Search

Starting with the observation that focus phrases can be dichotomized on the basis of whether they are exhaustive or not, this paper shows that exhaustive focus triggers intervention effects for wh-in-situ argument questions, while non-exhaustive focus does not. Empirical data from typologically unrelated languages are cited to support this generalization; exceptional cases are argued to arise on independent grounds. No existing analysis of intervention effects takes into consideration the distinguished behaviors of exhaustive and non-exhaustive focus in terms of triggering intervention effects for wh-in-situ argument questions. I argue that with some modifications, the semantic approach which draws on the interpretation similarity between focus phrases and wh-phrases constitutes the most promising analysis. Intervention effects arise when the focus operator associated with exhaustive focus phrases unselectively evaluates both the focus semantic value and exhaustivity of the in-situ argument wh-phrase in its scope. [ABSTRACT FROM AUTHOR]

A novel feature extraction technique is presented in this paper for an offline handwritten Gurmukhi character recognition system. Handwritten character recognition is a complex task because of various writing styles of different individuals. To select a set of features is an important step for implementing a handwriting recognition system. In this work, we have extracted various topological features, namely, peak-extent features, shadow features and centroid features. A new feature set is also proposed by using horizontal peak extent features and the vertical peak extent features. For classification, we have used k-NN and Linear-SVM classifiers. In view of learning and simplification capabilities of multi layer perceptrons (MLPs), MLPs based pattern classifier is also used for classification. In the present work, we have taken 7,000 samples of offline handwritten Gurmukhi characters for training and testing. Proposed system achieves a maximum recognition accuracy of 95.62% using SVM with linear kernel classifier. By using k-NN and MLPs, a maximum recognition accuracy of 95.48% and 94.74%, respectively, has been achieved with five-fold cross validation. [ABSTRACT FROM AUTHOR]

Highlights: [⿢] Provides a detailed discussion of gaps and resumptive pronouns in Welsh unbounded dependencies. [⿢] Highlights problems for transformational approaches. [⿢] Develops an HPSG analysis in which resumptive pronouns are ordinary pronouns. [Copyright &y& Elsevier]

Built on the basis of the methods developed for Princeton WordNet and EuroWordNet, Arabic WordNet (AWN) has been an interesting project which combines WordNet structure compliance with Arabic particularities. In this paper, some AWN shortcomings related to coverage and usability are addressed. The use of AWN in question/answering (Q/A) helped us to deeply evaluate the resource from an experience-based perspective. Accordingly, an enrichment of AWN was built by semi-automatically extending its content. Indeed, existing approaches and/or resources developed for other languages were adapted and used for AWN. The experiments conducted in Arabic Q/A have shown an improvement of both AWN coverage as well as usability. Concerning coverage, a great amount of named entities extracted from YAGO were connected with corresponding AWN synsets. Also, a significant number of new verbs and nouns (including Broken Plural forms) were added. In terms of usability, thanks to the use of AWN, the performance for the AWN-based Q/A application registered an overall improvement with respect to the following three measures: accuracy (+9.27 % improvement), mean reciprocal rank (+3.6 improvement) and number of answered questions (+12.79 % improvement). [ABSTRACT FROM AUTHOR]

This paper describes the organization and results of the automatic keyphrase extraction task held at the Workshop on Semantic Evaluation 2010 (SemEval-2010). The keyphrase extraction task was specifically geared towards scientific articles. Systems were automatically evaluated by matching their extracted keyphrases against those assigned by the authors as well as the readers to the same documents. We outline the task, present the overall ranking of the submitted systems, and discuss the improvements to the state-of-the-art in keyphrase extraction. [ABSTRACT FROM AUTHOR]

Copyright of Studies in Polish Linguistics is the property of Jagiellonian University Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

We propose a facial trait code (FTC) to encode human facial images, and apply it to face recognition. Extracted from an exhaustive set of local patches cropped from a large stack of faces, the facial traits and the associated trait patterns can accurately capture the appearance of a given face. The extraction has two phases. The first phase is composed of clustering and boosting upon a training set of faces with neutral expression, even illumination, and frontal pose. The second phase focuses on the extraction of the facial trait patterns from the set of faces with variations in expression, illumination, and poses. To apply the FTC to face recognition, two types of codewords, hard and probabilistic, with different metrics for characterizing the facial trait patterns are proposed. The hard codeword offers a concise representation of a face, while the probabilistic codeword enables matching with better accuracy. Our experiments compare the proposed FTC to other algorithms on several public datasets, all showing promising results. [ABSTRACT FROM PUBLISHER]

This paper examines the peculiar aspects of the British English (BE) do, in comparison to those of regular VP ellipsis. In particular, the BE do allows A-extraction out of a verbal domain, but it does not allow A'-extraction out of it, though regular VP ellipsis does not have this kind of asymmetry, permitting both types of extraction. This paper argues that the BE do involves elision of vP, unlike regular VP ellipsis that involves elision of VP. The restriction on A'-movement out of the elided vP, it is shown, follows from the identity/parallelism condition on ellipsis and the chain uniformity condition. [ABSTRACT FROM AUTHOR]

The article focuses on the representation of embedded documents that emphasis on what aspects of the units of retrieval can be used as a basis for classification. It states about feature engineering that involves two tasks including construction of useful features and transformation of the resulting feature set to accommodate the capabilities and limitations of the classification technique that will be employed.

