43中英文雙語外文文獻翻譯成品網絡新聞熱點話題中文標題用詞分析---一個復雜的網絡視角

1、<p>　　外文標題：Words Analysis of Online Chinese News Headlines about Trending Events: A Complex Network Perspective</p><p>　　外文作者：Huajiao Li, Wei Fang, Haizhong An, Xuan Huang</p><p>　　文獻出處:《Pl

2、os One》 , 2015 , 10 (3)</p><p>　　英文2309單詞，13699字符，中文3155漢字。</p><p>　　此文檔是外文翻譯成品，無需調整復雜的格式哦！下載之后直接可用，方便快捷！只需二十多元。</p><p>　　Words Analysis of Online Chinese News Headlines about Trend

3、ing Events: A Complex Network Perspective</p><p>　　Huajiao Li, Wei Fang, Haizhong An, Xuan Huang</p><p><b>　　Abstract</b></p><p>　　Because the volume of information avai

4、lable online is growing at breakneck speed, keeping up with meaning and information communicated by the media and netizens is a new challenge both for scholars and for companies who must address public relations crises.

5、 Most current theories and tools are directed at identifying one website or one piece of online news and do not attempt to develop a rapid understanding of all websites and all news covering one topic. This paper repres

6、ents an effort to inte</p><p>　　Introduction</p><p>　　With the development and popularization of information and network technology, the Internet has become the main medium from which people ob

7、tain information and news. Helping solve a serious information overload problem [1], search engines are recognized as one of the most useful and popular services on the web [2, 3]. Generally, the web (and a search engine

8、) is the first source a person turns to for information or news [4]. People have grown accustomed to inputting a few keywords into search en</p><p>　　Method of headlines’ word segmentation</p><p&g

9、t;　　We used the open source word segmentation software called Simple Chinese Word Segmentation (http://www.xunsearch.com) based on the scripting language PHP. Simple Chinese Word Segmentation employs a dictionary contai

10、ning more than 260 thousand Chinese words. The part-of-speech tagging used in this software is Peking University annotation, which contains 47 parts of speech. The input information is the headlines and the serial number

11、s of the headlines, whereas the output information consists of </p><p>　　Method of constructing words network</p><p>　　As described above, the main job of constructing the word network is to de

12、termine the nodes and edges as well as the weights of the edges. There are different ways of constructing networks, such as equivalence relationships (complete graph) [30], affiliation relationships (bipartite graph) [33

13、, 42], and so on. In this paper, in order to show the words contextual relationships in the title, we gleaned the segmented words from the news headlines according to the features of the study subject (them</p>&l

14、t;p>　　Fig. 4 shows the linear network for one title. Next, the linear networks of different headlines were superimposed; the weights of the edges are the times of the appearance of the edges between two nodes in diff

15、erent linear networks. Let graph G = (V,E,W) represent the directed weighted network in which V and E are the set of nodes and edges, and W represents the after each occurred, and then faded away to be talked about in th

16、e media only occasionally thereafter. Meanwhile, there is one notable</p><p>　　Results and Analysis</p><p>　　The topological features of the whole-sample words network</p><p>　　The

17、visualization of the whole-sample words network. After application of the Simple Chinese Word Segmentation software, we obtained 5,661 words regarding the 2010 Gulf of Mexico oil spill and 6,821 words regarding the 2011

18、Bohai Bay oil spill (after eliminating punctuations). After cleaning duplicate words, there were 1,288 different words in all the online Chinese news headlines regarding the 2010 Gulf of Mexico oil spill and 1,572 diffe

19、rent words in all the online Chinese news headlines rega</p><p>　　Discussion and Conclusion</p><p>　　Complex network method has been well used in different empirical areas [44-48]. In this paper

20、, we studied an infrequently considered but quite important method for developing a rapid and deep understanding of all the websites and all the news regarding one topic which integrates statistics, word segmentation, c

21、omplex network theory and visualization to analyze all the online news headlines’ keywords and their evolution regarding two trending events, the 2010 Gulf of Mexico oil spill and the 201</p><p>　　We present

22、ed an integrated method to analyze both the whole-sample words network and monthly-words network regarding the online news headlines of the two trending events. Through our research, we found that, as with most empirical

23、 complex networks, the words networks of online news headlines regarding the two trending events have scale-free characteristics and small-world properties, and the degree assortativity coefficients of the two whole- s

24、ample words networks are very low. By calculating </p><p>　　References</p><p>　　Chen DB, Wang GN, Zeng A, Fu Y, Zhang YC. Optimizing Online Social Networks for Information Propagation. PloS one

