Báo cáo khoa học: "Learning Surface Text Patterns for a Question Answering System" doc

... Learning Surface Text Patterns for a Question Answering System Deepak Ravichandran and Eduard Hovy Information Sciences Institute University of Southern California 4676 Admiralty Way Marina ... few hand-crafted examples of each question type to Altavista. Patterns are then automatically extracted from the returned documents and standardized. We calculate the precision of each pattern, ... acquiring patterns automatically and to measure their accuracy. It has been noted in several QA systems that certain types of answer are expressed using characteristic phrases (Lee et al., 01; Wang...

Tài liệu Báo cáo khoa học: "Learning Sub-Word Units for Open Vocabulary Speech Recognition" doc

... 1AA5AA_N_JH_AA_N4AA_N_JH_AA3AA_N_JH2AA_NN_JH_AA_NN_JH_AAN_JHN6N_JH_AA_N_IYIYNAA_NAAAA_N_IYJH_AA_NJH_AAJHJH_AA_N_IYFigure 3: FSM representing all segmentations for the word ANJANI ... ofdetection task performance. In Eurospeech.Carolina Parada, Abhinav Sethy, and Bhuvana Ramab-hadran. 2009. Query-by-example spoken term detec-tion for oov terms. In ASRU.Carolina Parada, Mark Dredze, ... 2011.c2011 Association for Computational LinguisticsLearning Sub-Word Units for Open Vocabulary Speech RecognitionCarolina Parada1, Mark Dredze1, Abhinav Sethy2, and Ariya Rastrow11Human Language...

Báo cáo khoa học: "Mining Association Language Patterns for Negative Life Event Classification" doc

... mining, where each sales transaction in a database can be considered as a sentence in the corpora, and each item in a transaction denotes a word in a sentence. An association language pat-tern ... Traditional approaches to sentence classifica-tion (Khoo et al., 2006; Naughton et al., 2008) or text categorization (Sebastiani 2002) usually adopt bag-of-words as baseline features to train classifiers. ... friend and was upset. Table 1. Classification of negative life events. 2022.2 Generate association patterns from fre-quent word sets Association language patterns can be generated via a confidence...

Báo cáo khoa học: "Espresso: Leveraging Generic Patterns for Automatically Harvesting Semantic Relations" docx

... Generic Patterns for Automatically Harvesting Semantic Relations Patrick Pantel Information Sciences Institute University of Southern California 4676 Admiralty Way Marina del Rey, CA 90292 ... Web and syntactic ex-pansions to compensate for lacks of redun-dancy in small corpora; • Generality: Espresso is amenable to a wide variety of binary relations, from classical is -a and part-of ... quality manually using two human judges (a total of 680 instances were annotated per judge). For each instance, judges may assign a score of 1 for correct, 0 for incorrect, and ½ for partially...

Báo cáo khoa học: "Dealing with distinguishing descriptions in a guided composition system" doc

... contrary, 'l'oiseau' ('the bird') is correct and can lead to at least three different a~ 'l'oiseau qui chante' (the bird that sings') which designate ... and only one entity among others in a context set. Many natural language interfaces need to deal with this sort of NPs. We develop an application where these NPs occur in a particular way: ... goal like in [Appelt 85]. The relations between Gricean maxims and the generation of definite NP are studied in [Passonneau 95] and [Dale & al. 96] for example. [Horacek 97] gives a...

Tài liệu Báo cáo khoa học: "Learning with Unlabeled Data for Text Categorization Using Bootstrapping and Feature Projection Techniques" doc

... data (Ko and Seo, 2004). How can labeled training data be automatically created from unlabeled data and title words? Maybe unlabeled data don’t have any information for building a text classifier ... labeled data. While labeled data are difficult to obtain, unlabeled data are readily available and plentiful. Therefore, this paper advocates using a bootstrapping framework and a feature projection ... can label all unlabeled documents for their category, we can finally obtain labeled training data (machine-labeled data). When the machine-labeled data is used to learn a text classifier,...

Tài liệu Báo cáo khoa học: "Learning to Find Translations and Transliterations on the Web" doc

... Phrase translation and transliteration is important for cross-language tasks. For example, Knight and Graehl (1998) describe and evaluate a multi-stage machine translation method for back transliterating ... to prepare the training data. 3 Method To find translations for a given term on the Web, a promising approach is automatically learning to extract phrasal translations or transliterations ... exploit mixed code webpages for word and phrase translation. Nagata et al. (2001) present a system for finding English translations for a given Japanese technical term using Japanese-English snippets...

Tài liệu Báo cáo khoa học: "Learning the Latent Semantics of a Concept from its Deﬁnition" pptx

... the last 10 sampling iterations. We also set a threshold to elesk similarity values, which yieldsbetter performance. Same as (Sinha and Mihalcea,2007), values of elesk larger than 240 are set ... senses acrossdictionaries, hence Wik is only used as augmenteddata for WMF to better learn the semantics of words.All data is tokenized, POS tagged (Toutanova et al.,2003) and lemmatized, ... challenge lies in that sense deﬁni-tions are typically too short/sparse for latent vari-able models to learn accurate semantics, since thesemodels are designed for long documents. For exam-ple, topic...

Tài liệu Báo cáo khoa học: "Learning Word-Class Lattices for Deﬁnition and Hypernym Extraction" doc

... but are rather uninformative, like“dynamic programming was the brainchild of anamerican mathematician”, as well as informativesentences that are not deﬁnitional (e.g., they do nothave a hypernym), ... UK.Judith Klavans and Smaranda Muresan. 2001. Eval-uation of the DEFINDER system for fully auto-matic glossary construction. In Proc. of the Amer-ican Medical Informatics Association (AMIA) Sym-posium.Michael ... 2003 Annual National Conference on DigitalGovernment Research, pages 1–6. Digital Govern-ment Society of North America.Adrian Iftene, Diana Trandab˘ a, and Ionut Pistol. 2007.Natural language...

Báo cáo khoa học: "Learning Context-Dependent Mappings from Sentences to Logical Form" doc

... to logical form. The training ex-amples are sequences of sentences anno-tated with lambda-calculus meaning rep-resentations. We develop an algorithm thatmaintains explicit, lambda-calculus ... logical form for Exam-ple 1 (a) , which is directly available in C, producesthe desired ﬁnal meaning.Elaborations Later statements can expand themeaning of previous ones in ways that are difﬁ-cult ... meaning representations.Researchers have developed approaches usingmodels and algorithms from statistical machinetranslation (Papineni et al., 1997; Ramaswamyand Kleindienst, 2000; Wong and...

Xem thêm