Background: One challenge in reusing clinical data stored in electronic medical records is that these data are heterogenous. Clinical Natural Language Processing (NLP) plays an important role in transforming information in clinical text to a standard representation that is comparable and interoperable. Information may be processed and shared when a type system specifies the allowable data structures. Therefore, we aim to define a common type system for clinical NLP that enables interoperability between structured and unstructured data generated in different clinical settings. Results: We describe a common type system for clinical NLP that has an end target of deep semantics based on Clinical Element Models (CEMs), thus interoperating with structured data and accommodating diverse NLP approaches. The type system has been implemented in UIMA (Unstructured Information Management Architecture) and is fully functional in a popular open-source clinical NLP system, cTAKES (clinical Text Analysis and Knowledge Extraction System) versions 2.0 and later. Conclusions: We have created a type system that targets deep semantics, thereby allowing for NLP systems to encapsulate knowledge from text and share it alongside heterogenous clinical data sources. Rather than surface semantics that are typically the end product of NLP algorithms, CEM-based semantics explicitly build in deep clinical semantics as the point of interoperability with more structured data types. [ABSTRACT FROM AUTHOR]

Automatic keyphrase extraction techniques play an important role for many tasks including indexing, categorizing, summarizing, and searching. In this paper, we develop and evaluate an automatic keyphrase extraction system for scientific documents. Compared with previous work, our system concentrates on two important issues: (1) more precise location for potential keyphrases: a new candidate phrase generation method is proposed based on the core word expansion algorithm, which can reduce the size of the candidate set by about 75% without increasing the computational complexity; (2) overlap elimination for the output list: when a phrase and its sub-phrases coexist as candidates, an inverse document frequency feature is introduced for selecting the proper granularity. Additional new features are added for phrase weighting. Experiments based on real-world datasets were carried out to evaluate the proposed system. The results show the efficiency and effectiveness of the refined candidate set and demonstrate that the new features improve the accuracy of the system. The overall performance of our system compares favorably with other state-of-the-art keyphrase extraction systems. [ABSTRACT FROM AUTHOR]

In the syntax literature, it is commonly assumed that a constraint on linguistic competence blocks extraction of wh-expressions (e.g. what or which book) from embedded questions, referred to as wh-islands. Furthermore, it is assumed that there is an argument/adjunct asymmetry in extraction from wh-islands. We report results from two acceptability judgment experiments on long and short wh-movement and wh-extraction from wh-islands in Danish. The results revealed four main findings: (1) No adjunct/argument asymmetry in extraction from wh-islands. (2) Long adjunct wh-movement is less acceptable than long argument wh-movement, and this difference is attributable to matrix verb compatibility and factivity, not D-linking. (3) Long movement reduces acceptability, but is more acceptable than island violations. (4) Training effects reveal that island violations, though degraded, are grammatical in Danish. Since the standard assumptions cannot account for the range of results, we argue in favor of a processing account referring to locality (processing domains) and working memory. [ABSTRACT FROM AUTHOR]

We report on TExSIS, a flexible bilingual terminology extraction system that uses a sophisticated chunk-based alignment method for the generation of candidate terms, after which the specificity of the candidate terms is determined by combining several statistical filters. Although the set-up of the architecture is largely language-independent, we present terminology extraction results for four different languages and three language pairs. Gold standard data sets were created for French-Italian, French-English and French-Dutch, which allowed us not only to evaluate precision, which is common practice, but also recall. We compared the TExSIS approach, which takes a multilingual perspective from the start, with the more commonly used approach of first identifying term candidates monolingually and then aligning the source and target terms. A comparison of our system with the LUIZ approach described by Vintar (2010) reveals that TExSIS outperforms LUIZ both for monolingual and bilingual terminology extraction. Our results also clearly show that the precision of the alignment is crucial for the success of the terminology extraction. Furthermore, based on the observation that the precision scores for bilingual terminology extraction outperform those of the monolingual systems, we conclude that multilingual evidence helps to determine unithood in less related languages. [ABSTRACT FROM AUTHOR]

Companies must translate their content if they want to operate multinationally. Both quality and speed of translation are key factors in determining market share in the target market. Proactively managing terminology, including pre-translating key terms for a translation project, has beneficial effects on these factors. However, in commercial environments, the volumes of content and subsequently of the required terminology are typically large. Therefore, integrating terminology into the translation pipeline requires a process that is as automated as possible. Term extraction is the cornerstone of this process, but to maximize efficiency it requires a post-processing strategy that repurposes existing lexical resources. Terms extracted from corpora and subsequently translated should be channeled into the company termbase so that they can be leveraged for other purposes. These and other effective practices for processing extracted terms are discussed, based on the author's experiences in one large company. [ABSTRACT FROM AUTHOR]

