網(wǎng)絡(luò)輿情進(jìn)行用戶情感傾向性分析研究,為政府有效掌握和監(jiān)控網(wǎng)絡(luò)輿情突發(fā)事件提供相應(yīng)的理論支持和對(duì)策建議。本文以“羅一笑”網(wǎng)絡(luò)熱門話題事件為例,對(duì)輿情信息進(jìn)行情感分析和輿情追蹤。主要的研究工作包括:第一,利用網(wǎng)絡(luò)爬蟲工具采集事件相關(guān)微博數(shù)據(jù),并進(jìn)行整理分析。第二,以知網(wǎng)HowNet等詞典為基礎(chǔ)對(duì)情感詞進(jìn)行擴(kuò)展,構(gòu)建一個(gè)比較全面的情感分類詞典,同時(shí)對(duì)各情感詞所表達(dá)的情感極性和強(qiáng)度進(jìn)行識(shí)別和標(biāo)記。第三,構(gòu)建情感傾向分析模型,判斷網(wǎng)絡(luò)輿情的情感類型和統(tǒng)計(jì)情感詞頻,并對(duì)該事件中的用戶情感進(jìn)行挖掘與可視化分析。第四,運(yùn)用實(shí)證分析研究,對(duì)該事件的輿情演化階段進(jìn)行劃分,分別對(duì)各階段用戶情感演化特征及規(guī)律進(jìn)行分析。為后續(xù)網(wǎng)絡(luò)輿情情感引導(dǎo)對(duì)策的提出提供參考依據(jù)。實(shí)驗(yàn)表明,網(wǎng)絡(luò)輿情從生成到最終消亡是一個(gè)完整的生命周期,通過對(duì)網(wǎng)絡(luò)輿情演化進(jìn)行科學(xué)的階段劃分,可以發(fā)現(xiàn)各階段特征:(1)開始期微博發(fā)布數(shù)量少,網(wǎng)民對(duì)網(wǎng)絡(luò)輿情事件的態(tài)度紛繁復(fù)雜,但是通過對(duì)文本中用戶情感的挖掘、觀點(diǎn)的提取有利于進(jìn)一步跟蹤事件的后續(xù)發(fā)展趨勢(shì);(2)爆發(fā)期微博發(fā)布數(shù)最多,用戶參與度最高,影響范圍和影響效果極大,網(wǎng)民對(duì)事件的態(tài)度、觀點(diǎn)、情感等信息能夠?yàn)榫W(wǎng)絡(luò)輿情分析和監(jiān)控提供足量的數(shù)據(jù)基礎(chǔ),同時(shí),爆發(fā)期的情感傾向很大程度上定義了網(wǎng)絡(luò)輿情事件的總體情感演化趨勢(shì),相關(guān)部門應(yīng)對(duì)爆發(fā)期的網(wǎng)絡(luò)輿情情感演化多加關(guān)注,并引導(dǎo)輿情朝著正確的方向發(fā)展;(3)發(fā)酵期網(wǎng)民對(duì)網(wǎng)絡(luò)輿情事件的新資訊、新動(dòng)態(tài)較為敏感,正面信息公開與輿情披露在此階段能夠起到良好的效果;(4)消解期和反思期用戶參與程度較低,但仍需要對(duì)網(wǎng)絡(luò)輿情事件進(jìn)行跟蹤報(bào)道,規(guī)避謠言,肅清網(wǎng)絡(luò)環(huán)境,避免網(wǎng)絡(luò)輿情事件的二次發(fā)酵。
[Abstract]:With the rapid development of mobile Internet, social network has become an important platform for users to obtain information, express opinions and exchange views. Once a hot event occurs, Internet users can express their attitude, cognition, opinion and emotion on a social event by means of text, picture, small video and so on. The information is transmitted by way of forwarding, commenting and liking, and if the user adds personal subjective emotion to transmit and comment on the information, it promotes the evolution of the event. In recent years, the number of network mass incidents has risen sharply, causing a huge public opinion response among Internet users. When emergencies break out, if they do not control and guide bad emotions, public opinion is easy to become extreme. Even endanger social security and stability. Therefore, it is necessary to analyze and study the emotional tendency of users in order to provide corresponding theoretical support and countermeasures for the government to effectively grasp and monitor the sudden events of network public opinion. Taking Luo Yixiao as an example, this paper analyzes and tracks public opinion information. The main research work is as follows: firstly, the Weibo data are collected and analyzed by using web crawler tools. Secondly, based on the HowNet dictionary, we construct a comprehensive emotion classification dictionary, and identify and mark the emotion polarity and intensity expressed by each emotion word. Thirdly, we construct an emotional tendency analysis model to judge the emotional types and statistical affective word frequency of network public opinion, and mine and visualize the user emotion in this event. Fourthly, using the empirical analysis, the public opinion evolution stage of the event is divided, and the characteristics and rules of user emotion evolution in each stage are analyzed respectively. It provides the reference for the following network public opinion emotion guidance countermeasure. The experiment shows that the network public opinion is a complete life cycle from the generation to the final extinction. By dividing the evolution of the network public opinion into scientific stages, we can find that the number of Weibo releases at the beginning of each stage is small. Internet users' attitude to network public opinion events is complicated, but through the mining of users' feelings in the text, the point of view extraction is conducive to further tracking the future development trend of events.) in the outbreak period, the number of Weibo releases is the most, and the participation of users is the highest. The influence range and the influence effect are great, the netizens' attitude, viewpoint, emotion and so on information can provide the sufficient data foundation for the network public opinion analysis and the monitoring, at the same time, The emotional tendency of the outbreak period has largely defined the overall emotional evolution trend of the network public opinion event, and the relevant departments should pay more attention to the evolution of the network public opinion emotion in the outbreak period. And guide the public opinion to develop in the right direction.) during the fermentation period, the netizens are more sensitive to the new information about the network public opinion events. Positive information disclosure and public opinion disclosure can play a good effect in this stage. (4) the level of user participation in the period of resolution and reflection is relatively low, but it is still necessary to track and report online public opinion events, to avoid rumors, and to eliminate the network environment. Avoid the secondary fermentation of network public opinion events.
【學(xué)位授予單位】：湘潭大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：C912.63