25、 2014; 9: e96614. doi: 10.1371/journal.pone.0096614 PMID: 24816894</p><p>　　Bharat K, BroderA. A technique for measuring the relative size and overlap of public web search engines. Computer Networks and ISD

26、N Systems 1998; 30: 379-388.</p><p>　　Risvik KM, Michelsen R. Search engines and web dynamics. Computer Networks 2002; 39: 289-302.</p><p>　　Morris MR, Teevan J, Panovich K. A Comparison of Info

27、rmation Seeking Using Search Engines and Social Networks.ICWSM 2010; 10: 23-26.</p><p>　　QiuT, Zhang ZK, Chen G. Information filtering via a scaling-based function. PloS one 2013; 8: e63531. doi: 10.1371/jou

28、rnal.pone.0063531 PMID: 23696829</p><p>　　Medo M, Zhang YC, Zhou T. Adaptive model for recommendation of news. EPL (Europhysics Letters) 2009; 88:38005.</p><p>　　Zhang ZK, Liu C. Hybrid recommen

29、dation algorithm based on two roles of social tags. International Journal of Bifurcation and Chaos 2012; 22:1250166</p><p>　　Chen D, Zeng A, Cimini G, Zhang YC. Adaptive social recommendation in a multiple c

30、ategory landscape. arXiv preprint2012; arXiv:1210.1441.</p><p>　　ShieJS. Conceptual metaphoras a news-story promoter: The cases of ENLand EILheadlines. Inter- cultural Pragmatics 2012; 9:1-21.</p>&l

31、t;p>　　Kleinnijenhuis J, Schultz F, Utz S, Oegema D. The mediating role of the news in the BP oil spill crisis 2010: How US news is influenced by public relations and in turn influences public awareness, foreign news,

32、and the share price. Communication Research 2013; 0093650213510940.</p><p>　　UtzS, Schultz F, GlockaS. Crisis communication online: How medium, crisis type and emotions affected public reactions in the Fuku

33、shima Daiichi nucleardisaster. Public Relations Review 2013; 39:4046.</p><p>　　Mahgoub H, RosnerD, Ismail N, Torkey F. AText Mining Technique Using Association Rules Extrac- tion.International journal of com

34、putational intelligence 2008; 4:21-28.</p><p>　　Tanabe L, Scherf U, Smith LH, Lee JK, Hunter L, Weinstein JN. MedMiner: an Internet text-mining tool forbiomedical information, with application to gene expres

35、sion profiling. Biotechniques 1999; 27: 12104. PMID: 10631500</p><p>　　Choi Y, JungY, Myaeng SH. Identifying controversial issues and theirsub-topics in news articles, In Intelligence and Security Informati

36、cs. SpringerBerlin Heidelberg 2010;140-153.</p><p>　　Balahur A, Steinberger R. Rethinking Sentiment Analysis in the News: fromTheoryto Practice and back. Proceeding of WOMSA 2009; 9</p><p>　　Bho

37、wmick PK. Reader perspective emotion analysis in text through ensemble based multi-label classification framework. Computer and Information Science 2009; 2:64-74.</p><p>　　Yoon J. Detecting weak signals for

38、 long-term business opportunities using text mining of web news. Expert Systems with Applications 2012; 39:12543-12550.</p><p>　　Huang CJ, Liao JJ, Yang DX, Chang TY, Luo YC. Realization of a news dissemina

39、tion agent based on weighted association rules and text mining techniques. Expert Systems with Applications 2010; 37: 6409-6413.</p><p>　　Tanasa D,Trousse B. Advanced data preprocessing for intersites web us

40、age mining. Intelligent Systems, IEEE2004; 19:59-65.</p><p>　　Li N, Wu DD. Using text mining and sentiment analysis foronline forums hotspotdetection andforecast. Decision Support Systems 2010; 48: 354-368.