Huang, Li & Li (2009) claim that (i) there exists in Chinese a type of bei constructions, the so-called unbounded or long-distance passives, in which the pre-bei patient can be indefmitely long from the site from which it is interpreted; (ii) the unbounded or long-distance dependency relation in the bei construction is somewhat likely to exist in the ba construction. By revisiting the structural and semantic properties of the bei construction, I show that Huang, Li & Li's examples are not really attested. I argue that there are two problems with Huang, Li & Li 's claim about unbounded passives: (i) it ignores the fundamental function of passivization; (ii) it mistakenly equates passivization with topicalization. I also point out that Huang, Li & Li 's invention of the ill-formed unbounded passives reveals that interpretive factors may sometimes play a crucial role in the linguistic judgments involving the Chinese language. [ABSTRACT FROM AUTHOR]

A single glance at your crowded desk is enough to locate your favorite cup. But finding an unfamiliar object requires more effort. This superiority in recognition performance for learned objects has at least two possible sources. For familiar objects observers might: 1) select more informative image locations upon which to fixate their eyes, or 2) extract more information from a given eye fixation. To test these possibilities, we had observers localize fragmented objects embedded in dense displays of random contour fragments. Eight participants searched for objects in 600 images while their eye movements were recorded in three daily sessions. Performance improved as subjects trained with the objects: The number of fixations required to find an object decreased by 64% across the 3 sessions. An ideal observer model that included measures of fragment confusability was used to calculate the information available from a single fixation. Comparing human performance to the model suggested that across sessions information extraction at each eye fixation increased markedly, by an amount roughly equal to the extra information that would be extracted following a 100% increase in functional field of view. Selection of fixation locations, on the other hand, did not improve with practice. [ABSTRACT FROM AUTHOR]

The article discusses the traditional window-based and the more recent Part-of-Speech-pattern approach as the most popular approaches to the computational extraction of collocations, methodology and software. It emphasizes the relation of the theoretical status and the notions of collocation into lexicography. It also notes the use of web interface as tools that allows lexicographers to utilize dependency-annotated versions of corpora.

In most statistical machine translation (SMT) systems, bilingual segments are extracted via word alignment. However, there is a need for systematic study as to what alignment characteristics can benefit MT under specific experimental settings such as the type of MT system, the language pair or the type or size of the corpus. In this paper we perform, in each of these experimental settings, a statistical analysis of the data and study the sample correlation coefficients between a number of alignment or phrase table characteristics and variables such as the phrase table size, the number of untranslated words or the BLEU score. We report results for two different SMT systems (a phrase-based and an n-gram-based system) on Chinese-to-English FBIS and BTEC data, and Spanish-to-English European Parliament data. We find that the alignment characteristics which help in translation greatly depend on the MT system and on the corpus size. We give alignment hints to improve BLEU score, depending on the SMT system used and the type of corpus. For example, for phrase-based SMT, dense alignments are required with larger corpora, especially on the target side, while with smaller corpora, more precise, sparser alignments are better, especially on the source side. Avoiding some long-distance crossing links may also improve BLEU score with small corpora. We take these conclusions into account to modify two types of alignment systems, and get 1 to 1.6 % relative improvements in BLEU score on two held-out corpora, although the improved system is different in each corpus. [ABSTRACT FROM AUTHOR]

In this paper, the system bgBell/OLAP for warehousing and online analytical processing data about unique Bulgarian bells is proposed. The implemented system provides possibility for retrieving summarized reports and analyzing different characteristics of the bells with the purpose of extracting the previously unknown and potentially useful information. [ABSTRACT FROM AUTHOR]

Abstract: In this work, a first approach to a robust phoneme recognition task by means of a biologically inspired feature extraction method is presented. The proposed technique provides an approximation to the speech signal representation at the auditory cortical level. It is based on an optimal dictionary of atoms, estimated from auditory spectrograms, and the Matching Pursuit algorithm to approximate the cortical activations. This provides a sparse coding with intrinsic noise robustness, which can be therefore exploited when using the system in adverse environments. The recognition task consisted in the classification of a set of 5 easily confused English phonemes, in both clean and noisy conditions. Multilayer perceptrons were trained as classifiers and the performance was compared to other classic and robust parameterizations: the auditory spectrogram, a probabilistic optimum filtering on Mel frequency cepstral coefficients and the perceptual linear prediction coefficients. Results showed a significant improvement in the recognition rate of clean and noisy phonemes by the cortical representation over these other parameterizations. [Copyright &y& Elsevier]

