本文關(guān)鍵詞：社交網(wǎng)絡(luò)中教育資源推薦的目標(biāo)用戶挖掘研究　出處：《中央民族大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

  更多相關(guān)文章： 社交網(wǎng)絡(luò) 資源推薦 用戶挖掘 特征提取 用戶分類(lèi)

【摘要】：互聯(lián)網(wǎng)和信息技術(shù)正處于飛躍發(fā)展時(shí)期,隨之產(chǎn)生的網(wǎng)絡(luò)信息內(nèi)容日漸豐富,網(wǎng)絡(luò)數(shù)據(jù)也日漸增加。在這個(gè)互聯(lián)網(wǎng)和教育并行發(fā)展的時(shí)代,人們更樂(lè)于借助社交網(wǎng)絡(luò)平臺(tái)尋求更多獲取教育資源的途徑,然而在有限的時(shí)間內(nèi)準(zhǔn)確快速獲取需要的教育資源成為研究的重點(diǎn)。于是,數(shù)據(jù)挖掘技術(shù)、信息推薦技術(shù)、分類(lèi)技術(shù)等前言技術(shù)應(yīng)運(yùn)而生。新浪微博社交網(wǎng)絡(luò)平臺(tái)提供了一個(gè)全面的分析用戶興趣的龐大數(shù)據(jù)源,成為近幾年研究的熱點(diǎn)。如何有效地給用戶推薦有用的信息,需要找到關(guān)注某一資源或主題的目標(biāo)用戶,其中重要的研究工作即如何能準(zhǔn)確分析目標(biāo)用戶的興趣偏向。論文的主要研究?jī)?nèi)容如下:(1)研究社交網(wǎng)絡(luò)目標(biāo)用戶數(shù)據(jù)采集相關(guān)技術(shù)和預(yù)處理的過(guò)程。其中,數(shù)據(jù)采集主要是利用基于Scrapy的網(wǎng)絡(luò)爬蟲(chóng)技術(shù),得到用于實(shí)驗(yàn)分析的用戶基本屬性信息和博文信息。數(shù)據(jù)預(yù)處理包括對(duì)語(yǔ)料去除停用詞、分詞等。(2)基于內(nèi)容的用戶興趣特征提取與目標(biāo)用戶挖掘研究。利用LDA模型進(jìn)行主題建模提取用戶特征,將直接使用LDA建模和改進(jìn)的LDA建模的實(shí)驗(yàn)數(shù)據(jù)進(jìn)行對(duì)比分析,結(jié)果發(fā)現(xiàn)改進(jìn)的LDA模型進(jìn)行主題建模提取的特征主題準(zhǔn)確性更高。選擇基于聚類(lèi)的半監(jiān)督算法作為目標(biāo)用戶挖掘研究的分類(lèi)算法,實(shí)驗(yàn)結(jié)果顯示分類(lèi)結(jié)果能較準(zhǔn)確表示用戶的關(guān)注傾向。(3)系統(tǒng)可視化技術(shù)和實(shí)現(xiàn)的研究。主要通過(guò)Java后端數(shù)據(jù)的處理和HTML CSS技術(shù)的前端展示實(shí)現(xiàn),將目標(biāo)用戶分類(lèi)結(jié)果界面化顯示。文章研究的最終目的是為小學(xué)階段學(xué)生以及關(guān)注小學(xué)教育資源的學(xué)生家長(zhǎng)服務(wù),系統(tǒng)能快速準(zhǔn)確獲取相關(guān)教學(xué)資源。同時(shí),更有助于學(xué)校教育管理者開(kāi)展工作,利用該系統(tǒng)可以將教學(xué)資源準(zhǔn)確且有針對(duì)性的推薦給目標(biāo)用戶。
[Abstract]:Internet and information technology are in a period of rapid development, resulting in the increasingly rich content of network information, network data is also increasing. In this era of parallel development of Internet and education. People are more willing to seek more access to educational resources with the help of social network platform. However, accurate and rapid access to educational resources in a limited time has become the focus of research. Therefore, data mining technology. Information recommendation technology, classification technology and other preface technology came into being. Sina Weibo social network platform provides a comprehensive analysis of user interests of a huge data source. It has become a hot topic in recent years. How to effectively recommend useful information to users needs to find the target users who pay attention to a certain resource or topic. The important research work is how to accurately analyze the interest bias of the target users. Research on the social network target user data acquisition technology and preprocessing process. Data acquisition is mainly based on Scrapy based web crawler technology to obtain user basic attribute information and blog information for experimental analysis. Data preprocessing includes the removal of discontinuation words from corpus. Segmentation, etc.) based on the content of user interest feature extraction and target user mining research. Using LDA model for topic modeling to extract user features. The experimental data of direct use of LDA modeling and improved LDA modeling will be compared and analyzed. The results show that the improved LDA model is more accurate in feature topic extraction. The clustering based semi-supervised algorithm is selected as the classification algorithm of target user mining research. The experimental results show that the classification results can accurately indicate the user's concern. System visualization technology and implementation research. Mainly through the Java back-end data processing and HTML CSS technology front-end display realization. The final purpose of this paper is to provide services for primary school students and parents concerned about primary education resources, and the system can quickly and accurately obtain relevant teaching resources. It is also helpful for school education administrators to carry out their work, and the system can be used to recommend the teaching resources to the target users accurately and pertinently.
【學(xué)位授予單位】：中央民族大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP311.13;G434

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 儲(chǔ)濤濤;;基于LDA主題模型的用戶興趣發(fā)現(xiàn)方法[J];軟件;2016年12期

2 陳科文;張祖平;龍軍;;文本分類(lèi)中基于熵的詞權(quán)重計(jì)算方法研究[J];計(jì)算機(jī)科學(xué)與探索;2016年09期

3 李佳亮;;淺析數(shù)據(jù)挖掘技術(shù)及其應(yīng)用[J];商業(yè)故事;2015年25期

4 潘澄;陳宏;;我國(guó)學(xué)習(xí)資源個(gè)性化推薦研究進(jìn)展[J];現(xiàn)代教育科學(xué);2015年04期

5 梁宏;許南山;盧罡;;新浪微博用戶及其微博特征分析[J];計(jì)算機(jī)工程與應(yīng)用;2015年07期

6 宗慧;劉金嶺;;基于短文本信息流的熱點(diǎn)話題檢測(cè)[J];數(shù)據(jù)采集與處理;2015年02期

7 俞忻峰;;社交網(wǎng)絡(luò)挖掘方案研究[J];現(xiàn)代電子技術(shù);2015年04期

8 彭希羨;朱慶華;劉璇;;微博客用戶特征分析及分類(lèi)研究——以“新浪微博”為例[J];情報(bào)科學(xué);2015年01期

9 李軼凡;;淺析新浪微博用戶體驗(yàn)優(yōu)勢(shì)與提升空間[J];中國(guó)報(bào)業(yè);2014年16期

10 歐衛(wèi);謝贊福;謝彬彬;歐繽憶;;基于LDA模型的社交網(wǎng)絡(luò)主題社區(qū)挖掘[J];計(jì)算機(jī)與現(xiàn)代化;2014年08期

，

本文編號(hào)：1399591