Báo cáo khoa học: "Template-Based Information Extraction without the Templates" pot

... Entity Extraction Once documents are labeled with templates, we nextextract entities into the template slots. Extraction oc-curs in the trigger sentences from the previous sec-tion. The extraction ... versa. We thus take the max-imum of either cosine score as our ﬁnal similaritymetric between two relations. We then back off to the average of the two cosine scores if the max is notconﬁdent ... represent templates, we induce their slots, notknowing a priori how many there are, and then ﬁllthem in by extracting entities as in the standard task.In our example above, the three bold verbs (use,...

Tài liệu Báo cáo khoa học: "Resume Information Extraction with Cascaded Hybrid Model" pdf

... instance, the error in the first pass may be transferred to the second pass when determining the extraction range of detailed information. Therefore the precision and recall of detailed information ... type of them appears in the text, then the weight of this feature is 1, otherwise is 0. 3.3 Block Selection Block selection is used to select the blocks generated from the first pass as the ... HMM for the educational detailed information extraction for the same reason. In addition, classification based method is selected for the personal detailed information extraction where information...

Tài liệu Báo cáo khoa học: "Integrating Information Extraction and Automatic Hyperlinking" docx

... This way, the user can, while reading a text, immediately link up textual information to the Internet or to any other docu-ment base without accessing a search engine. The quality of the link ... Further-more, coreferences serve as a means of information transport into the output description on the RHS of the rule. Finally, the choice of feature structures as primary citizens of the ... and gender for the matched items. Eventually, the description on the RHS creates a feature structure of type phrase, where the category is coreferent with the category Noun of the right-most...

Báo cáo khoa học: "Open Information Extraction using Wikipedia" pdf

... Patterns of the <type>”: The matcher ﬁrstidentiﬁes the type of the entity (e.g., “city” for“Ithaca”), then instantiates the pattern to create the string the city.” Since the ﬁrst sentence ... sentence if the subject and/or attribute value are not heads of the noun phrases containing them. Third, it discards the sentence if the subject and the attribute valuedo not appear in the same ... sentence, the matcher picks the onewhich is closest to the matched infobox attributevalue in the parser dependency graph.Matching Sentences: The matcher seeks a uniquesentence to match the attribute...

Báo cáo khoa học: "Sparse Information Extraction: Unsupervised Language Models to the Rescue" pptx

... becorrect based on the conjunction of the KnowItAlland the distributional hypotheses. The contributions of the paper are as follows:• The paper introduces the insight that the sub-ﬁeld of language ... methodology, and then present our re-sults. The ﬁrst experiment tests the hypothesis thatHMM-T outperforms an n-gram-based method on the task of type checking. The second experimenttests the hypothesis ... whether the extracted arguments share the appropriate relation.Again, this information is not given to the system,so REALM compares the context distributions of the extractions to those of the...

Báo cáo khoa học: "Learning Information Structure in The Prague Treebank" docx

... MOD,MANN, ATT, OTHER. The derived features are computed using the de-pendency information from the tectogrammaticallevel of the treebank and the surface order of the words corresponding to the nodes5. ... itemsbelong to the Topic and f items to the Focus.Before the manual annotation, the corpus hasbeen preprocessed to mark all nodes with the TFAattribute of f, as it is the more common value. Then the ... divides the linguis-tic meaning of the sentence into parts that link the content to the context, and others that advance the discourse, i.e. add or modify information; and (ii)We use the Praguian...

Báo cáo khoa học: "Extracting Hypernym Pairs from the Web" potx

... in em-ploying the web for the extraction of hypernym re-lations. We are especially curious about whether the size of the web allows to achieve meaningful resultswith basic extraction techniques.In ... make the same assumption as Snow et al. (2005): the hy-pernymy relations in the WordNets are complete for the terms that they contain. This means that if twowords are present in the lexicon without ... as their hypernyms,outperforms the corpus approaches.ods, both with respect to precision and recall.3.2 Extraction from the webFor our web extraction work, we used the same in-dividual extraction...

Báo cáo khoa học: "Synonymous Collocation Extraction Using Translation Information" ppt

... 1998; Gasperin et al., 2001). The methods used the contexts around the investigated words to discover synonyms. The problem of the methods is that the precision of the extracted synonymous words ... are the highest. In order to compare our methods with other methods under the same recall value, we conduct another experiment on the type <verb, OBJ, noun>4. We set the recalls of the ... than the other two methods even when their recalls are the same. It proves that our method of using translation information to select the candidates is effective for synonymous collocation extraction. ...

Báo cáo khoa học: "Unsupervised Relation Extraction by Mining Wikipedia Texts Using Information from the Web" pdf

... leveraging the vast size of the Web. Our hypothesis is that there exist some keyterms and patterns that provide clues to the rela-tions between pairs. From the snippets retrievedby the search ... information contributegreatly to the coverage of relation extraction. • The combination of these patterns producesa clustering method to achieve high pre-cision for different Information Extraction applications, ... sequence of words. The distanceof two surface patterns is deﬁned as the fraction of the LD value to the length of the longer sequence.For estimating the number of clusters k, we ap-ply the stability-based...

Tài liệu Báo cáo khoa học: "Semantic Information and Derivation Rules for Robust Dialogue Act Detection in a Spoken Dialogue System" pptx

... λAis the weight for the ASR score and the lexical score, λLis the weight of the history score,and λA+ λL= 1. Table 4 shows the results thathistory information will effect on the DA ... pivotal in the computa-tion of the lexical score. Finally, the lexical, the his-tory, and the ASR scores are combined to decide the 603optimal dialogue act, and a proper action by the sys-tem ... represents the re-sults of the experiment using the ASR transcriptsdirectly. The fourth column (REF) uses the refer-ence transcripts, so it represents the case with per-fect ASR. The ﬁrst (40%-sim)...

Xem thêm