This article argues for an ellipsis approach to ATB-movement on the basis of morphosyntactic mismatches and reconstruction asymmetries between conjuncts. I will argue that ATB-moved constituents are present in each conjunct; while those in the first conjunct undergo asymmetric extraction, those in the second conjunct are elided under identity with those in the first. Ellipsis in ATB-movement is licensed by means of an [E]atb-feature whose selectional restrictions correctly single out the elements that can undergo ATB-movement. Ellipsis applies derivationally and makes the elided constituents inaccessible for further syntactic operations. Asymmetric extraction is sanctioned by a representational definition of the Coordinate Structure Constraint: At LF, the asymmetrically extracted operator can bind both its own trace as well as the trace left behind by the operator in the second conjunct. Binding into the second conjunct is only possible if both operators bear the same index; this in turn is guaranteed by the recoverability condition on ellipsis, which requires an identical antecedent. Since the operator can bind both variables, the single-identity reading characteristic of ATB is correctiy derived. Finally, the reconstruction asymmetries follow from mismatches between pronouns/R-expressions and their counterparts in the ellipsis site which ellipsis is famous for, so-called vehicle change effects. [ABSTRACT FROM AUTHOR]

In recent years, genre studies have attracted the attention of many researchers. The aim of the present study was to observe the differences in generic structure of doctoral dissertation acknowledgements texts written by English native and non-native (Chinese) PhD students. To this end, thirty native English students' acknowledgement texts and the same number of non-native (Chinese) dissertations acknowledgement texts were selected. The model that was used in this study was Hyland (2003) model. After analyzing each text, the main moves and their steps were extracted and the frequencies of each one were calculated and compared. The results showed that there were some differences but these differences were not statistically significant. In addition, a new step called ''thanking God'' step was discovered in both texts throughout this study. [ABSTRACT FROM AUTHOR]

This article explores a statistical, language-independent methodology for the construction of taxonomies of specialized domains from noisy corpora. In contrast to proposals that exploit linguistic information by searching for lexicosyntactic patterns that tend to express the hypernymy relation, our methodology relies entirely upon the distributional semantics of terms as captured by their lexical co-occurrence in large scale corpora. In a first stage, we analyze the syntagmatic relations of terms that serve as seeds of the taxonomy to be constructed and we obtain, thus, the first batch of hypernym candidate terms for our seed terms. In a second stage, we analyze the paradigmatic relations of the terms by inspecting which terms show a prominent frequency of co-occurrence with the terms that, as we found in the previous stage, are syntagmatically related to our seed terms which allows us to refine the first batch of hypernym candidate terms and obtain new ones. In a third and final stage, we build a taxonomy from the obtained hypernym candidate lists, exploiting the asymmetric statistic association between terms that is characteristic of the hypernymy relation. [ABSTRACT FROM AUTHOR]

This paper evaluates the possibilities of recognizing term candidates through their formal and semantic characteristics. From a cognitive-linguistic stance, the semantic motivation of the word-formation patterns of collocations and compounds in domain-specific texts is assumed to promote their termhood. The semantic motivation to integrate into a multi-word unit is assumed to originate in the generally agreed on generic reference of the modifying constituent. This hypothesis is investigated empirically in a corpus-linguistic experiment. Term candidates instantiating different derivational patterns of adjective-noun constructions have been manually sorted by experts into approved and non-approved terms. Our subsequent linguistic categorization into the morpho-semantic constructions instantiated by the term candidates could be verified quantitatively in terms of relative frequencies. These frequencies clearly divide the term candidates into instances of term-promoting and term-impeding constructions. Obviously, the experts have approved the terms of those constructions which come up with a high relative frequency in the corpus. [ABSTRACT FROM AUTHOR]

In text summarization, relevance and coverage are two main criteria that decide the quality of a summary. In this paper, we propose a new multi-document summarization approach SumCR via sentence extraction. A novel feature called Exemplar is introduced to help to simultaneously deal with these two concerns during sentence ranking. Unlike conventional ways where the relevance value of each sentence is calculated based on the whole collection of sentences, the Exemplar value of each sentence in SumCR is obtained within a subset of similar sentences. A fuzzy medoid-based clustering approach is used to produce sentence clusters or subsets where each of them corresponds to a subtopic of the related topic. Such kind of subtopic-based feature captures the relevance of each sentence within different subtopics and thus enhances the chance of SumCR to produce a summary with a wider coverage and less redundancy. Another feature we incorporate in SumCR is Position, i.e., the position of each sentence appeared in the corresponding document. The final score of each sentence is a combination of the subtopic-level feature Exemplar and the document-level feature Position. Experimental studies on DUC benchmark data show the good performance of SumCR and its potential in summarization tasks. [ABSTRACT FROM AUTHOR]

