Yorick was a great

Natural Language Engineering really came about from a meeting between Roberto Garigliano (then of Durham University) and myself in his office in late 1992 or early 1993. I had returned to academia the previous year after a spell doing a variety of jobs in industry, and had become aware of Roberto and the Natural Language Group at Durham (just about 15 miles from the University of Sunderland where I was working). Roberto and I discussed several possible avenues of cooperation, including sponsorship by Durham of students on existing Sunderland masters degrees, a joint Durham/Sunderland specialist Masters in Language Engineering (which came to nothing) and a new journal focused on practical, engineering work in the language domain. Incidentally, one of the sponsored master’s students was Siobhan Devlin, now Head of Computing at Sunderland.

Abstract The paper reviews the state of the art of natural language engineering (NLE) around 1995, when this journal first appeared, and makes a critical comparison with the current state of the art in 2018, as we prepare the 25th Volume. Specifically the then state of the art in parsing, information extraction, chatbots, and dialogue systems, speech processing and machine translation are briefly reviewed. The emergence in the 1980s and 1990s of machine learning (ML) and statistical methods (SM) is noted. Important trends and areas of progress in the subsequent years are identified. In particular, the move to the use of n-grams or skip grams and/or chunking with part of speech tagging and away from whole sentence parsing is noted, as is the increasing dominance of SM and ML. Some outstanding issues which merit further research are briefly pointed out, including metaphor processing and the ethical implications of NLE.

David L. Waltz died on March 22, 2012 after suffering from brain cancer. Dave was a good friend to Natural Language Engineering, and provided some sage advice when Roberto Garigliano and I started working on the proposed journal in the early 1990s; he subsequently agreed to serve as a founding editorial board member.

The 3rd International Workshop on Patent Information Retrieval builds on the experiences of the first two workshops, to provide its participants an exciting, scientifically challenging and interactive event, where the specific issues of patent retrieval may be put into the general context of Information Retrieval and Knowledge Management, in order to explore innovative solutions to new and old problems, but also to evaluate and adapt traditional or classic approaches to new problems. Between the scientific presentations and posters, distinguished keynote speakers and a panel discussion, PaIR 2010 shapes itself into a significant landmark in the field of domain specific information retrieval.

Karen Spärck Jones died peacefully on 4 April 2007 after a number of months battling cancer.Karen was President of the Association for Computation Linguistics in 1994 during a difficult period, and saw through a transitional phase in which a solid basis for the long-term health of the association was established.Karen was an exceptional individual who made very substantial contributions in two separate fields, computational linguistics (CL) and information retrieval (IR), as well as in computing and artificial intelligence more broadly.In IR, her greatest contribution was probably her invention (1972) of the concept of inverse document frequency (IDF), in which the importance of terms is weighted according to the proportion of documents in the corpus in which they occur; the intuition being that terms which occur in many documents are poor index terms.This is the partial basis of all weighting schemes adopted by widely used Internet search engines, and will undoubtedly be a lasting contribution to the field.Her primary inspiration and interest was always language, especially language in practical use.Her long involvement with IR arose (as someone who subsisted for an inordinately long time on soft money) by the need to find a new line of research in the aftermath of the ALPAC Report and the subsequent difficulties in getting machine translation work funded.However, she was always well qualified to work in IR, a topic addressed in her very early publications (see, for example, Masterman, Needham, and Spärck Jones 1958), although in a rather different context to her later work.Given this focus on language it is therefore perhaps odd that her main contributions to CL or natural language processing are harder to pin down.They include contributions to summarization, natural language interfaces to structured databases, dialogue, semantics, and more.In the long term perhaps her greatest contribution will come to be seen to be her Ph.D. thesis (1964), which was far ahead of its time.It brought together statistical or machine learning approaches with the use of an existing resource (Roget's Thesaurus-punched onto cards!) and still can be read with profit today (see Wilks and Tait 2005).It can be viewed as the forerunner of a whole range of more recent attempts to derive semantic information on an empirical basis.It is perhaps as an empiricist that Karen will be best remembered.Her long associations with the US TREC initiatives and the more recent DUC summary evaluations are well known, as is her earlier association with DARPA TIDES, but her desire to work on well-founded experimental bases runs through from her earliest days, initially through involvement with Cyril Cleverdon and the Cranfield project, and then later in the work reported, for example, in Spärck Jones and van Rijsbergen (1976) and Spärck Jones and

Many users of image retrieval systems would prefer to express initial queries using keywords. However, manual keyword indexing is very time-consuming. Therefore, a content-based image retrieval system which can automatically assign keywords to images would be very attractive. Unfortunately, it has proved very challenging to build such systems, except where either the image domain is restricted or the keywords relate only to low-level concepts such as color. This article presents a novel image indexing and classification system, called CLAIRE (CLAssifying Images for REtrieval), composed of one image processing module and three modules of support vector machines for color, texture, and high-level concept classification for keyword assignment. The experimental prototype system described here assigns up to five keywords selected from a controlled vocabulary of 60 terms to each image. The system is trained offline by 1639 examples from the Corel stock photo library. For evaluation, five judges reviewed a sample of 800 unknown images to identify which automatically assigned keywords were actually relevant to the image. The system proved to have an 80% probability to assign at least one relevant keyword to an image.

We report on two experiments performed to test the importance of Term Order in automatic summarisation. Experiment one was undertaken as part of DUC 2004 to which three systems were submitted, each with a different summarisation approach. The system that used document Term Order outperformed those that did not use Term Order in the ROUGE evaluation. Experiment two made use of human evaluations of search engine results, comparing our Query Term Order summaries with a simulation of current Google search engine result summaries in terms of summary quality. Our QTO system's summaries aided users' relevance judgements to a significantly greater extent than Google's.

