首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 31 毫秒
1.
Zusammenfassung. Anhand zweier Anwendungsbeispiele aus der Induktiven Logischen Programmierung (ILP) wird die Fragestellung des Erlernens von Klassifikatoren für relational strukturierte Beispiele betrachtet. Es wird auf wesentliche Techniken und Fragen des graphbasierten und logischen Lernens, insbesondere auf Fragen des Aufwands eingegangen. Darauf aufbauend wird das Verfahren Indigo zur effizienten graphentheoretischen Transformation des relationalen in ein merkmalsbasiertes Lernproblem vorgestellt, welches den Einsatz von klassischen Lernverfahren wie CAL3 oder ID3 erlaubt. Am Beispiel von Indigo wird gezeigt, daß spezifisch graphentheoretische Ansätze und Algorithmen einen wertvollen Beitrag zur Behandlung von relationalen Lernproblemen erbringen können, der mit rein logischen Methoden, wie sie derzeit in der ILP untersucht werden, nicht entsprechend erzielt werden kann. Eine Aufgabenstellung dieser Arbeit ist deshalb die Gegenüberstellung von graphbasierten und logischen Lerntechniken am Beispiel des hybriden Lernverfahrens Tritop, welches Grundideen aus der Graphentheorie in einem logischen Gewand verwirklicht. Tritop verwendet die sog. a\alpha-Subsumtion, eine Spezialisierung der bekannten q\theta-Subsumtion, zur Klassifikation von Beispielen. Neben den beiden Lernverfahren gehen wir auf die effiziente Konstruktion von Klassenprototypen mit einem konnektionistischen Verfahren ein. Abstract. In this article, we discuss the problem of learning a classifier for relational data with two application examples from Inductive Logic Programming (ILP). We present important techniques and questions of graph based and logical learning, especially its computational complexity. Based on this discussion, the learning system Indigo is described, which relies on the efficient graph based transformation of the relational learning task into a feature based problem, which can be solved with classical feature based methods like CAL3 or ID3. Using Indigo as an example, it can be seen that graph theoretical approaches can make an important contribution to relational machine learning, that cannot be achieved in the same way with logical methods stemming from the field of ILP. Therefore, one concern of this work is the comparison of graph theoretical and logical learning methods. This is done by describing the hybrid learning system Tritop that uses graph theoretical methods together with a logical notation and ILP concepts like subsumtion of clauses. Tritop uses the so called a\alpha-subsumption, a restriction of the well known q\theta-subsumption. As an extension of the two learning systems, we also discuss the efficient construction of class prototypes using a neual network.  相似文献   

2.
This study investigates the relationship between media coverageof Helmut Kohl in seven leading German print media and the opinionsof the German general public about the politician between 1975and 1984. For the content analysis evaluative assessments aboutHelmut Kohl on six different dimensions of characteristics werecoded. The analysis of public opinion is based on 72 representativesurveys in which respondents expressed their evaluations ofthe politician. The two time series were compared by means ofcross-lagged correlations. The whole period was first examinedwith aggregations of three-months-intervals; then closer attentionwas paid to the time period since Helmut Kohl took office aschancellor on the basis of monthly intervals. In both cases,the results show that evaluation shifts in the media precedesimilar evaluation shifts in public opinion with a time lagof about three to six months for the whole period of investigation,and a somewhat shorter time lag for the time of his chancellorship.Evaluation shifts in the political magazines Der Spiegel andStern were more closely related to public opinion than evaluationshifts in the national dailies.  相似文献   

3.
In this section the International Journal of Public OpinionResearch reviews in a very condensed manner articles which haverecently been published in other journals and which relate tothe field of public opinion research. This section does notaim to give readers a comprehensive insight to particular studies,but rather to alert them to ideas and results that may be linkedto their own work. In order to facilitate orientation to thereader, the reviews are grouped according to the following topics: (1) Politics, Elections and Social Values; (2) Public Opinion on Current Issues; (3) Methodology of Public Opinion Polling; (4) Polls as a Social and Political Factor; (5) Mass Media and Public Opinion.  相似文献   