It is well-known that wh-pronouns may pied-pipe their containing host phrases as they move to their final scope positions. In Finnish, such pied-piping requires further that a wh-element is situated at the left edge of host phrases, a position in which it ends up either through base generation or through wh-movement. This article investigates which independent properties define such pied-piping domains. An empirical generalization will be defended according to which a phrase constitutes such pied-piping domain if and only if it is adjoinable. The hypothesis that pied-piping domains are islands is put into question. Secondary wh-movement, pied-piping and adjunction are thus intrinsically linked with each other. [ABSTRACT FROM AUTHOR]

Defining extraction domains has been a central issue for questions of locality in syntax for decades. The topic thus exhibits a long-lasting research tradition in various paradigms and models. Syntactic approaches, however, get challenged when restrictions on well-formed extraction constructions are encountered which do not allow an explanation based on structural differences straightforwardly. A case in point are that-complement clauses for which it is known that their transparency for extraction gets influenced by the verb which embeds them: So called bridge verbs allow extraction from their complement, non-bridge verbs do not. Because of the fact that an analysis of good and bad extraction verbs shows that the property which makes out a verb's 'bridgedness' can best be accounted for in semantic terms, a non-structural account seems to be called for. Even more evidence for non-structural influence comes from the observation that different subtypes of erothetic illocutionary force have an impact on the acceptability of the extraction construction. Based on these observations and the intention to account for them, this paper presents an analysis of opaque and transparent that-complements from a discourse semantic point of view focusing on the context update induced by the extraction constructions at issue. [ABSTRACT FROM AUTHOR]

This paper examines English unbounded dependency constructions and concludes that the syntactic mechanism responsible for extraction operates in the same way in all constructions, regardless of the structures being subordinate or coordinate. In other words, this work argues that there is a general and fully uniform mechanism behind the linkage of 'gaps' and their 'fillers' in the family of constructions that allow such unbounded dependencies. Crucially, this mechanism is devoid of exceptions, and interacts with other aspects outside the grammar of extraction to predict the Coordinate Structure Constraint, ATB extractions, parasitism, and at least certain islands. Although the underlying syntactic mechanism for extraction is argued to work in the same way for all constructions, certain parts of the grammar can locally impose additional conditions on the possible extraction patterns. Along with recent psycholinguistic and neurolinguistic research, it is argued that many apparent exceptions result from independently motivated performance factors, also observed in garden-path sentences and center-embedded constructions. [ABSTRACT FROM AUTHOR]

Variation and change in relativization strategies has been well documented (e.g. Ball 1996: 46, Biber and Clark 2002, Biber, Johansson, Leech, Conrad and Finegan 1999, Johansson 2006, Lehmann 2002). Certain types of relative clause, namely that-relatives and zero relatives, were difficult to retrieve from plain-text corpora. Studies therefore either relied on manual extraction of data or a subset of possible relativization strategies. In some text types, however, the zero relative is an important member of the class of possible relativizers. Recent advances in syntactic annotation should have made that-relatives and zero relatives more accessible to automatic retrieval. In this article, we test precision and recall of searches on a modest-sized corpus, i.e. scientific texts from ARCHER (A Representative Corpus of Historical English Registers), as a preliminary to future work on the large corpora which are increasingly becoming available. The parser retrieved some false positives and at the same time missed some relevant data. We discuss structural reasons for both kinds of shortcoming as well as the possibilities and limitations of parser adaptation. [ABSTRACT FROM AUTHOR]

A vast amount of usable electronic data is in the form of unstructured text. The relation extraction task aims to identify useful information in text (e.g. PersonW works for OrganisationX, GeneY encodes ProteinZ) and recode it in a format such as a relational database or RDF triplestore that can be more effectively used for querying and automated reasoning. A number of resources have been developed for training and evaluating automatic systems for relation extraction in different domains. However, comparative evaluation is impeded by the fact that these corpora use different markup formats and notions of what constitutes a relation. We describe the preparation of corpora for comparative evaluation of relation extraction across domains based on the publicly available ACE 2004, ACE 2005 and BioInfer data sets. We present a common document type using token standoff and including detailed linguistic markup, while maintaining all information in the original annotation. The subsequent reannotation process normalises the two data sets so that they comply with a notion of relation that is intuitive, simple and informed by the semantic web. For the ACE data, we describe an automatic process that automatically converts many relations involving nested, nominal entity mentions to relations involving non-nested, named or pronominal entity mentions. For example, the first entity is mapped from ‘one’ to ‘Amidu Berry’ in the membership relation described in ‘Amidu Berry, one half of PBS’. Moreover, we describe a comparably reannotated version of the BioInfer corpus that flattens nested relations, maps part-whole to part-part relations and maps n-ary to binary relations. Finally, we summarise experiments that compare approaches to generic relation extraction, a knowledge discovery task that uses minimally supervised techniques to achieve maximally portable extractors. These experiments illustrate the utility of the corpora.1 [ABSTRACT FROM PUBLISHER]

