共查询到20条相似文献，搜索用时 46 毫秒
1.
Word embeddings and convolutional neural networks (CNN) have attracted extensive attention in various classification tasks for Twitter, e.g. sentiment classification. However, the effect of the configuration used to generate the word embeddings on the classification performance has not been studied in the existing literature. In this paper, using a Twitter election classification task that aims to detect election-related tweets, we investigate the impact of the background dataset used to train the embedding models, as well as the parameters of the word embedding training process, namely the context window size, the dimensionality and the number of negative samples, on the attained classification performance. By comparing the classification results of word embedding models that have been trained using different background corpora (e.g. Wikipedia articles and Twitter microposts), we show that the background data should align with the Twitter classification dataset both in data type and time period to achieve significantly better performance compared to baselines such as SVM with TF-IDF. Moreover, by evaluating the results of word embedding models trained using various context window sizes and dimensionalities, we find that large context window and dimension sizes are preferable to improve the performance. However, the number of negative samples parameter does not significantly affect the performance of the CNN classifiers. Our experimental results also show that choosing the correct word embedding model for use with CNN leads to statistically significant improvements over various baselines such as random, SVM with TF-IDF and SVM with word embeddings. Finally, for out-of-vocabulary (OOV) words that are not available in the learned word embedding models, we show that a simple OOV strategy to randomly initialise the OOV words without any prior knowledge is sufficient to attain a good classification performance among the current OOV strategies (e.g. a random initialisation using statistics of the pre-trained word embedding models). 相似文献
2.
《Communication Teacher》2013,27(4):220-223
Courses: Hybrid/Introduction to Communication, Public Speaking, Advanced Public Speaking (online and face-to-face) Objectives: Students will understand the importance of repetition, style and focus in speeches. Students will evaluate the effectiveness of their language use by creating and reviewing word clouds 相似文献
3.
4.
《Communication monographs》2012,79(3):143-156
This re‐examination of Richard Brinsley Sheridan's two speeches against Warren Hastings appraises the spectacular success of the impeachment speech of 1787 and the relative success of the rephrased version delivered at Hastings’ trial in 1788. It also assesses factors that may have influenced later observers who reacted to the speeches with considerably less enthusiasm than did Sheridan's listeners. In maintaining that Sheridan's view of Hastings was consistent and justified, the author questions the continuing criticisms of Sheridan's motives in the Hastings affair. The conclusion considers why Sheridan's two speeches have probably failed as literature, and hence indirectly suggests why other speeches may have gained literary stature. 相似文献
5.
William E. Beattie 《Communication Studies》2013,64(3):16-18
Many commentators on the 1956 presidential campaign have asserted that Stevenson's speeches in that year were both less interesting and simpler than those in 1952. The author has applied the Flesch Readability Formulas to selected speeches in the two campaigns and has found that the charges are supported by application of the tests. 相似文献
6.
《Communication monographs》2012,79(1):39-47
This article seeks to isolate factors associated with effective and ineffective campaign speeches made by Adlai E. Stevenson in 1956. To distinguish effective from ineffective speeches, the author solicited the judgments of 105 campaign associates, Democratic and Republican politicians, journalists, and students of public address. Eighty‐seven replies were received and sixty‐four were sufficiently complete for analysis. There was high agreement that six speeches were effective and six ineffective. Factors associated with effectiveness (their opposites with ineffectiveness) included the following: single issue discussed; relatively uncontroversial subject; constructive material emphasized and refutation de‐emphasized; responsive and sympathetic audience; speaker spent great deal of time editing and revising but little on basic draft; revisions were to clarify and identify with audience rather than to qualify remarks; emphasis on solutions rather than problems; introductions contained humor, irony, and satire; presence of direct ethical appeals; emphasis on values rather than needs; frequent alterations in text during delivery. The author also found several factors common to both effective and ineffective speeches. These included, among others: foreign or domestic issue; size of audience and size of auditorium; length of speech of introduction; length of program prior to speech; logical organization; poor use of summaries and transitions. 相似文献
7.
Pamela A. Hayward 《Communication Teacher》2017,31(4):239-244
Courses: Introduction to public speaking, advanced public speaking, hybrid/survey introduction to communication.
Objectives: At the end of this activity, students will be able to (1) explain the elements of a speaking outline and discover these elements in real-world speech examples, (2) recreate outline formats effectively in their personal speeches, (3) appraise both their own presentations and the presentations of others in order to assess the effectiveness of these presentations, (4) recognize the needed elements of different speech genres and apply those elements when developing and delivering their own speeches, and (5) summarize and synthesize course concepts and apply them to a critical analysis of a real-world speech example. 相似文献
8.
Don Waisanen 《Communication Teacher》2018,32(2):82-86
Courses: Public speaking; communication courses requiring speeches.
Objective: Students will learn how to apply humor principles to speeches through a slideshow method supportive of this goal, and to become more discerning about the possibilities and pitfalls of humorous communication. 相似文献
9.