41、</p><p>　　Lin C, Xie R, Guan X, Li L, Li T. Personalized news recommendation via implicitsocial experts. Information Sciences 2014; 254:1-18.</p><p>　　Wagner H, Dlotko P, MrozekM. Computational

42、 topology in text mining, In Computational Topology in Image Context. Springer Berlin Heidelberg 2012; 68-78.</p><p>　　Afzal S, Maciejewski R, Jang 丫，曰mqvist N, Ebert DS. Spatial text visualization using aut

43、omatic typographic maps. lEEETransactions on Visualization&ComputerGraphics 2012; 18: 2556-2564. PMID: 24783264</p><p>　　Gurkan A, landoli L, Klein M, Zollo G. Mediating debate through on-line large-sca

44、le argumentation: Evi- dencefrom the field. Information Sciences 2010; 180: 3686-3702.</p><p>　　Chen RC, Hsieh CH. Web page classification based on a supportvector machine usingaweighted vote schema. Expert

45、Systems with Applications 2006; 31:427-435.</p><p>　　Magerman T, Looy BV, Song X. Exploring the feasibility and accuracy of Latent Semantic Analysis based text mining techniques to detect similarity between

46、patent documents and scientific publications. Scientometrics 2010; 82: 289-306.</p><p>　　Dodds PS, Watts DJ, Sabel CF. Information exchange and the robustness of organizational networks. Proceedings ofthe Na

47、tional Academy of Sciences 2003; 100:12516-12521. PMID: 14528009</p><p>　　Barabasi AL, Jeong H, Neda Z, Ravasz E, Schubert A, Vicsek T. Evolution of the social network of scientific collaborations. Physica

48、A: Statistical Mechanics and its Applications 2002; 311:590-614.</p><p>　　Hanaki N, Peterhansl A, Dodds PS, Watts DJ. Cooperation in evolving social networks. Management Science 2007; 53:1036-1050.</p>

49、<p>　　Li HJ, An HZ, HuangJC, Gao XY, ShYL. Correlation of the holding behaviourof the holding-based network of Chinese fund management companies based on the node topological characteristics. Acta Phys.Sin.2014;

50、63:48901-048901.</p><p>　　Gao X, An H, Zhong W. Features of the Correlation Structure of Price Indices. PLoSone2013; 8: e61091. doi: 10.1371/journal.pone.0061091 PMID: 23593399</p><p>　　Serrano

51、MA, BogunaM. Topology of the world trade web. Physical ReviewE2003; 68: 015101. PMID: 12935184</p><p>　　Zhang CJ, Zeng A. Behavior patterns of online users and the effect on information filtering. Physica A:

52、 Statistical Mechanics and its Applications 2012; 391:1822-1830.</p><p>　　Hu H., Wang X. Evolution of a large online social network. Physics LettersA2009; 373:1105-1110.</p><p>　　Piraveenan M, P

53、rokopenko M, Zomaya A. Assortative mixing in directed biological networks. IEEE/ ACM Transactions on Computational Biology and Bioinformatics (TCBB) 2012; 9: 66-78.</p><p>　　Newman MEJ. The structure andfunc

54、tion of complex networks. SIAM review2003; 45:167-256.</p><p>　　Li H, An H, Gao X, Huang J, Xu Q. On the topological properties of the cross-shareholding networks of listed companies in China: Taking shareho

55、lders’ cross-shareholding relationships into account. Physica A: Statistical Mechanics and its Applications 2014; 406:80-88.</p><p>　　Brandes U. A faster algorithm for betweenness centrality. Journal of Mat

56、hematical Sociology 2001; 25:163-177.</p><p>　　Ebel H, Mielsch LI, Bornholdt S. Scale-free topology of e-mail networks. Physical ReviewE2002; 66: 035103. PMID: 12366171</p><p>　　Blondel VD, Guil

57、laumeJL, Lambiotte R, Lefebvre E. Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment 2008; 10:10008.</p><p>　　Palla G, Barabasi A L, VicsekT. Quantifying

58、 social group evolution. Nature 2007; 446: 664-667.</p><p>　　PMID: 17410175</p><p>　　Li H, Fang W, An H, Yan L. The shareholding similarity of the shareholders of the worldwide listed energy co

59、mpanies based on a two-mode primitive network and a one-mode derivative holding-based network. PhysicaA: Statistical Mechanics and its Applications 2014; 415:525-532.</p><p>　　Newman MEJ, Assortative mixing

60、 in networks. Physical review letters 2002; 89: 208701. PMID: 12443515</p><p>　　Qi H, An H, Hao X, Zhong W, Zhang Y. Analyzing the International Exergy Flow Network of Ferrous Metal Ores. PloS one 2014; 9: e

61、106617. doi: 10.1371/journal.pone.0106617 PMID: 25188407</p><p>　　Hao X, An H, Liu X, Gao X, Cong L. Analysis on main mineral products in international trade. Resources & Industries 2013; 15:3543</p&g