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉培玉;荀靜;費(fèi)紹棟;朱振方;;基于隱馬爾可夫模型的主觀句識(shí)別[J];中文信息學(xué)報(bào);2016年04期

2 周志杰;劉功申;;基于投票機(jī)制的文本主客觀分類系統(tǒng)研究[J];微型電腦應(yīng)用;2015年02期

3 侯敏;滕永林;李雪燕;陳毓麒;鄭雙美;侯明午;周紅照;;話題型微博語言特點(diǎn)及其情感分析策略研究[J];語言文字應(yīng)用;2013年02期

4 柳軍;蔡淑琴;;微內(nèi)容的網(wǎng)絡(luò)輿情傳播特征分析[J];情報(bào)雜志;2013年01期

5 唐超;;網(wǎng)絡(luò)情緒演進(jìn)的實(shí)證研究[J];情報(bào)雜志;2012年10期

6 肖強(qiáng);朱慶華;;Web2.0環(huán)境下的“網(wǎng)絡(luò)推手”現(xiàn)象案例研究[J];情報(bào)雜志;2012年09期

7 羅旭;;網(wǎng)絡(luò)政治參與中的輿論領(lǐng)袖研究:演進(jìn)與規(guī)制[J];探索;2012年04期

8 代大明;王中卿;李壽山;李培峰;朱巧明;;基于情緒詞的非監(jiān)督中文情感分類方法研究[J];中文信息學(xué)報(bào);2012年04期

9 李本陽;關(guān)毅;董喜雙;李生;;基于單層標(biāo)注級(jí)聯(lián)模型的篇章情感傾向分析[J];中文信息學(xué)報(bào);2012年04期

10 朱恒民;李青;;面向話題衍生性的微博網(wǎng)絡(luò)輿情傳播模型研究[J];現(xiàn)代圖書情報(bào)技術(shù);2012年05期

相關(guān)重要報(bào)紙文章 前1條

1 尹潘嶸;陶建華;;情感與認(rèn)知 相互觸發(fā)[N];計(jì)算機(jī)世界;2005年

相關(guān)博士學(xué)位論文 前4條

1 黃勝;Web評(píng)論文本的細(xì)粒度意見挖掘技術(shù)研究[D];北京理工大學(xué);2014年

2 紀(jì)雪梅;特定事件情境下中文微博用戶情感挖掘與傳播研究[D];南開大學(xué);2014年

3 李澍淞;基于蜂擁策略的網(wǎng)絡(luò)輿論演化模型研究[D];復(fù)旦大學(xué);2011年

4 張玉強(qiáng);網(wǎng)絡(luò)輿情危機(jī)的政府適度反應(yīng)研究[D];中央民族大學(xué);2011年

相關(guān)碩士學(xué)位論文 前1條

1 趙城利;基于Web的信息智能感知技術(shù)及應(yīng)用[D];國防科學(xué)技術(shù)大學(xué);2004年

，

本文編號(hào)：1990163