With this first issue of Volume 10 of Natural Language Engineering we would like to take this opportunity to reflect on the development of the journal, especially over the last year or two, and to look forward to a number of future developments.

This paper describes two studies that looked at users' ability to formulate visual queries with a Content-Based Image Retrieval system that uses dominant image colour as the primary indexing key. The first experiment examined users' performance with two visual search tools, a sketch tool and a structured browsing tool, with different types of image query. The results showed that while users were able to successfully search on the basis of colour, and were able to formulate visual queries, their ability to do so was affected by search task type. Search task type was also shown to be related to search tool choice. However, the results of study two showed that while users were able to complete all of the tasks, there was evidence to suggest that a degree of compromise was present in the users' choice of image that was largely due to problems relating to query formulation.

Word sense ambiguity is recognized as having a detrimental effect on the precision of information retrieval systems in general and web search systems in particular, due to the sparse nature of the queries involved. Despite continued research into the application of automated word sense disambiguation, the question remains as to whether less than 90% accurate automated word sense disambiguation can lead to improvements in retrieval effectiveness. In this study we explore the development and subsequent evaluation of a statistical word sense disambiguation system which demonstrates increased precision from a sense based vector space retrieval model over traditional TF*IDF techniques.

Use of semantic content is one of the major issues which needs to be addressed for improving image retrieval effectiveness. We present a new approach to classify images based on the combination of image processing techniques and hybrid neural networks. Multiple keywords are assigned to an image to represent its main contents, i.e. semantic content. Images are divided into a number of regions and colour and texture features are extracted. The first classifier, a self-organising map (SOM) clusters similar images based on the extracted features. Then, regions of the representative images of these clusters were labeled and used to train the second classifier, composed of several support vector machines (SVMs). Initial experiments on the accuracy of keyword assignment for a small vocabulary are reported.

This is the first issue of Volume 8, and we thought we would take this opportunity to bring readers of Natural Language Engineering up-to-date with various developments with the journal.

Content Based Image Retrieval (CBIR) presents special challenges in terms of how image data is indexed, accessed, and how end systems are evaluated. This paper discusses the design of a CBIR system that uses global colour as the primary indexing key, and a user centered evaluation of the systems visual search tools. The results indicate that users are able to make use of a range of visual search tools, and that different tools are used at different points in the search process. The results also show that the provision of a structured navigation and browsing tool can support image retrieval, particularly in situations in which the user does not have a target image in mind. The results are discussed in terms of their implications for the design of visual search tools, and their implications for the use of user-centered evaluation for CBIR systems.

No abstract available.

No abstract available.

Abstract Nowadays, we are witnessing the development of new information-seeking environments and applications such as hypermedia digital libraries. Information Retrieval (IR) is increasingly embedded in these environments and plays a cornerstone role. However, in hypermedia digital libraries IR is a part of a large and complex user-centred information-seeking environment. In particular, information seeking is also possible using non-analytical, opportunistic and intuitive browsing strategies. This paper discusses the particular evaluation problems posed by these current developments. Current methods based on Recall (R) and Precision (P) for evaluating IR are discussed, and their suitability for evaluating the performance of hypermedia digital libraries is examined. We argue that these evaluation methods cannot be directly applied, mainly because they do not measure the effectiveness of browsing strategies; the underlying notion of relevance ignores the highly interconnected nature of hypermedia information and misses the reality of how information seekers work in these environments. Therefore, we propose a new quantitative evaluation methodology, based on the structural analysis of hypermedia networks and the navigational and search state patterns of information seekers. Although the proposed methodology retains some of the characteristics (and criticisms) of R and P evaluations, it could be more suitable than them for measuring the performance of information-seeking environments where information seekers can utilize arbitrary mixtures of browsing and query-based searching strategies.

The expansion of the World Wide Web (WWW) has created an increasing need for tools capable of supporting WWW authors in composing documents using the HyperText Markup Language (HTML). Currently, most web authors use tools which are basically ordinary text editors and have additional features to facilitate the easy and correct use of HTML tags. This approach places the burden on the web author to design and then create the entire web site in a top-down fashion, without any explicit support for the structural design of the site. In this paper we discuss an alternative structural approach to Web authoring, which is based on the use of the HyperTree hypermedia system as the central authoring tool. The advantages of using HyperTree are two-dimensional. Firstly, web authors can manage a web site as a single complete hypermedia database. For example, HyperTree provides facilities like the automatic creation of indices and the discovery of link inconsistencies. Additionally, it organizes the web pages in an easy to understand hierarchy without using any HTML directly. Secondly, web end-users can benefit from the use of HyperTree, since seeking information in structured web sites is generally less disorientating and develops fewer cognitive overheads. ©1997 John Wiley & Sons, Ltd.

A linguistically motivated approach to indexing, that is the provision of descriptive terms for texts of any kind, is presented and illustrated. The approach is designed to achieve good, i.e. accurate and flexible, indexing by identifying index term sources in the meaning representations built by a powerful general purpose analyser, and providing a range of text expressions constituting semantic and syntactic variants for each term concept. Indexing is seen as a legitimate form of shallow text processing, but one requiring serious semantically based language processing, particularly to obtain well-founded complex terms, which is the main objective of the project described. The type of indexing strategy described is further seen as having utility in a range of applications environments.