62、t;<p>　　An H, Gao X, Fang W, Huang X, Ding Y. The role of fluctuating modes of autocorrelation in crude oil prices. PhysicaA: Statistical Mechanics and its Applications 2014; 393:382-90.</p><p>　　An H

63、, Zhong W, Chen Y, Li H, Gao X. Features and evolution of international crude oil trade relationships: A trading-based network analysis. Energy 2014; 74: 254-259.</p><p>　　An J, An H, Yang G. Relation of fi

64、nanceial institutions and listed mining entities in equity financing based on complex network. Resources & Industries 2014; 16:124-1</p><p>　　網絡新聞熱點話題中文標題用詞分析---一個復雜的網絡視角</p><p>　　Huajiao Li

65、, Wei Fang, Haizhong An, Xuan Huang</p><p><b>　　摘要</b></p><p>　　由于網絡上的信息量以是爆炸式的速度進行增長，因此要跟上媒體和網民傳達的意思和信息，對于學者和那些必須解決公關危機的公司來說都是一個全新的挑戰(zhàn)。當前的大多數(shù)理論和工具都是針對某一個網站或者是某一條在線新聞的，而不是去嘗試快速了解所有網站和

66、所有涉及同一個主題新聞報道的情況。在本文中，通過2011年渤海灣漏油事件和2010年墨西哥灣漏油事件這兩個樣本事件，整合統(tǒng)計數(shù)據、詞的切分、復雜網絡環(huán)境以及可視化去嘗試分析中國在線新聞標題中的關鍵字和詞語的關系。我們搜集了來自中國最受歡迎的搜索引擎--百度搜索結果中關于這兩個熱點事件的所有新聞頭條。我們使用簡體中文分詞軟件將所有標題分割成單詞，然后以單詞作為節(jié)點，以相鄰詞的關系為邊，利用整個樣本和每月的用詞量去搭建詞匯網。最后，基于新聞

67、標題，我們開發(fā)了一個綜合機制來分析詞匯網絡的特征，這些新聞標題可以記錄關于特定事件的新聞中的所有關鍵字，并因此可以深入而迅速地追蹤新聞的動態(tài)發(fā)展情況。</p><p><b>　　引言</b></p><p>　　伴隨著信息和網絡技術的快速發(fā)展和普及，互聯(lián)網已經成為人們獲取信息和新聞的主要媒介。在幫助解決嚴重的信息過載問題[1]方面，搜索引擎被認為是網絡上最有用和最受

68、歡迎的服務之一[2,3]。通常來說，網絡（和搜索引擎）是向人們傳遞信息或新聞的第一來源[4]。人們已經習慣于在搜索引擎中輸入幾個關鍵詞，然后點擊一個或多個標題，更多人意識到網絡新聞在輿論傳播中起著重要的作用。因此，了解不同新聞來源呈現(xiàn)信息的方式就非常重要。標題是新聞的重要組成部分，不僅是提供或關聯(lián)新聞內容的要點，而且要必須吸引讀者的注意力[9]。有學者已經提供相關證據表明公共關系、公眾意識和新聞之間存在關聯(lián)[10]。</p>

69、<p><b>　　新聞標題的分詞方法</b></p><p>　　我們使用了基于腳本語言PHP的開源分詞軟件--簡體中文分詞（http://www.xunsearch.com）。 簡體中文分詞使用詞典中超過26萬個中文詞匯。 本軟件中使用的詞性標注是北大的注釋，其中包含47個詞類。 要輸入信息是頭條新聞的標題和序列號，而輸出的信息是由詞匯的序號、詞匯、詞類的詞語部分和標題的序

70、列號組成。</p><p><b>　　構建詞匯網絡的方法</b></p><p>　　如上所述，構建詞匯網絡的主要工作是確定詞匯的節(jié)點以及詞匯邊界的權重。 構建詞匯網絡有不同的方式，如等價關系（完整圖）[30]、從屬關系（二分圖）[33,42]等等。 在本文中，為了顯示標題中詞匯的上下文關系，我們根據研究主題（標題）的特征從新聞標題中搜集了分詞，然后根據標題中詞匯的

71、序列，即前一個節(jié)點作為起始節(jié)點，和前一節(jié)點之后的節(jié)點作為終止節(jié)點，我們將每個詞作為節(jié)點并將節(jié)點與詞匯的邊界建立聯(lián)系。</p><p>　　圖4顯示的是一個標題的線性網絡。 接下來，我們疊加了不同標題的線性網絡; 詞與詞之間邊的權重是在不同線性網絡中兩個節(jié)點之間邊的出現(xiàn)次數(shù)。 假設圖G =（V，E，W）表示有向加權網絡，其中V和E是節(jié)點和邊的集合，W表示其發(fā)生之后，然后在媒體中逐漸消失，這在之后會略微談到。 與此同