In this paper potential active contours are presented as a new method of image segmentation. The concept of potential contour is a result of the relationship between active contour techniques and the methods of classifiers' construction. The proposed method can be extended by the adaptation mechanism that allows changing the available class of the shapes dynamically. An original contribution is also the method of evaluation of segmentation results and methodology used for the parameters selection. The described method is illustrated by two examples. [ABSTRACT FROM AUTHOR]

We describe a system for extracting complex events among genes and proteins from biomedical literature, developed in context of the BioNLP'09 Shared Task on Event Extraction. For each event, the system extracts its text trigger, class, and arguments. In contrast to the approaches prevailing prior to the shared task, events can be arguments of other events, resulting in a nested structure that better captures the underlying biological statements. We divide the task into independent steps which we approach as machine learning problems. We define a wide array of features and in particular make extensive use of dependency parse graphs. A rule-based postprocessing step is used to refine the output in accordance with the restrictions of the extraction task. In the shared task evaluation, the system achieved an F-score of 51.95% on the primary task, the best performance among the participants. Currently, with modifications and improvements described in this article, the system achieves 52.86% F-score on Task 1, the primary task, improving on its original performance. In addition, we extend the system also to Tasks 2 and 3, gaining F-scores of 51.28% and 50.18%, respectively. The system thus addresses the BioNLP'09 Shared Task in its entirety and achieves the best performance on all three subtasks. [ABSTRACT FROM AUTHOR]

This paper presents the preparation, results and analysis of the BioNLP'09 shared task on event extraction, organized to address the automatic extraction of fine-grained information from the scientific literature on molecular biology. A representation of this information was defined taking into account both the biological and computational requirements of the task, and corpus resources manually annotated by domain experts provided to task participants. To create a basis for further progress, emphasis was placed on providing fine-grained evaluation that isolates different subtasks and allows the analysis of different aspects of the results through various evaluation criteria. In introducing this new task to the community, we made an effort to reduce the cost of participation by making common natural language processing tools, data, and evaluation methods easily accessible. The task received community-wide participation, establishing the state-of-the-art performance at fine-grained event extraction as well as allowing the identification of remaining challenges and suggesting directions for future improvements. All the resources and results of the shared task are publicly available and an online evaluation on blind test data accessible at . [ABSTRACT FROM AUTHOR]

This paper describes our bio-event extraction system developed for the BioNLP 2009 Shared Task 2, with focus on its capability of extracting secondary biological event arguments from literature. Shared Task 2 is particularly interesting because when browsing literature, biologists often need to understand conditions surrounding biological events, which are usually expressed by secondary event arguments (e.g., binding sites). To achieve our goal, we take an approach that extracts n-ary relations from text using event extraction constraints automatically generated from a training corpus. Event constraints consist of sequences of trigger words and semantic roles which we automatically identify using Conditional Random Fields (CRFs). Unlike most other systems participating in this shared task, our system is light-weight and relies on neither external resources (e.g., Ontologies and dictionaries) nor natural language processing software (e.g., POS taggers and parsers). The official test results show that our approach performed well on extracting secondary arguments in Task 2, yielding the highest precision at 76.62% and the second highest F-measure at 43.22%. [ABSTRACT FROM AUTHOR]

The scientific literature is the main source for comprehensive, up-to-date biological knowledge. Automatic extraction of this knowledge facilitates core biological tasks, such as database curation and knowledge discovery. We present here a linguistically inspired, rule-based and syntax-driven methodology for biological event extraction. We rely on a dictionary of trigger words to detect and characterize event expressions and syntactic dependency based heuristics to extract their event arguments. We refine and extend our prior work to recognize speculated and negated events. We show that heuristics based on syntactic dependencies, used to identify event arguments, extend naturally to also identify speculation and negation scope. In the BioNLP'09 Shared Task on Event Extraction, our system placed third in the Core Event Extraction Task (F-score of 0.4462), and first in the Speculation and Negation Task (F-score of 0.4252). Of particular interest is the extraction of complex regulatory events, where it scored second place. Our system significantly outperformed other participating systems in detecting speculation and negation. These results demonstrate the utility of a syntax-driven approach. In this article, we also report on our more recent work on supervised learning of event trigger expressions and discuss event annotation issues, based on our corpus analysis. [ABSTRACT FROM AUTHOR]

We approached the problems of event detection, argument identification, and negation and speculation detection in the BioNLP'09 information extraction challenge through concept recognition and analysis. Our methodology involved using the OpenDMAP semantic parser with manually written rules. The original OpenDMAP system was updated for this challenge with a broad ontology defined for the events of interest, new linguistic patterns for those events, and specialized coordination handling. We achieved state-of-the-art precision for two of the three tasks, scoring the highest of 24 teams at precision of 71.81 on Task 1 and the highest of 6 teams at precision of 70.97 on Task 2. We provide a detailed analysis of the training data and show that a number of trigger words were ambiguous as to event type, even when their arguments are constrained by semantic class. The data is also shown to have a number of missing annotations. Analysis of a sampling of the comparatively small number of false positives returned by our system shows that major causes of this type of error were failing to recognize second themes in two-theme events, failing to recognize events when they were the arguments to other events, failure to recognize nontheme arguments, and sentence segmentation errors. We show that specifically handling coordination had a small but important impact on the overall performance of the system. The OpenDMAP system and the rule set are available at . [ABSTRACT FROM AUTHOR]

