Harold L. Somers

Handbook of Natural Language Processing

Symbolic approaches to natural language processing tokenisation and sentence segmentation lexical analysis parsing techniques semantic analysis discourse structure and intention recognition natural language generation intelligent writing assistance database interfaces information extraction the generation of reports from databases the generation of multimedia presentations machine translation dialogue systems - from theory to practice in TRAINS-96 empirical approaches to natural language processing corpus creation for data-intensive linguistics part-of-speech tagging alignment contextual word similarity computing similarity collocations statistical parsing authorship identificaiton and computational stylometry lexical knowledge acquisition example-based machine translation word-sense disambiguation NLP based on artificial neural-networks - introduction.

An Overview of EBMT

In the last ten to fifteen years there has been a significant amount of research in Machine Translation within a ‘new’ paradigm of empirical approaches, often labelled collectively as ‘Example-based’ approaches. The first manifestation of this approach caused some surprise and hostility among observers more used to different ways of working, but the techniques were quickly adopted and adapted by many researchers, often creating hybrid systems. This paper reviews the various research efforts within this paradigm reported to date, and attempts a categorisation of different manifestations of the general approach. This paper first appeared in 1999 in Machine Translation 14:113-157. It has been updated with a small number of revisions, and references to more recent work.

EBMT Seen as Case-Based Reasoning

This paper looks at EBMT from the perspective of the Case-based Reasoning (CBR) paradigm. We attempt to describe the task of machine translation seen as a potential application of CBR, and attempt to describe MT in standard CBR terms. The aim is to see if other applications of CBR can suggest better ways to approach EBMT.

Current research in Machine Translation

This paper, accompanied by peer group commentary and authors response, is a discussion paper concerning the state of the art in Machine Translation. The current orthodoxy is first summarized, then criticized. A number of research projects based on the standard architecture are discussed: they involve the use of Artificial Intelligence techniques, advanced linguistic theories, and sublanguage. Alternative approaches discussed are systems which develop or update their grammars semi-automatically, dialogue MT, and corpus-based MT including example-based and statistical approaches.

Multilingual Generation and Summarization of Job Adverts: the TREE Project

A multilingual Internet-based employment advertisement system is described. Job ads are submitted as e-mail texts, analysed by an example-based pattern matcher and stored in language-independent schemas in an object-oriented database. Users can search the database in their own language and get customized summaries of the job ads. The query engine uses symbolic case-based reasoning techniques, while the generation module integrates canned text, templates, and grammar rules to produce texts and hypertexts in a simple way.

Computer-based support for patients with limited English

The paper describes a proposal for computer-based aids for patients with limited or no English. The paper describes the barriers to health-care experienced due to linguistic problems, then suggests some computer-based remedies incorporating a multi-engine machine translation system based on a corpus of doctor-patient interviews which provides a dialogue model for the system. The doctors and patients interfaces are described. Ideas from Augmentative and Alternative Communication and in particular picture-based communication are incorporated. The initial proposal will focus on Urdu- and Somali-speaking patients with respiratory problems.

Spontaneous and imitated responses in articulation testing

SummaryThis paper represents a renewed investigation into the question of spontaneous and imitated responses in articulation testing. It first reviews the work done in this field in a critical manner, thereby forming the basis of a new experimental design. The subjects of the experiment were children in the reception classes of two Bolton infant schools. Apart from comparing the two modes of testing, other interactions were also investigated, namely sex, age (c. 4; -9 versus c. 5; -9), and order of testing. The test used was the Edinburgh Articulation Test. This experiment differs from all its predecessors in that it involves British children and a British test: to the best of our knowledge this is the first time this has been done. The results suggested that there is a significant difference between the two modes, that older children perform better on both modes when given the imitation test first, and that girls of this age are overall slightly better articulators than boys. We conclude that elicitation...

A Practical Approach to Using Machine Translation Software

AbstractHuman-Aided Machine Translation (HAMT) relies on the traditional roles of pre-editing, interaction and post-editing. ‘Post-editing’ is an activity generally understood to involve revision of the output ‘in the light of errors’ made by the system. This article proposes a new type of activity which involves changing the source text in the light of errors made by the software in order to improve the performance of the system. This activity involves adjusting the ‘input’ (rather than output) to the system and combines the possibility of controlling the errors that an MT system produces by manipulating its input (the source text) with the possibility of reacting to the errors once they have been seen, which is normally thought of as post-editing. The novelty lies in ‘post-editing’ the input, not the output. Some examples illustrating this approach are given and the implications for improving the efficiency of HAMT systems are discussed.

The Stanford Machine Translation Project

This chapter contains sections titled: Introduction, The Structure of the Translation and Organization System, The Processing of English Text, Fragmentation and Isolation, PICKUP and EXTEND, The Interlingual Representation, The Dictionary Format, The Generation of French, References

EXPERIENCES WITH AN ON-LINE TRANSLATING DIALOGUE SYSTEM

An English-Japanese bi-directional machine translation system was connected to a keyboard conversation function on a workstation, and tested via a satellite link with users in Japan and Switzerland. The set-up is described, and some informal observations on the nature of the bilingual dialogues reported.