72、時，有關這兩個熱電事件的新聞有一個顯著的區(qū)別： 在2010年墨西哥灣漏油事故發(fā)生后媒體就首次報道這事件，但2011年渤海灣漏油事件是在其發(fā)生一個月后媒體再進行報道的。</p><p>　　詞匯網絡的構建（根據標題）</p><p><b>　　結果與分析</b></p><p>　　全樣本詞匯網絡的拓撲特征</p><p&g

73、t;　　全樣本詞匯網絡的可視化。在應用簡體中文分詞軟件后，我們獲得了關于2010年墨西哥灣漏油事件的5,661個詞匯和關于2011年渤海灣漏油事件（標點符號除外）的6,821詞匯。在清理重復詞語后，2010年所有在線中文新聞標題中關于2010年墨西哥灣漏油事件以及所有關于2011年渤海灣漏油事件的在線中文新聞標題中一共有1,572個不同詞語，這意味著有1,288個節(jié)點是關于墨西哥的全樣本詞網絡以及1,572個節(jié)點是關于渤海的全樣本詞網絡

74、。圖中給出了關于墨西哥和渤海的兩個全樣本詞匯網絡的可視化結果（節(jié)點的顏色由節(jié)點所屬的同一ID來確定）。</p><p>　　兩個熱點事件全樣本詞匯網絡的可視化結果</p><p><b>　　探討與結論</b></p><p>　　復雜網絡法已被很好地用于不同的實證領域[44-48]。 在本文中，我們研究了一種不常用的但相當重要的方法，用于快速

75、深入地了解所有網站和同一主題的所有新聞，這其中要去整合數(shù)據統(tǒng)計、分詞、復雜網絡理論以及可視化以分析所有在線新聞標題中的關鍵詞及其關于2010年墨西哥灣漏油事件和2011年渤海灣漏油事件兩個熱點事件的演變趨勢。</p><p>　　我們提出了一個綜合性的方法來分析整個樣本詞匯網絡和每月詞匯網絡關于這兩個熱點事件的在線新聞標題。通過我們的研究我們發(fā)現(xiàn)，與大多數(shù)實證的復雜網絡一樣，關于這兩個熱點事件的在線新聞頭條網絡具

76、有無標度特征和微觀屬性，并且這兩個全樣本詞匯網絡的同配性系數(shù)程度非常低。通過計算節(jié)點的拓撲特征，我們得到了全樣本詞網絡的關鍵詞和月詞網絡的關鍵詞。同時，我們也得到了詞的內在關系和演變。與搜索引擎中關于2010年墨西哥灣事件相比，如果我們想要更準確地收集關于詞網的信息，我們必須探索更多搜索新聞的方法。因此，今后我們可以擴展數(shù)據搜索的方法，并根據實際情況嘗試構建頭條新聞詞匯網絡。當然，有些標題是具有煽動性或誤導性的，并不能反映新聞內容的真實

77、內容。因此，在下一步的工作中我們可以鑒定出一種判斷新聞標題與內容之間相關程度的新方法。</p><p><b>　　參考文獻</b></p><p>　　Chen DB, Wang GN, Zeng A, Fu Y, Zhang YC. Optimizing Online Social Networks for Information Propagation. Pl

78、oS one 2014; 9: e96614. doi: 10.1371/journal.pone.0096614 PMID: 24816894</p><p>　　Bharat K, BroderA. A technique for measuring the relative size and overlap of public web search engines. Computer Networks a

79、nd ISDN Systems 1998; 30: 379-388.</p><p>　　Risvik KM, Michelsen R. Search engines and web dynamics. Computer Networks 2002; 39: 289-302.</p><p>　　Morris MR, Teevan J, Panovich K. A Comparison o

80、f Information Seeking Using Search Engines and Social Networks.ICWSM 2010; 10: 23-26.</p><p>　　QiuT, Zhang ZK, Chen G. Information filtering via a scaling-based function. PloS one 2013; 8: e63531. doi: 10.13

81、71/journal.pone.0063531 PMID: 23696829</p><p>　　Medo M, Zhang YC, Zhou T. Adaptive model for recommendation of news. EPL (Europhysics Letters) 2009; 88:38005.</p><p>　　Zhang ZK, Liu C. Hybrid re

