Advances in language engineering may be dependent on theoretical principles originating from linguistics, since both share a common object of enquiry, natural language structures. We outline an approach to term extraction that rests on theoretical claims about the structure of words. We use the structural properties of compound words to specifically elicit the sets of terms defined by type hierarchies such as hyponymy and meronymy. The theoretical claims revolve around the head-modifier principle, which determines the formation of a major class of compounds. Significantly it has been suggested that the principle operates in languages other than English. To demonstrate the extendibility of our approach beyond English, we present a case study of term extraction in Chinese, a language whose written form is the vehicle of communication for over 1.3 billion language users, and therefore has great significance for the development of language engineering technologies.

Named entity recognition identifies and classifies entity names in a text document into some predefined categories. It resolves the “who”, “where” and “how much” problems in information extraction and leads to the resolution of the “what” and “how” problems in further processing. This paper presents a Hidden Markov Model (HMM) and proposes a HMM-based named entity recognizer implemented as the system PowerNE. Through the HMM and an effective constraint relaxation algorithm to deal with the data sparseness problem, PowerNE is able to effectively apply and integrate various internal and external evidences of entity names. Currently, four evidences are included: (1) a simple deterministic internal feature of the words, such as capitalization and digitalization; (2) an internal semantic feature of the important triggers; (3) an internal gazetteer feature, which determines the appearance of the current word string in the provided gazetteer list; and (4) an external macro context feature, which deals with the name alias phenomena. In this way, the named entity recognition problem is resolved effectively. PowerNE has been benchmarked with the Message Understanding Conferences (MUC) data. The evaluation shows that, using the formal training and test data of the MUC-6 and MUC-7 English named entity tasks, and it achieves the F-measures of 96.6 and 94.1, respectively. Compared with the best reported machine learning system, it achieves a 1.7 higher F-measure with one quarter of the training data on MUC-6, and a 3.6 higher F-measure with one ninth of the training data on MUC-7. In addition, it performs slightly better than the best reported handcrafted rule-based systems on MUC-6 and MUC-7.

With growing interest in Chinese Language Processing, numerous NLP tools (e.g., word segmenters, part-of-speech taggers, and parsers) for Chinese have been developed all over the world. However, since no large-scale bracketed corpora are available to the public, these tools are trained on corpora with different segmentation criteria, part-of-speech tagsets and bracketing guidelines, and therefore, comparisons are difficult. As a first step towards addressing this issue, we have been preparing a large bracketed corpus since late 1998. The first two installments of the corpus, 250 thousand words of data, fully segmented, POS-tagged and syntactically bracketed, have been released to the public via LDC (www.ldc.upenn.edu). In this paper, we discuss several Chinese linguistic issues and their implications for our treebanking efforts and how we address these issues when developing our annotation guidelines. We also describe our engineering strategies to improve speed while ensuring annotation quality.

Multimodal interfaces are systems that allow input and/or output to be conveyed over multiple channels such as speech, graphics, and gesture. In addition to parsing and understanding separate utterances from different modes such as speech or gesture, multimodal interfaces also need to parse and understand composite multimodal utterances that are distributed over multiple input modes. We present an approach in which multimodal parsing and understanding are achieved using a weighted finite-state device which takes speech and gesture streams as inputs and outputs their joint interpretation. In comparison to previous approaches, this approach is significantly more efficient and provides a more general probabilistic framework for multimodal ambiguity resolution. The approach also enables tight-coupling of multimodal understanding with speech recognition. Since the finite-state approach is more lightweight in computational needs, it can be more readily deployed on a broader range of mobile platforms. We provide speech recognition results that demonstrate compensation effects of exploiting gesture information in a directory assistance and messaging task using a multimodal interface.

The symbolic location of EUROCALL's 2004 conference in Vienna offered to both new members from Eastern Europe and established members from the West an opportunity to review the relationship between computer assisted language learning (CALL) and language teaching in general. CALL is defined as an ‘academic field that explores the role of information and communication technologies in language learning and teaching’ (EUROCALL 1999; for a discussion of CALL as an interdisciplinary research domain, see Levy, 1997). CALL practitioners and researchers have long been aware of the importance of recognition within the broader discipline of language learning and teaching, as the joint EUROCALL/CALICO/IALL Research Policy Statement (EUROCALL 1999) explicitly noted. Yet CALL in fact remains marginalised in several ways which this article will explore. In seeking to promote more effective dissemination of good teaching practices and especially of research in CALL, the article will evoke the UKÕs predominant role in introducing Quality Assurance (QA) to higher education teaching and research – a trend which the Bologna Process will surely intensify throughout Europe. The author will draw on his current role as language research coordinator at the UK's Open University, and on substantial experience as a QA insider in both teaching and research, to analyse successes and failures in dissemination of both research and good teaching/learning practices. He will propose strategies for moving CALL from the margins towards the centre of language learning. In so doing, he will also provide an incidental overview of some key journals and conferences in the domain.

This paper describes some of the pedagogical and technical issues involved in adopting a hybrid approach to CALL materials development. It illustrates some of these issues with reference to a vocational CALL project, LANCAM, which took such a hybrid approach. It describes some of the benefits and considerations involved in hybrid development and evaluates the commercial web development tool, ‘Dreamweaver’ and its extension application ‘Coursebuilder’, which was used to author the LANCAM materials. It argues for a reconsideration of generic web development tools in CALL authoring and also for language experts to be actively involved in the instructional design process.

The potential of corpora for language learning and teaching has been widely acknowledged and their ready availability on the Web has facilitated access for a broad range of users, including language teachers and learners. However, the integration of corpora into general language learning and teaching practice has so far been disappointing. In this paper, I will argue that the shape of many existing corpora, designed with linguistic research goals in mind, clashes with pedagogic requirements for corpus design and use. Hence, a ‘pedagogic mediation of corpora’ is required (cf. Widdowson, 2003). I will also show that the realisation of this requirement touches on both the development of appropriate corpora and the ways in which they are exploited by learners and teachers. I will use a small English Interview Corpus (ELISA) to outline possible solutions for a pedagogic mediation. The major aspect of this is the combination of two approaches to the analysis and exploitation of a pedagogically relevant corpus: a corpus-based and a discourse-based approach.

In this paper we present the Report Manager of our Web-based tutoring system for German – an interface to a persistent learner report that students can inspect and manipulate. The learner report collects and retains information on the learner’s progress and performance that is saved between visits. It also provides access to a journaling system that records prior inputs along with detailed error analysis, and a bookmarking system that tracks exercise completion. Finally, results can be printed or emailed to an instructor. In the Fall semester 2003, we conducted a study with 87 learners of German to investigate how their study habits are affected by an inspectable learner report. Our study indicates that students reviewed their learner profiles frequently – 33 times on average over the course of a semester – and that 70% of the students repeated exercises after having inspected their learner profile. So, while detailed user profiles and performance tracking are invaluable tools for CALL researchers, our study suggests that students also utilize and are influenced by the information.