Ivan Andrei Uemlianin, Ph. D.
This c.v. can also be viewed online at: http://www.llaisdy.com/static/cv.html.
See also pdf version here.
Skills and experience |
Back to Contents |
Employment |
Back to Contents |
I am a self-employed software consultant, specialising in social networking and natural language processing. Below I list some recent projects. See also my open-source projects on github.
Corporate social networking (2015-2016)
For a company in south Cheshire, I developed a Django-based social networking platform based around knowledge-sharing and professional development. This is a small site with a user base of around 300.
SMPP gateway (2013)
For an SMS marketing company based in Guernsey, with offices in London, Cape Town and Melbourne, I developed an SMPP router in erlang. This delivered marketing SMS messages in volume to Europe and Africa.
Cosmo (2011-2014)
This was a therapeutically-oriented social networking platform, designed in collaboration with the couselling staff at Beatbullying, and implemented by me in python and erlang. In 2014 it was running in eight countries across Europe. It was effective and highly regarded. Beatbullying presented a paper about this platform (which describes some of the erlang tech, as well as the therapeutic function) at CHI 2012, entitled "We're Altogether Better".
Big March (2010)
This was a digital marketing campaign which involved a "virtual march" across the web. I was the
main server-side developer (python). The campaign won several prestigious awards.
Research Officer
Language Technologies Unit, Canolfan Bedwyr, University of Wales Bangor; Oct. 2004 - Sept. 2006
details
Speech Recognition Research and Development, Project Leader
Jura Technology Limited; Jul. 2002 - Jul. 2003.
details
Head of Topic Map Development.
Jura Technology Limited; Apr. 2001 - Dec. 2001.
details
Teaching fellow.
School of Psychology, University of Wales Bangor; Dec. 1994 - Mar. 1999.
details
Research and publications |
Back to Contents |
Liaison with Russian researchers
A grant from the Royal Society funded a visit to Pomor State
University, Arkhangelsk in March 1996. During my visit I was able to
make contact with research staff in Artificial Intelligence, English,
Linguistics and Psychology departments, to discover areas of common
interest, and to establish the potential for future collaboration.
Since then, in spite of Russia's difficulties, I have kept in touch with
contacts at the university.
See also employment details
Reviews for CVu can be accessed via the ACCU website: http://accu.org.
Uemlianin, I. A.Education |
Back to Contents |
Diploma in Statistics. Open University. December 2005.
Ph.D. The development of the subject category in first language acquisition. University of Edinburgh, Department of Linguistics. January 1995.
M.A. (Hons) 2i. Linguistics with artificial intelligence. University of Edinburgh, Department of Linguistics. June 1989.
higher education details
Five `A' levels (1985); nine `O' levels (1983).
secondary education details
Conversational Russian, Welsh.
Elementary Chinese (Mandarin).
Some understanding of French, Spanish and Turkish; Classical Latin and Greek.
Teaching in Higher Education. Centre for Learning Development, University of Wales, Bangor. July 1998.
Introduction to philosophical enquiry in education. Society for the advancement
of philosophical enquiry in education. August 1995.
Personal information |
Back to Contents |
Date of birth: 27th December 1966
Nationality: British
Appendix 1: details |
Back to Contents |
Language Technologies Unit, Canolfan Bedywr, University of Wales, Bangor, Bangor, Gwynedd LL57 2DG, Wales, U.K. Oct 2004 - Sept 2006.
Working on the WISPR project I developed the infrastructure used to develop Welsh synthetic voices. This was essentially a set of python scripts wrapping festival and HTK; see Uemlianin (2005a). On the conclusion of WISPR I worked on maintenance and improvement of Cysgliad, Canolfan Bedwyr's flagship package of spelling and grammar checkers, and electronic dictionaries. This involved primarily C++ and C#.
Jura Technology Limited, 35 Norroy Road, London SW15.
Successfully applied for, won and completed a DTI Smart project to assess the feasibility of developing a speech-enabled plug-in for Jura Technology's categorisation and knowledge management software. The project focussed on automatic speech recognition (ASR) but also involved some multimedia processing (i.e., mpeg). Used the following technologies:
Jura Technology Limited, 35 Norroy Road, London SW15. Apr. 2001 - Dec.2001.
Developed standards-compliant Topic Map processor as a middle-ware component of Jura's automatic categorisation and knowledge management software. Participated in software development and strategic business planning. Represented Jura in international discussions, e.g. at the Organisation for the Advancement of Structured Information Standards (OASIS). Used the following technologies:
School of Psychology, University of Wales Bangor, Bangor, Gwynedd LL57 2DG, Wales, U.K. Dec. 1994 - Mar. 1999.
Organised and taught modules on practical phonetics, psycholinguistics and philosophy. Methods included web-based teaching and computer-assisted assessment. This work was instrumental in the School's being awarded an Excellent in the 1996 Teaching Quality Assessment. Supervised final year experimental psychology projects: this involved designing and running experiments, and analysing data using SPSS. Organised the first School of Psychology mini-conference, and produced the conference proceedings. Involved in pastoral care of students and in School administration at all levels.