82、commendation algorithm based on two roles of social tags. International Journal of Bifurcation and Chaos 2012; 22:1250166</p><p>　　Chen D, Zeng A, Cimini G, Zhang YC. Adaptive social recommendation in a mult

83、iple category landscape. arXiv preprint2012; arXiv:1210.1441.</p><p>　　ShieJS. Conceptual metaphoras a news-story promoter: The cases of ENLand EILheadlines. Inter- cultural Pragmatics 2012; 9:1-21.</p&g

84、t;<p>　　Kleinnijenhuis J, Schultz F, Utz S, Oegema D. The mediating role of the news in the BP oil spill crisis 2010: How US news is influenced by public relations and in turn influences public awareness, foreign

85、news, and the share price. Communication Research 2013; 0093650213510940.</p><p>　　UtzS, Schultz F, GlockaS. Crisis communication online: How medium, crisis type and emotions affected public reactions in th

86、e Fukushima Daiichi nucleardisaster. Public Relations Review 2013; 39:4046.</p><p>　　Mahgoub H, RosnerD, Ismail N, Torkey F. AText Mining Technique Using Association Rules Extrac- tion.International journal

87、of computational intelligence 2008; 4:21-28.</p><p>　　Tanabe L, Scherf U, Smith LH, Lee JK, Hunter L, Weinstein JN. MedMiner: an Internet text-mining tool forbiomedical information, with application to gene

88、expression profiling. Biotechniques 1999; 27: 12104. PMID: 10631500</p><p>　　Choi Y, JungY, Myaeng SH. Identifying controversial issues and theirsub-topics in news articles, In Intelligence and Security Inf

89、ormatics. SpringerBerlin Heidelberg 2010;140-153.</p><p>　　Balahur A, Steinberger R. Rethinking Sentiment Analysis in the News: fromTheoryto Practice and back. Proceeding of WOMSA 2009; 9</p><p>

90、;　　Bhowmick PK. Reader perspective emotion analysis in text through ensemble based multi-label classification framework. Computer and Information Science 2009; 2:64-74.</p><p>　　Yoon J. Detecting weak signa

91、ls for long-term business opportunities using text mining of web news. Expert Systems with Applications 2012; 39:12543-12550.</p><p>　　Huang CJ, Liao JJ, Yang DX, Chang TY, Luo YC. Realization of a news dis

92、semination agent based on weighted association rules and text mining techniques. Expert Systems with Applications 2010; 37: 6409-6413.</p><p>　　Tanasa D,Trousse B. Advanced data preprocessing for intersites

93、web usage mining. Intelligent Systems, IEEE2004; 19:59-65.</p><p>　　Li N, Wu DD. Using text mining and sentiment analysis foronline forums hotspotdetection andforecast. Decision Support Systems 2010; 48: 35

94、4-368.</p><p>　　Lin C, Xie R, Guan X, Li L, Li T. Personalized news recommendation via implicitsocial experts. Information Sciences 2014; 254:1-18.</p><p>　　Wagner H, Dlotko P, MrozekM. Computa

95、tional topology in text mining, In Computational Topology in Image Context. Springer Berlin Heidelberg 2012; 68-78.</p><p>　　Afzal S, Maciejewski R, Jang 丫，曰mqvist N, Ebert DS. Spatial text visualization usi

96、ng automatic typographic maps. lEEETransactions on Visualization&ComputerGraphics 2012; 18: 2556-2564. PMID: 24783264</p><p>　　Gurkan A, landoli L, Klein M, Zollo G. Mediating debate through on-line lar

97、ge-scale argumentation: Evi- dencefrom the field. Information Sciences 2010; 180: 3686-3702.</p><p>　　Chen RC, Hsieh CH. Web page classification based on a supportvector machine usingaweighted vote schema. E

98、xpert Systems with Applications 2006; 31:427-435.</p><p>　　Magerman T, Looy BV, Song X. Exploring the feasibility and accuracy of Latent Semantic Analysis based text mining techniques to detect similarity be

99、tween patent documents and scientific publications. Scientometrics 2010; 82: 289-306.</p><p>　　Dodds PS, Watts DJ, Sabel CF. Information exchange and the robustness of organizational networks. Proceedings of

100、the National Academy of Sciences 2003; 100:12516-12521. PMID: 14528009</p><p>　　Barabasi AL, Jeong H, Neda Z, Ravasz E, Schubert A, Vicsek T. Evolution of the social network of scientific collaborations. Ph