4.
Current Critical Problems in Agenda-Setting Research   总被引:2,自引:0,他引:2  
After experiencing steady development over the past three decades,agenda-setting research appears to be in a period of flux. Thispaper discusses three current and critical problems that agenda-settingresearch has been facing recently: the problems of process,identity, and environment. These problems are critical becauseeach has implications that might call into question the valueof agenda-setting theory. The process problem concerns the natureof the agenda-setting process, specifically, the degree to whichthe agenda-setting process is automatic and unthinking. Theidentity problem asks whether the new concept of attribute agendasetting will become indistinguishable from framing or traditionalpersuasion research. The environment problem asks if the developmentof communication technology and the subsequent growth in thenumber and variety of news outlets will minimize the impactof media agenda setting at the social level, leading to fragmentationof the public agenda. After examining each of the problems,I suggest that the agenda-setting perspective is still worthpursuing, and I present an agenda that agenda-setting researchshould address for its future development.  相似文献   

5.
The Media and Economic Voting in Israel   总被引:2,自引:0,他引:2  
The premise of the economic voting hypothesis is that citizensvote for the government if the national economy is doing well;otherwise, they vote against it. The causal chain of effectsin the economic voting hypothesis starts from the objectiveeconomic indicators, moves on to voter perceptions of the economy,and then to the vote. Under the competence hypothesis that isanalyzed here, this causal chain is slightly modified, withthe objective economy as the first step, voter perceptions ofthe economic competence of the government as the second step,and then the vote. It is argued that citizens learn about thestate of the national economy mainly from the media. Yet, onlyfew studies of economic voting include empirical analyses ofthe media's role in it. The findings support the competencehypothesis and the important role of the media in it. They underlinethe valuable contribution of media effects theories, and especiallymedia framing and priming, to our understanding of economicvoting.  相似文献   

6.
For some time now, theorists and investigators in the fieldof public opinion have discussed how opinion, public and thecombination of the two into a single term should be conceptualized.With regard to the first concept mentioned, the difference betweenattitude and opinion has been a central focus of dispute. Thisarticle demonstrates how a structural equation model, whereopinions are included as dependent variables, and attitudes,among other relevant variables, as explanatory variables, maycontribute to the discussion. Application of such a model, whichmay be considered the reflection of an attitude and opinionsystem, again forces the researcher to conceptualize the publicinvolved.  相似文献   

7.
Positive news is advantageous for a party, but the effects ofnegative news are less obvious, especially in a multiparty system.One possibility is an indecisionmodel in which negative informationwould push voters to the undecided category. An alternativeis a proportional model in which voters would move to otherparties in proportion to their share of the vote. A third scenariowould correspond to a proximity model in which the shift wouldbe to parties holding viewpoints similar to those of the criticizedparty. The three models were tested using the ideodynamic modelor the 1994 elections in the Netherlands and Germany. The explanatoryvariables were content analysis data obtained daily for majorprint and electronic news media. The dependent variable wasweekly survey data of party preferences of voters. The analysisshowed the proportional model to be the least plausible withthe proximity model being promising. Campaign strategists startingfrom the proximity model will both promote negative news aboutother parties and adopt their issue positions.  相似文献   

8.
Since their beginnings pre-election polls have been under attackfrom politicians and journalists. One of the most fundamentalcriticisms of polls is that they can influence the outcome ofelections. This article investigates news media reporting ofpoll results and comments on public opinion research beforeFederal elections in Germany. It presents empirical findingsfor the quantity as well as the formal and substantial qualityof this press coverage. The database is a content analysis of443 pre-election poll articles published in Germany's leadingdailies Frankfurter Allgemeine Zeitung (FAZ), Frankfurter Rundschau(FR), Süddeutsche Zeitung (SZ), and Die Welt (Welt) between1980 and 1994. The quantity of news media reporting of publicopinion polls has improved over the years. Now, poll storiesare a standard feature of German newspapers. The frequency ofreports depends on the conditions of each election—e.g.on the expected closeness of the election outcome. The formalquality could be better—that is the conformity to AAPORstandards. Horse-race journalism isn't found as often as inthe USA. Journalists often use poll results to predict an electionoutcome. Moreover, the analysis revealed an ambivalent relationshipbetween liberal journalists and polls, while the reporting ofconservative journalists is more in favor of opinion research.  相似文献   

