... facilitatethe development of better user interfaces and im-prove the performance of summarization and in-formation retrieval systems. Discourse segmentationof the documents com-posed of parallel parts ... to the problem of segmenting parallelparts of documents. The task of aligning each sentence of an abstractto one or more sentences of the body has beenstudied in the context of summarization ... Proceedings of the ACL 2010 Conference Short Papers, pages 151–155,Uppsala, Sweden, 11-16 July 2010.c2010 Association for Computational Linguistics Unsupervised Discourse Segmentation of Documents...
... prosodic anal-ysis ofdiscourse segments in direction-giving mono-logues. In Proc. of the ACL.J. Hirschberg and C. Nakatani. 1998. Acoustic indicators of topic segmentation. In Proc. of ICSLP.A. ... error of Pk= 15.79, while the average performance of the algorithm is Pk= 15.31 on the WSJ test corpus (unknown number of segments).mean and the variance of the hypothesized probabil-ities of ... approaches tothe segmentation problem applied to spoken andwritten documents. In Section 3, we describe thecorpus of recorded meetings intended to be seg-mented, and the annotation of its discourse...
... so our segmentation approach shouldwork well for Discourse GraphBank style parsing.The Penn Discourse Treebank (Miltsakaki et al.,2004), or PDTB, uses a relatively flat annotation of discourse ... Proceedings of the 43rd Annual Meeting of ACL, pages 173–180.E. Charniak. 2000. A maximum-entropy-inspired parser. InProceedings of the 1st Conference of the North AmericanChapter of the Association ... the publicly available SPADE discourse parser1– make use of the output of the Charniak(2000) parser to derive syntactic indicator featuresfor segmentation and discourse parsing.Sporleder...
... Technology{igorm,malex,regina,glass}@csail.mit.eduAbstractWe address the task ofunsupervised topic segmentation of speech data operating overraw acoustic information. In contrast to ex-isting algorithms for topic segmentation of speech, our approach ... the impact of various design deci-sions on the performance of our algorithm.Comparison with Transcript-Based Segmenta-tion Table 2 shows the segmentation accuracy of the audio-based segmentation ... speaker-independent (SI) transcripts. Forall of the algorithms, the target number of segmentsis set to the reference number of segments.boundaries along the span of the lecture and select-ing random...
... [%]ProbabilisticRecursive MDLLinguisticaNo segmentation Figure 2: Expectation of the percentage of recog-nized morphemes for English data.a baseline of no segmentation fairly high. The no- segmentation baseline ... obtain a segmentation of the words in the corpus, since wecan rewrite every word as a sequence of morphs.2.1 Size of the morph lexiconWe start the generation process by deciding the num-ber of ... segmenta-tion of a set of words consists of the following steps:(1) Segment the words in the corpus using the au-tomatic segmentation algorithm.(2) Divide the segmented data into two parts of equal...
... contexts. Perspec-tives of this study concern the application of these evaluationcriteria for the choice of the segmentation method parame-ters or the definition of new segmentation methods by ... a segmentation result. Such anevaluation criterion can be useful for different applications: the comparison ofsegmentation results, the automatic choice of thebest fitted parameters of a segmentation ... given image, or the definition of new segmentation methods by optimization.We first present the state of art ofunsupervised evaluation, and then, we compare six unsupervised evaluation criteria....
... on Critical Discourse Analysis …………………………………………81.2.1.What is Critical Discourse Analysis …………………………………………… 91.2.2.Key notions of CDA …………………………………………………………… 91.2.3.Methodology of CDA ……………………………………………………………111.2.4.Principles ... ………………………………………………………………………iList of Tables …………………………………………………………………………… iiINTRODUCTION ……………………………………………………………………….11. Rationale ………………………………………………………………………… 12. Scope of the research …………………………………………………………… 23. Aims of ... ………………………………………………………… 46. Design of the research …………………………………………………………… 5CHAPTER 1 – THEORETICAL BACKGROUND ………………………………… 61.1. The history of Critical Linguistics and Critical Discourse Analysis ……………...
... do. The importance of the historical contexts ofdiscourse therefore should be highlighted in the processes of interpretation and explanation of discourses.Ideology is one of the most controversial ... reflection of ideology, thus CDA can help revealing the hidden effects of language embodied in various types of discourse. With the aim of dealing with social problems by means of language, ... way of discourse practice. That is, properties of socio-cultural practice shape texts, but indirectly through shaping the nature ofdiscourse practice, which is realized in features of texts....
... researchWithin the limitation of a minor thesis, the research focuses only on the written text of the news discourse in a relatively short period of time from two sources of news and studies certain ... both the groups of participants. In a broad sense, the analysis obviously proves that discourse or the use of language is a medium for expressing ‘attitudes’ or various forms of social relations.2. ... in two stages: the analysis of headlines and the analysis of full-text news reports. In the first stage, we will make a detailed description of the representation of the US- Japan coalition...
... from two sources of news: the Voice of America and Nhan Dan Analysis of headlines- VOAPositively represent the US-Japan coalition 76.9% of the processesThe role of actor and sayer ... Scope of the research Aims of the research and research questions Methodology Background information Design of the research lexicalizationVOA Negativization of NK’s activitiesPositivization ... over-lexicalization of NK’s missile launchesNhan Dan neutral over-lexicalization of NK’s missile launches Data analysis and discussionAnalysis of headlines Analysis of full-text news...