Báo cáo khoa học: "Summarizing multiple spoken documents: finding evidence from untranscribed audio" potx

... 549–557, Suntec, Singapore, 2-7 August 2009. c 2009 ACL and AFNLP Summarizing multiple spoken documents: finding evidence from untranscribed audio Xiaodan Zhu, Gerald Penn and Frank Rudzicz University ... Canada {xzhu,gpenn,frank}@cs.toronto.edu Abstract This paper presents a model for summa- rizing multiple untranscribed spoken doc- uments. Without assuming the availabi...

Ngày tải lên: 23/03/2014, 16:21

... randomly selected words (17.7%) does not differ substantially from the overall average F-Measure (17.1%) to be sure that we have chosen words from all F-Measure ranges. In particular, we asked different ... Karlsruhe jowenderoth@googlemail.com Abstract This paper presents an approach for the au- tomatic acquisition of qualia structures for nouns from the Web and thus opens the pos- sibi...

Ngày tải lên: 08/03/2014, 02:21

... hand, most of CNN’s story high- lights are not taken from the beginning of the ar- ticles. In fact, more than 50% of the highlights stem from sentences that are not among the first 100 words of ... story highlights, we gathered statistics from 1,200 CNN newswire articles. An additional 300 articles were set aside to serve as a test set later on. The arti- cles were taken from a wide ra...

Ngày tải lên: 24/03/2014, 03:20

... language modeling toolkit. In Proceedings International Con- ference on Spoken Language Processing, pages 257– 286. Jorg Tiedemann. 2009. News from opus - a collection of multilingual parallel corpora with ... theory, it is assumed that the training and test datasets are drawn from the same distribu- tion, or in other words, they are from the same do- main. However, bilingual corpora...

Ngày tải lên: 19/02/2014, 19:20

... some labeled data for multiple other domains, des- ignated as the source domains. We automat- ically create a sentiment sensitive thesaurus using both labeled and unlabeled data from multiple source ... automatically created sentiment sensitive thesaurus. We use la- beled data from multiple source domains and unla- beled data from source and target domains to rep- resent the distr...

Ngày tải lên: 20/02/2014, 04:20

... In addition, the corpus created from re- views is often noisy as we discuss in Section 2. This paper proposes a novel method of building polarity-tagged corpus from HTML documents. The idea behind ... that express opinion (opinion sentences) from HTML documents. Because this method is fully auto- matic and can be applied to arbitrary HTML doc- uments, it does not suffer from the same...

Ngày tải lên: 20/02/2014, 12:20

... discourse. Nevertheless, we believe that the entire spoken discourse is to be represented by one discourse structure. Evidence for this assumption comes from the observation that anaphoric refer- ences ... the producer. In order to obtain a better understand- ing of the planning process we analyse spoken dis- course elicited in an experimental setting. Subjects describe the pi...

Ngày tải lên: 20/02/2014, 18:20

... we collected from NEONATE, to build the micro planner and realizer. 4.1 Content Selection and Segmentation The most important question in summarization is 'what data points from the input ... the domain constraints such as limits on parameter values. It is clear from our own studies on data summari- zation and also from the earlier studies by others (Shahar, 1997; Boyd, 1998; K...

Ngày tải lên: 22/02/2014, 02:20

Ngày tải lên: 08/03/2014, 01:20

... Context-Free Grammars and Multiple Context-Free Grammars", Trans. IEICE, J71-D-I, 5:758-765. Kasami, T. et al. 1988b. "On the Membership Prob- lem for Head Language and Multiple Context-Free ... languages. Among them are parallel multiple context-free grammars (pmcfg's) and lexical-functional gram- mars (lfg's). Pmcfg's and their subclass called multiple...

Ngày tải lên: 08/03/2014, 07:20