Search

Knowledge is assumed by cognitive science to consist of concepts that are organised and maintained by complex processes taking place in human minds. These processes are not yet accessible directly. Language is still the primary medium for communicating knowledge and presumably linguistic objects and structures are expressions of knowledge and its organisation in mind. Collecting terms (i.e., creating a specialised vocabulary) and capturing their relationships are thus important mechanisms for distilling knowledge from specialised texts and for formalising it for machines. The approach taken in this thesis is to analyse the co-hyponymy relationships between terms as an organisational mechanism. Co-hyponyms are sets of lexical units sharing a common hypernym; bank and building society, for example, are co-hyponyms of the hypernym financial organisation. Analysing the co-hyponymy relationships between terms is important because it bridges the semantic gap between a) specialised lexical knowledge, b) the quantitative interpretation of meanings in specialised discourse, and c) machine-accessible conceptualisation of knowledge. This thesis proposes the use of a vector-based distributional representation of terms in order to construct a quantitative conceptual model of kinds-sorts in a given field of knowledge. Among empirical methods for analysing linguistic structures, distributional approaches to semantics encode language data to models that should correspond to the meanings of linguistic entities. The meaning of an entity, such as a word or a phrase, is assumed to be a function of its statistical distribution in contexts. In order to use these methods we thus need to define (a) the contexts, that is, which statistical information must be collected; and (b) the functions, that is, how this information must be used to correlate with a meaning. This thesis is a study of corpus-based distributional methods for characterising co-hyponymy between terms. Terms are represented as vectors to form a so-called term-space model. To obviate the curse of dimensionality and to facilitate the construction of models, novel methods employing sparse random projections are proposed. Random Manhattan indexing is used to construct L1-normed spaces and random indexing for L2-normed spaces. Following these steps a memory-based classifier exploits the distance between vectors to identify the presence of targeted co-hyponymy relationships. An evaluation is also performed to assess any reciprocal influences of the method's parameters on its performance. Userfriendliness, flexibility in updating and maintenance, and an innate capacity to resemble conceptual structures in a domain knowledge are the advantages of this method.

In this paper, we propose a method that combines the principles of automatic term recognition and the distributional hypothesis to identify technology terms from a corpus of scientific publications. We employ the random indexing technique to model terms surrounding words, which we call the context window, in a vector space at reduced dimension. The constructed vector space and a set of reference vectors, which represents manually annotated technology terms, in a k-nearest-neighbour voting classification scheme are used for term classification. In this paper, we examine a number of parameters that influence the obtained results. First, we inspect several context configurations, i.e. the effect of the context window size, the direction in which co-occurrence counts are collected, and information about the order of words within the context windows. Second, in the k-nearest-neighbour voting scheme, we study the role that neighbourhood size selection plays, i.e. the value of k. The obtained results are similar to word space models. The performed experiments suggest the best performing context are small (i.e. not wider than 3 words), are extended in both directions and encode the word order information. Moreover, the accomplished experiments suggest that the obtained results, to a great extent, are independent of the value of k. peer-reviewed

Conference paper This paper introduces ACL RD-TEC: a dataset for evaluating the extraction and classification of terms from literature in the domain of computational linguistics. The dataset is derived from the Association for Computational Linguistics anthology reference corpus (ACL ARC). In its first release, the ACL RD-TEC consists of automatically segmented, part-of-speech-tagged ACL ARC documents, three lists of candidate terms, and more than 82,000 manually annotated terms. The annotated terms are marked as either valid or invalid, and valid terms are further classified as technology and non-technology terms. Technology terms signify methods, algorithms, and solutions in computational linguistics. The paper describes the dataset and reports the relevant statistics. We hope the step described in this paper encourages a collaborative effort towards building a full-fledged annotated corpus from the computational linguistics literature. peer-reviewed

Conference paper Vector space models (VSMs) are mathematically well-defined frameworks that have been widely used in text processing. In these models, high-dimensional, often sparse vectors represent text units. In an application, the similarity of vectors and hence the text units that they represent is computed by a distance formula. The high dimensionality of vectors, however, is a barrier to the performance of methods that employ VSMs. Consequently, a dimensionality reduction technique is employed to alleviate this problem. This paper introduces a new method, called Random Manhattan Indexing (RMI), for the construction of L1 normed VSMs at reduced dimensionality. RMI combines the construction of a VSM anddimension reduction into an incremental, and thus scalable, procedure. In order to attain its goal, RMI employs the sparse Cauchy random projections. peer-reviewed

Vector space models (VSMs) are mathematically well-defined frameworks that have been widely used in the distributional approaches to semantics. In VSMs, high-dimensional vectors represent linguistic entities. In an application, the similarity of vectors and thus the entities that they represent is computed by a distance formula. The high dimensionality of vectors, however, is a barrier to the performance of methods that employ VSMs. Consequently, a dimensionality reduction technique is employed to alleviate this problem. This paper introduces a novel technique called Random Manhattan Indexing (RMI) for the construction of L1 normed VSMs at reduced dimensionality. RMI combines the construction of a VSM and dimension reduction into an incremental and thus scalable two-step procedure. In order to attain its goal, RMI employs the sparse Cauchy random projections. We further introduce Random Manhattan Integer Indexing (RMII): a computationally enhanced version of RMI. As shown in the reported experiments, RMI and RMII can be used reliably to estimate the L1 distances between vectors in a vector space of low dimensionality. peer-reviewed

Annotation Guidelines for the ACL RD-TEC (ver 1.0) is set out in this document. The annotator is required to understand the meaning of term, technology term, and invalid term before commencing the annotation task. A de nition of each item is presented here. non-peer-reviewed

Conference paper We propose and evaluate the task of technology term recognition: a method to extract technology terms at a synchronic level from a corpus of scientific publications. The proposed method is built on the principles of terminology extraction and distributional semantics. It is realized as a regression task in a vector space model. In this method, candidate terms are first extracted from text. Subsequently, using the random indexing technique, the extracted candidate terms are represented as vectors in a Euclidean vector space of reduced dimensionality. These vectors are derived from the frequency of co-occurrences of candidate terms and words in windows of text surrounding candidate terms in the input corpus (context window). The constructed vector space and a set of manually tagged technology terms (reference vectors) in a k-nearest neighbours regression framework is then used to identify terms that signify technology concepts. We examine a number of factors that play roles in the performance of the proposed method, i.e. the configuration of context windows, neighborhood size (k) selection, and reference vector size. peer-reviewed

In this paper, we address the problem of extracting technical terms automatically from an unannotated corpus. We introduce a technology term tagger, that is based on Liblinear Support Vector Machines and employs linguistic features including Part of Speech tags and Dependency Structures, in addition to user feedback to perform the task of identification of technology related terms. Our experiments show the applicability of our approach as witnessed by acceptable results on precision and recall. peer-reviewed

International audience; This paper describes the PARSEME Shared Task 1.1 on automatic identification of verbal multi-word expressions. We present the annotation methodology, focusing on changes from last year's shared task. Novel aspects include enhanced annotation guidelines, additional annotated data for most languages, corpora for some new languages, and new evaluation settings. Corpora were created for 20 languages, which are also briefly discussed. We report organizational principles behind the shared task and the evaluation metrics employed for ranking. The 17 participating systems, their methods and obtained results are also presented and analysed.