This article presents a novel approach to event extraction from biological text using Markov Logic. It can be described by three design decisions: (1) instead of building a pipeline using local classifiers, we design and learn a joint probabilistic model over events in a sentence; (2) instead of developing specific inference and learning algorithms for our joint model, we apply Markov Logic, a general purpose Statistical Relation Learning language, for this task; (3) we represent events as relations over the token indices of a sentence, as opposed to structures that relate event entities to gene or protein mentions. In this article, we extend our original work by providing an error analysis for binding events. Moreover, we investigate the impact of different loss functions to precision, recall and F-measure. Finally, we show how to extract events of different types that share the same event clue. This extension allowed us to improve our performance our performance even further, leading to the third best scores for task 1 (in close range to the second place) and the best results for task 2 with a 14% point margin. [ABSTRACT FROM AUTHOR]

Acronym extraction for business documents has been neglected in favor of acronym extraction for biomedical documents. Although there are overlapping challenges, the semi-structured and non-predictive nature of business documents hinder the effectiveness of the extraction methods used on biomedical documents and fail to deliver the expected performance. A classifier-based extraction subsystem is presented as part of the wider project, Binocle, for the analysis of French business corpora. Explicit and implicit acronym presentation cases are identified using textual and syntactical hints. Among the 7 features extracted from each candidate instance, we introduce 'similarity' features, which compare a candidate's characteristics with average length-related values calculated from a generic acronym repository. Commonly used rules for evaluating the candidate (matching first letters, ordered instances, etc.) are scored and aggregated in a single composite feature that permits a supple classification. One hundred and thirty-eight French business documents from 14 public organizations were used for the training and evaluation corpora, yielding a recall of 90.9% at a precision level of 89.1% for a search space size of 3 sentences. [ABSTRACT FROM AUTHOR]

Why and its counterparts in some languages are argued to be externally merged in a low left-peripheral specifier of a dedicated functional category and subsequently moved to a criterial position higher in the left-peripheral space. The analysis considers both short- and longdistance construals of why, asymmetries of why-extraction in finite and nonfinite complement clauses, the position of 'why' in a multiple wh-movement language like Romanian, and the differences between why and how come. The analysis sharpens the formal differences between movement to a criterial position and movement out of cyclic domains via escape hatches. [ABSTRACT FROM AUTHOR]

When first approaching an unfamiliar domain or requirements document, it is often useful to get a quick grasp of what the essential concepts and entities in the domain are. This process is called abstraction identification, where the word abstraction refers to an entity or concept that has a particular significance in the domain. Abstraction identification has been proposed and evaluated as a useful technique in requirements engineering (RE). In this paper, we propose a new technique for automated abstraction identification called relevance-based abstraction identification (RAI), and evaluate its performance-in multiple configurations and through two refinements-compared to other tools and techniques proposed in the literature, where we find that RAI significantly outperforms previous techniques. We present an experiment measuring the effectiveness of RAI compared to human judgement, and discuss how RAI could be used to good effect in requirements engineering. [ABSTRACT FROM AUTHOR]

This paper deals with the difficult problem of indexing ancient graphic images. It tackles the particular case of indexing drop caps (also called Lettrines) and specifically, considers the problem of letter extraction from this complex graphic images. Based on an analysis of the features of the images to be indexed, an original strategy is proposed. This approach relies on filtering the relevant information, on the basis of Meyer decomposition. Then, in order to accommodate the variability of representation of the information, a Zipf's law modeling enables detection of the regions belonging to the letter, what allows it to be segmented. The overall process is evaluated using a relevant set of images, which shows the relevance of the approach. [ABSTRACT FROM AUTHOR]

Abstract: The existential construction with the predicate –isy in Malagasy is shown to have an unaccusative structure in which the predicate may take either (i) a DP-complement, with an optional locative adjunct and an optional PP or (ii) an IP-complement. The structure is empirically supported by various facts about adverb placement, coordination, word-order, extraction, certain cases of tense-matching between –isy and a following thematic predicate, as well as the lack of the stage-level vs individual-level predicate distinction in the construction. It is argued that the predicate –isy introduces an existential quantifier binding (i) a variable that certain DPs or the locative phrase can in principle provide or (ii) the event or unprojected arguments of a thematic predicate. The definiteness effect arises from existential quantification interacting with the structure of headless relative clauses and the constraint barring strong quantifiers in predicate position. [Copyright &y& Elsevier]