9.
In this article the authors argue that studies of the influenceof the mass media on public support for presidential candidatesin primary elections should return to the social pyschologicalmodel of attitude change. This model should take into account,however, that in the real world, unlike laboratory settings,randomization and complete controls of variables are not possible,and rational and strategic choice considerations come into play,along with the cumulative effects of new information on aggregateopinion. The variables found to be relevant to attitude change,especially the characteristics of the communicator and the communication,have not been studied systematically in natural settings. Littleis known about how new information unrelated to electoral ‘momentum’affects the public's evaluations and support for candidates.A fuller explanation is needed of what are complex media influenceson candidate support in primary elections—including especiallythose messages conveyed through the media which tend to be distinctiveand most accessible to memory: information about the viabiltiyof candidates; positive messages and images conveyed directlyfrom the candidates themselves and (to a lesser extent) fromother major party leaders who support or oppose them; and, ofspecial theoretical importance, the ostensibly credible newscommentary about the candidates.  相似文献   

10.
This article documents public opinion research activities inMexico in the 1940s and the role played by Hungarian professorLászló Radványi, who immigrated to thatcountry at the height of World War II. Our research relies onseveral of Radványi's publications archived in differentcountries, as well as on interviews with family, acquaintances,and experts on the work of his wife, the German poet Anna Seghers.During his years in Mexico, Radványi founded the ScientificInstitute of Mexican Public Opinion, in 1941, and the InternationalJournal of Opinion and Attitude Research, in 1947—a forefatherof today's IJPOR. He was also a founding member of WAPOR. Hisearly "sample surveys" raised important methodological issuesand recorded opinion results that reflect the vibrant timesof war and policy making in a modernizing country. However,Radványi's contribution to the profession has been virtuallyforgotten. Until now, accounts about how public opinion researchbegan in Mexico either ignored Radványi's works or reducedhis ten years of survey research to a single footnote. Thisarticle is an attempt to fill this enormous omission and highlightsome of Radványi's contributions to these early stagesof survey research.  相似文献   

11.
‘Here is a book that makes you want to shout, "Finally!"’That is the opening sentence of a recent book review that theauthor of this review wrote for the International Journal ofPublic Opinion Research about Campaigning for Hearts and Minds:How Emotional Appeals in Political Ads Work by Ted Brader (2006).The remark refers to the fact that Brader’s book focuseson an issue that has thus far received very little attentionin the field of communication studies, or, for that matter,in the fields of political science or sociology, namely, themajor role emotions play in explaining collective behavior,social decisions, opinion formation and—especially inthe case at hand—the effects of media coverage on opinionformation. Now, in the form of Katrin Döveling’scomprehensive new  相似文献   

12.
Ambivalent Social Networks and Their Consequences for Participation   总被引:2,自引:0,他引:2  
Does perceived disagreement in political discussion help orhinder citizens’ political participation? Some argue thatdisagreement prompts reflection, perspective-taking, and tolerance.Challengers argue that disagreement fosters ambivalence andhinders participatory activities and turnout. One seminal studythat tackled this dilemma formulated the ‘cross-pressures’hypothesis (Lazarsfeld, Berelson, & Gaudet, 1944/1968),which posited that the more individuals are betwixt and betweenconflicting social positions, the longer the time for theirvote intention to crystallize (and the lower the likelihoodthey would vote). This paper offers a critique and refinementof the cross-pressures hypothesis. First, previous studies confoundedintra-individual and structural sources of cross-pressures.Second, past operationalizations of exposure to disagreementfocused on the sheer amount of opposition to the individual’spoint of view, rather than his or her exposure to two conflictingpoints of view. A new measure—network ambivalence—isproposed to capture the latter dynamic. Conceptual and methodologicalrefinements of the cross-pressures hypothesis are tested ona representative sample of voting-age respondents in the UnitedStates, interviewed on the American National Election Study2000 panel (N=1,555). Results suggest that not only were thesepressures hardly detrimental to participation, but they alsofacilitated the formation of considered electoral preferences.  相似文献   

13.
This study content analyzed a leading Korean newspaper to examinethe different framing of presidents Kim Young-sam and Roh Moo-hyun,and to look at whether there was a relationship between mediaframing and public opinion of the two presidents. The news storiesabout the presidents were analyzed in terms of topics, sources,the wording the presidents were referred to, and the contentof presidential quotations. Public opinion was investigatedwith the help of two surveys. Results show that President Rohwas predominantly referred to within a politics frame and hada lower approval among the population, mostly in terms of hisbeing negatively characterized as a politician, while PresidentKim was represented more within a policy frame, and was evaluatedin light of his policies among the public. The study discussesthe significance of examining media framing of a president,and the necessity for further experimental study of the primingeffect of media frames of a president.  相似文献   