Department of English Philology, Pomor State University, Lomonosov Pr. 4, 163006 Arkhangelsk, Russia. Mar. - May 1996.
Advised staff on course content; held seminars and tutorials on English phonetics and rhetoric.
Department of Linguistics, University of Edinburgh, Edinburgh, EH8 9LL. Oct. 1989 - Jun. 1994.
Held tutorials on linguistics and phonetics; marked exams and term papers; acted as personal tutor; trained staff on the CHILDES child language database; was elected post-graduate representative.
The encyclopaedia of language and linguistics, Pergamon Press Ltd., Headington Hill Hall, Oxford, OX3 0BW. Oct. 1990 - Jul. 1993.
Carried out editorial work including proof-reading and verification of bibliographies, enabling publication of the encyclopaedia by deadline.
Centre for Speech Technology Research, 80 South Bridge, Edinburgh, EH1 1HN. Jul. 1989 - Sept. 1990.
Designed, compiled, managed and analysed a database of spoken British English, providing a valuable resource for researchers at CSTR; trained staff in spectrograph analysis.
Survey of the Gaelic dialects of Scotland. Dublin Institute for Advanced Studies, School of Celtic Studies, 10 Burlington Road, Dublin 2, Ireland. Jun. - Sept. 1988.
Assisted in the compilation of phonetic data from paper to electronic media for the Survey.
Appendix 2: abstracts |
Back to Contents |
Back to Education |
The concept of the `Subject' is a central one in both formal linguistics and psychology. Formal linguistics tends to speak of `Subject' as a purely grammatical object which has identity only in relation to that over which it predicates. Psychology on the other hand tends to consider `the Subject' as more closely to do with notions of the `Self' and meaningful activity. The difference of these views becomes most profound in the study of first language acquisition.
This area seems to be one in which both perspectives are at a similar weakness, each displaying the drawbacks of its own mode of analysis. The rigidly fixed lexicon characteristic of formal linguistics finds the description of change problematic; psychology, which characteristically eschews `theory' for empirical veracity, can sometimes become too impressionistic to comfortably describe a tendency for systematicity.
The critical review of literature which comprises Chapter 1 takes issue with the Chomskyan formal linguistic assertion that much linguistic ability, including a formally defined Subject, must be innate. It is argued and demonstrated here that the assertion is not a conclusion based on well-founded empirical evidence, but is a consequence of an adherence to formal principles.
Chapter 2 is rooted in an exposition of the developmental ideas of the early Soviet Psychologist Lev Vygotsky - especially in critical relationship with Jean Piaget - whose dialectical forms allow the description of a constantly changing and developing device. Beginning with a recognition that the `Subject' of early child language does not fulfil formal linguistic criteria for Subjecthood, the chapter goes on through a close interpretation of infant behaviour to explore the changing ways in which the child relates to the physical world around it, to its social environment, and to itself. The goal of this chapter is to build an image of the child's growing sense of Self - and thus to a reconstructed psychological notion of `Subject'.
The final chapter 3 returns to formal linguistics and demonstrates the compatibility of the above psychological approach to early cognitive development and the formal linguistic description of the more principled and systematic changes of later child language. The formal linguistic concept of `Subject' is seen to be a development of the earlier psychological notion.
In conclusion, the thesis is intended to be a dialectical synthesis of psychological and formal linguistic methods, which is able to overcome inherent problems of both. The portrait of the development of the `Subject' category is intended to be a portrait in little of the development of language as a whole.
Taking Saussure as our point of departure, motivated by his rational and structural approach to language, we can trace a pattern of ideas in the history of twentieth century linguistics. Analysis of this pattern is instructive in its illustration of the changing conceptions of language and their consequences on aspects of theory and practice.
Saussure's system of dichotomies is portrayed as the foundation of twentieth century structuralist linguistics and also a product of the Comparative and Neogrammarian study of texts. Its re-interpretation in America is described both in terms of the early Structuralists' changed empirical and theoretical circumstances, and Chomsky's later re-assessment of the Structuralist framework as a whole.
Most of the dissertation is then concerned with the development of Generative Grammar, and an investigation into the complex relationship between theory and practice. The biological determinist implications of Government-Binding theory are exposed, and their roots within the framework are examined.
An alternative to biological determinism - a dialectical overview - is suggested, and its implications to linguistic theory and practice are explored - answering questions and raising problems of its own.
A major goal of modern linguistic theory is the development of a theory of Universal Grammar and sentences are parsed using general principles of language. This paper explores Berwick's (1987) ideas about `principles' in relation to existing linguistic theories.
Two theories of grammar are compared in their relative efficiency and coherence in the treatment of Warlpiri, a free word order language of Central Australia. This experiment is used, along with ideas in Berwick's paper, to construct a principle-based system, which reflects the Lexical Functional Grammar (LFG) model of language [parser implemented in Prolog]. This parser is contrasted with Kashket's (1986) parser for Warlpiri which is faithful to the Government-Binding paradigm.
It is concluded that an LFG model of language is closer to what Berwick calls a Principle Based system (notwithstanding his loyalty to GB) and questions are raised about the use of LFG and Unification in the study of Universal Grammar.
.