《Communication monographs》2012,79(4):339-347
Josephus’ narrative of Masada, confirmed by archeological findings, included an account of two speeches by Eleazar to the beseiged urging mass suicide rather than surrender. Application of Durkheim's etiological theory to the event suggested strongly that the speaker's arguments and appeals directed to shared social and religious beliefs produced a state of extreme anomie. This condition, combined with altruistic predispositions and pressure of group resolve, accounted in large part for the effectiveness of the speeches. 相似文献
10.
11.
《Communication Teacher》2013,27(4):271-275
Courses: Public Speaking, Advanced Public Speaking, Capstone Courses.Objective: This activity is designed to help students identify different components of an elevator speech and assemble elevator speeches highlighting their strengths. The activity ultimately increases students’ sense of self-competence. 相似文献
12.
Phillip K. Tompkins 《Communication Studies》2013,64(2):90-95
Rhetorical criticism (i.e., textual analysis of speeches) is severely handicapped because speech cannot be adequately represented in writing; even if it could be so represented, it is illogical and presumptuous to study critically oral communication received from an inappropriate medium (printed page) via an inappropriate sensory channel (vision). A hierarchy of research priorities—the criterion for relative worth being the degree to which methodologies extend our knowledge of rhetorical theory—is proposed by the author. 相似文献
13.
融合主题与情感特征的突发事件微博舆情演化分析 总被引：1，自引：0，他引：1
[目的/意义]微博是突发事件网络舆情传播的重要媒介。面向突发事件的微博主题和情感挖掘对掌握突发事件的网络舆情、识别与预测潜在问题及风险等方面具有重要的实践意义。尝试提出一种融合主题与情感特征的突发事件微博舆情演化分析方法。[方法/过程]以寨卡事件为例,通过划分微博舆情演化的生命周期,基于word2vec技术分别提取该事件生命周期各阶段的微博主题,采用基于词典的情感分析方法,引入情感词、表情符号等多情绪源,对不同主题下的评论情感做细粒度划分,并计算情感强度,最终实现微博主题与情感的协同分析。[结果/结论]所提出的微博舆情演化分析方法能够揭示面向特定事件的微博在突发事件生命周期各阶段的主题特征、情感类型与强度,剖析网络舆情主题与情感特征的协同演化规律。 相似文献
14.
《Communication Teacher》2013,27(4):224-227
Courses: Basic Course, Public Speaking, Hybrid Objectives: Students will execute a small group activity designed to improve the quality of their presentations through peer feedback and video “drafting” and to enhance the creative possibilities for their informative speeches 相似文献
15.
16.
ABSTRACTCourses: This activity is designed specifically for public-speaking courses, but it could be used in the general introductory communication course. It also holds potential for use in persuasion, argumentation, or strategic communication courses.Objectives: This activity helps students understand audience as a more complicated concept—one that recognizes the power of a speaker in creating and challenging communities and interrogating power through their performative speeches in the courses named above. 相似文献
17.
《Slavic & East European Information Resources》2013,14(1-2):145-148
ABSTRACT The article describes the multimedia website Tito's Home Page, which is devoted to Marshal Josip Broz Tito of Yugoslavia, his life, work, and associates, as well as others' reactions to him. Photos, songs, recordings of speeches, a filmography, and a bibliography are among the items included. The site is in English. 相似文献
18.
《Communication monographs》2012,79(1):38-43
Factor analyses were made of instructor ratings of 442 freshman speeches. Each speech was rated on 11 scales. It was found that only 3 factors, rather than 11, were actually being discriminated among by the'raters. These were an analysis‐content factor, a delivery factor, and a language factor. It was also demonstrated that the variance not attributable to one of these three factors could be accounted for in terms of rating unreliability. 相似文献
19.
《Communication Teacher》2013,27(3):122-126
Courses: Public Speaking, Hybrid Basic Course Objectives: Students will realize the importance of invention and arrangement when developing speeches, use PowerPoint presentational aids, and manage communication apprehension. 相似文献
20.
[目的/意义]微博已成为大众情感表达的重要平台,微博的情感分析在舆情分析、用户体验、商机挖掘等方面有着重要的作用。[方法/过程]提出的情感倾向分类算法WE_SDAE使用单词嵌入的方式将微博表示成一个低维稠密向量,然后通过添加正则项和加噪处理的方式将基本的自动编码器算法优化成深层噪音自动编码器,并在顶层添加分类器,实现情感倾向分类。考虑到微博用词灵活,还从单字和词语两个粒度训练模型。[结果/结论]实验结果表明,基于单字粒度的模型表现优于基于词语粒度的模型。此外,对比实验显示WE_SDAE算法优于传统的SVM、Naive-Bayes、XgBoost等相关算法;单词嵌入的方式优于传统的向量空间模型表示方法,能在微博情感分析中取得较好的效果。 相似文献