This article presents several acceptability rating experiments concerned with crossing wh-movement in German multiple questions. Our results show that there is no general superiority effect in German, thus refuting claims to the contrary by Featherston (Linguistics 43: 667--711, 2005). However, acceptability is reduced when a wh-phrase crosses a wh-subject with which it agrees in animacy. We explain this finding in terms of the availability of different sorting keys for the answers to the multiple questions. [ABSTRACT FROM AUTHOR]

One of the major problems in the study of language is to understand its biological roots. An effective way to tackle this issue is to explore to which extent we share with other animals some of the mechanisms responsible for its processing. In this paper we present a series of experiments that demonstrate that the abilities required to extract rhythmic and statistical regularities from speech are already present in rodents. Other studies also suggest that some animals might be able to process certain abstract structures that have been traditionally linked to grammar. Together, these studies point towards the idea that many of the basic components of language processing are shared across different species. Nevertheless, there is still a wide array of linguistic phenomena that should be studied from a comparative perspective to determine the extent to which human language uses abilities found in other animals [ABSTRACT FROM AUTHOR]

With the growth of social media, document sentiment classification has become an active area of research in this decade. It can be viewed as a special case of topical classification applied only to subjective portions of a document (sources of sentiment). Hence, the key task in document sentiment classification is extracting subjectivity. Existing approaches to extract subjectivity rely heavily on linguistic resources such as sentiment lexicons and complex supervised patterns based on part-of-speech (POS) information. This makes the task of subjective feature extraction complex and resource dependent. In this work, we try to minimize the dependency on linguistic resources in sentiment classification. We propose a simple and statistical methodology called review summary (RSUMM) and use it in combination with well-known feature selection methods to extract subjectivity. Our experimental results on a movie review dataset prove the effectiveness of the proposed methodology. [ABSTRACT FROM AUTHOR]

This paper presents two original methods for text summarization (by extraction) of a single source document. The first method fulfills the two desiderata of summaries on the base of Formal Concept Analysis (FCA): 1. the relevance of a selected sentence is given by the introduction of the weight for a FCA concept, and 2. the minimal similarity to sentences previously selected is assured by a different coverage in the Concept Lattice. The second method realizes summarization by clustering the sentences. The new measure of similarity, com, has the roots also in FCA and provides the results close to the classical cosine measure. At least to our knowledge, there are no previous attempts to solve the summarization of a text by FCA. [ABSTRACT FROM AUTHOR]

Field Association (FA) Terms-words or phrases that serve to identify document fields are effective in document classification, similar file retrieval and passage retrieval. But the problem lies in the lack of an effective method to extract and select relevant FA Terms to build a comprehensive dictionary of FA Terms. This paper presents a new method to extract, select and rank FA Terms from domain-specific corpora using part-of-speech (POS) pattern rules, corpora comparison and modified tf-idf weighting. Experimental evaluation on 21 fields using 306 MB of domain-specific corpora obtained from English Wikipedia dumps selected up to 2,517 FA Terms (single and compound) per field at precision and recall of 74-97 and 65-98. This is better than the traditional methods. The FA Terms dictionary constructed using this method achieved an average accuracy of 97.6% in identifying the fields of 10,077 test documents collected from Wikipedia, Reuters RCV1 corpus and 20 Newsgroup data set. [ABSTRACT FROM AUTHOR]

Chapter 7 of the book "Foundations and Trends in Information Retrieval" is presented. It explores modern summarization systems, which mostly depend on sentence extraction, and their advances that have been pushed forward due to the establishment of document understanding conference (DUC)/text analysis conference (TAC). It highlights several problematic issues with the use of the systems at DUC and TAC evaluations.

This paper investigates gaps in degree phrases with too, as in John is too rich [ for the monastery to hire ___ ]. We present two curious restrictions on such gapped degree phrases. First, the gaps must ordinarily be anteceded by the subject of the associated gradable adjective. Second, when embedded under intensional verbs, gapped degree phrases are ordinarily restricted to surface scope, unlike their counterparts without gaps. Just as puzzlingly, we show that these restrictions are lifted when there is overt wh-movement in the main clause, revealing a striking similarity between the distribution of gapped degree phrases and so-called parasitic gap constructions. These findings, we argue, suggest that the theory of parasitic gaps needs to accommodate gapped degree phrases. Specifically, they argue that parasitic constituents are null operator structures-and under the right conditions need not be accompanied by matrix wh-movement. [ABSTRACT FROM AUTHOR]

Computational models of semantic memory exploit information about co-occurrences of words in naturally occurring text to extract information about the meaning of the words that are present in the language. Such models implicitly specify a representation of temporal context. Depending on the model, words are said to have occurred in the same context if they are presented within a moving window, within the same sentence, or within the same document. The temporal context model (TCM), which specifies a particular definition of temporal context, has proved useful in the study of episodic memory. The predictive temporal context model (pTCM) uses the same definition of temporal context to generate semantic memory representations. Taken together pTCM and TCM may prove to be part of a general model of declarative memory. [ABSTRACT FROM AUTHOR]