Báo cáo khoa học: "Multi-Document Summarization using Sentence-based Topic Models" docx

... how much using ASR output degrades detection of decision regions. 3 The authors used the AMI DA annotations. 309 The authors conducted experiments using the AMI corpus and found that when using ... (Fern ´ andez et al., 2008). The HGM result presented in Table 2 was computed using the three-level DBN model (see Fig. 4b) using the combination of UTT and DA features. Without DA featur...

Ngày tải lên: 23/03/2014, 16:20

... to improve both topic- word distributions (by biasing topics to pro- duce appropriate seed words) and to im- prove document -topic distributions (by bi- asing documents to select topics related ... five topics identified by the LDA are shown in Table 1. A brief observation of the topics reveals that LDA has roughly allo- cated topics 1 & 2 for the most frequent class (“Earn”) and one to...

Ngày tải lên: 31/03/2014, 20:20

... Event-based summarization has been investi- gated in recent research. It was first presented in (Daniel, Radev and Allison, 2003), who treated a news topic in multi-document summarization ... transi- tion sequences, i.e. entity-based summarization, we are particularly interested in relevance be- tween events in event-based summarization. Extractive summarization requires rankin...

Ngày tải lên: 31/03/2014, 01:20

... (Felber, 1984). The recent works on ATR identify the candidate terms using shallow syntactic information and score the terms using statistical measure such as frequency. The candidate terms are ... engineering and chemistry are highly related. 2.3 Scoring Terms Using Dictionary Hierarchy The main idea for scoring terms using the hierarchy is based on the premise that terms i...

Ngày tải lên: 08/03/2014, 05:20

... METEOR scores were computed using 16 references. The TER and TERp were com- puted using a single reference. 4.3 Reordering model comparison Tables 1 and 2 show the scores using the differ- ent reordering ... weighted alignment matrices. First, we test a simple approach by using the 1-best alignment to generate the reordering model, while using the alignment matrix to produce the tra...

Ngày tải lên: 23/03/2014, 16:20

... Coreference Resolution Using Competition Learning Approach Xiaofeng Yang *+ Guodong Zhou * Jian Su * Chew Lim Tan ... generated for each anaphor encountered in the training data set, a classifier can be trained using a certain machine learning algorithm, such as C4.5, RIPPER, etc. Given the feature vector ... classification re- sult for an instance )ana,cj,ci(inst . The antecede...

Ngày tải lên: 23/03/2014, 19:20

... (83.3%) Table 1. Comparison of segmentation algorithm using different linguistic features. Discussion: The segmentation algorithm using the linguistic features word repetition and collocation ... Nottingham NG1 4BU, UK lje@doc.ntu.ac.uk Abstract A method is presented for segmenting text into subtopic areas. The proportion of related pairwise words is calculated between adjacen...

Ngày tải lên: 31/03/2014, 04:20

... time. 1 Introduction Topic models, exemplified by latent Dirichlet alloca- tion (LDA) (Blei et al., 2003), discover latent themes present in text collections. “Topics” discovered by topic models are ... n k|d )  (i→j)∈λ β i→j + n i→j|k  j (β i→j  + n i→j  |k ) where n k|d is topic k ’s count in the document d ; α k is topic k ’s prior; Z − and L − are topic and path assignments...

Ngày tải lên: 23/03/2014, 14:20

... independent of e. Most SMT systems approximate the summa- tion over all possible derivations by using 1-best derivation for efficiency. They search for the 1- best derivation and take its target ... structures (e.g., phrase pairs) usually generate target sentences from left to right while those using rules with hier- archical structures (e.g., SCFG rules) often run in a bottom-up style. In...

Ngày tải lên: 23/03/2014, 16:21

... proteins were collected onto glass fiber filters using a cell harvester, washed with 5% trichloroacetic acid (w/v), and radioactivity was counted using a liquid scintil- lation counter. Background ... was determined by UV spectroscopy, using an extinction coefficient of 10 280 M )1 Æcm )1 [22]. AAP concentrations were also determined spectrophotometri- cally using an extinction coefficien...

Ngày tải lên: 19/02/2014, 12:20