14.
An analysis of a multiyear General Social Survey data set supportsprevious findings (Gaddy 1986; Gortmaker et al. 1990) that,with rigorous controls in place, the negative correlation betweentelevision viewing and achievement disappears. Results of ahierarchical regression procedure reveal that television viewingis not a significant predictor of adult vocabulary, over andabove demographic variables and newspaper readership. But thefindings suggest that new methods of measuring television consumptionmay be needed. Based on 1993 GSS data, the analysis revealsthat frequent viewing of certain types of television programmingis related positivelyand significantly to adult vocabulary scores.The finding highlights the need for new methods of measuringtelevision consumption, for the traditional hours-per-day methodof measuring television viewership may be preventing researchersfrom understanding the true relationship that exists betweentelevision and learning.  相似文献   

15.
In this section the International Journal of Public OpinionResearch reviews articles that have recently been publishedin peer-refereed journals and which broadly relate to the fieldof public opinion. The intention is not to give an exhaustiveoverview of a given study but rather to alert our readers tointeresting ideas and research in our field. For this issuethanks are due to Ken’ichi Ikeda (University of Tokyo)for help in compiling the reviews. Arai, Kiichiro (2006). A mechanism of political participation:Experience and evaluation. Review of Electoral Studies, 6, 5–24. This paper examines how citizens are engaged in political activities,with a focus on their experience of political participationand their evaluation of the experience. Using data from theJapanese Election and Democracy Study 2000 survey, an empiricalanalysis shows that people who positively evaluate their experienceof participating in political activity (such as helping election  相似文献   

16.
This study provides a quantitative content analysis of the depictionsof sex roles in the American news magazine Time and the WestGerman news magazine Stern from 1969 to 1988. The underlyingassumption of the study is that advertisements reflect changesin social reality. While most studies focus on explicit aspectsof gender role depictions such as jobs, activities, interactionsand situations, this content analysis also covers implicit aspectsof role depictions such as posture, gestures, and facial expression.Some authors have claimed that these are powerful means of transmittingtraditional sex role stereotypes of women as being submissiveand of men as being dominant. Results indicate that while therehave been considerable changes in the explicit presentationof male and female roles, the implicit presentation remainedquite consistent over the 20-year period. According to thesefindings, sex roles may have changed primarily on the surface,while nonverbal sex role stereotypes still linger on. Differencesbetween the developments in Germany and the US are outlined.  相似文献   

17.
18.
Exploring the micromotivations (Williams, 1979, 1988; Aguiar, 1991)or internal reasons (Williams, 1979; Lupia, McCubbins, & Popkin, 2000)that mold public preferences for either democracy or authoritarianism,this paper aims to discuss the types of rationality that liebehind people’s choices in survey studies in Latin America.From this perspective, we examine the balance between surveyrespondents’ evaluation of democratic government and theirviews of the efficacy of democracy to solve their country’sproblems, and their joint impact on the molding of citizens’preferences for a particular type of government. Results showthat satisfaction with how democratic government performs standsout as one of the reasons underlying individuals’ preferencesfor democracy. Also the belief that democracy does not solvethe problems significantly determines the people’s choice.Conclusions favor the hypothesis that, in the region, a utilitarianrationality prevails over an axiological rationality (Weber, 1922;Boudon, 1996) in the way citizens form preferences for or againstdemocracy.  相似文献   

19.
Editorial     
We are at an important transition point in the editorial guidanceand management of the International Journal of Public OpinionResearch, and I want to bring you up to date on those developments.Wolfgang Donsbach has served two tours of duties as ManagingEditor, spanning a period of not <20 years, and during  相似文献   

20.
The capacity of the mass media to cover issues as well as thecapacity of recipients to be concerned about issues is limited.The coverage on and the concern about new issues will consequentlyremove old issues from the agenda of both the mass media andthe general public. The present study investigates two modelsof this process of issue competition. The equal-displacementmodel assumes that a rise of one issue in the media by a givennumber of stories is matched by an equal fall in all other issuessumming up to a similar number of issues. The restructuringmodel assumes that unexpected, surprising or otherwise newsworthyevents create killer issues that move several other issues completelyoff the agenda and leave others untouched. A content analysisof all news shows of the two major TV stations in Germany and53 weekly surveys regarding 16 different issues covering thewhole year 1986 shows that—within the media agenda—therewere no killer issues affecting the coverage of the TV stations.In the public agenda, however, some killer issues could be identified.Coverage of these issues increased public concern about themand decreased concerns about other issues. Consequences of theseresults for agenda-setting theory and for politics are discussed.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号