okay now sorry for this small delay I had some special issues so some some
issues to solve first welcome to the first session of this course here
today we are going to go through unit one the introduction you can download
your PowerPoint the PowerPoint the presentation I prepared because there
you also have some extra links and some extra references and for your
personal use we are going to just see briefly what the contents are of the
first chapter and also some other concepts that I think are important for
for for this course to have some clear concepts some concepts in in your
mind clear and if you have any doubts we will we will also answer them you
can ask your doubts here or in the forum in the forum for unit one okay so
this book and this course is going to focus on the construction of
dictionaries and the construction of ontologies but we are also going to
see in the first units the main concepts and also the construction of
lexical products or lexicological databases but especially ontologies and
dictionaries we will also go through the construction of terminological
applications in unit six although you know unit six is not mandatory for
for the for the course let's say that it doesn't um this this information
is not part of the exam but it's very useful for you and if you ever have
to do a bachelor's or uh or a master's degree thesis a master's thesis you
can always use the contents of the last chapter which are written by a
specialized specialist in terminology which is veronique host from Kent
University so we had the pleasure to invite her to write a chapter for for
the book and it's always useful to know about terminology and we will also
have some extra sessions in the future in the near future uh talking about
terminology and be and about the distinction between terminology and
terminograph um you should download the PowerPoint with uh which is
original okay you can also download the one with annotations uh but it's
just some uh some notes that I included um that are just copy paste from
the book okay so when you download this um it's useful for all of you uh
you are asking me whether to download the PowerPoint with annotations or
without annotations the annotations are just copy paste a text from the
from the book that has to do with the part that I include in the PowerPoint
so if you want the annotations is fine but you have everything in the book
or in the course because you know that the first two chapters are a
reproduced in the in the course and you can download them there so what is
lexicography what is the difference between lexicography and lexicology
lexicography is just that the craft of dictionary compilation so everything
that has to do with um with that we will see this in detail in further
units of the book and of the course it is interesting to know that it is
the oldest subdiscipline of linguistics so that's let's say how linguistics
started by the compilation of dictionaries okay there are many types of
dictionaries as you know uh you can have lexicological dictionaries lexical
listeners the normal type of dictionaries let's say the most frequent ones
are the ones that include a lexical unit and then you have some information
about the pronunciation about the the different cognates or variations then
you have the definition we will see all this you also have bilingual
dictionaries you have um specific dictionaries for a specific type of
languages such as um law uh language for instance or technology Etc uh we
will see this more in detail as I said what is the difference between
lexicology and lexicography lexical lexicography as I said is the craft of
a dictionary compilation which is very interesting and then lexicology is
something that is used also in the craft of dictionary compilation because
it includes all kinds of investigation uh for uh in the lexicon of a
language no problem welcome all the ones who arrive a bit late it doesn't
matter uh well um and the ones who cannot attend uh online now
synchronically I will this session is being recorded so welcome everybody
again um so as I said lexicology is the science the scientific
investigation of the lexicon of a language this is a definition uh that you
can find in this reference International Encyclopedia of the social and
behavioral Sciences that has had several editions I I think this definition
is quite clear and interesting uh because it has to do it it is quite wide
because lexicology is white it has to do with historical development uh we
you will see examples in the book for instance in the course book uh I I
myself did my PhD on a lexic the development of a lexicological database of
for old English so uh it had to do with the morphological structure of uh
of English lexical items it sounds complicated you will see some examples
in the book the the database we created is called nerds nerds but there are
many other types of studies um in the field of lexicology they can have to
do with gender with language diversity and uh language change uh
codification of meaning metal language Etc so it's as you see it's quite
more uh let's say abstract compared to lexicography which is more a craft
um focusing on lexicography we have a to call the attention to digital
lexicography obviously this is the nowadays trend a digital lexicography is
what is done nowadays and it's not uh it's a an ongoing development in the
last few decades everything has moved from a paper lexicography from
crafting creating dictionaries that are going to become paper versions to
digital lexicography and there are five main lines of fine lines of these
development that can be distinguished uh first of all the creation of
specific dictionaries for various tasks tasks um and in this part of this
area of digital lexicography has to do a lot with lexicology because these
computer dictionaries are based on a machine readable version of a printed
of an existing printed dictionary uh but they have they can be given
different tasks for instance natural language processing machine
translation the automatic text analysis a data mining sentiment analysis uh
so it is very wide the tasks these dictionaries can have a and uh the what
is important is that all the lexical information is stored on the computer
and it has they have uh they include uh formal representation so it is this
formal representation is different from natural language it's language
independent we will see this now in this presentation a bit further a
formal representation is what we call a meta language for instance we have
the most famous type of um structures or database that have been created
and that are used for this digital development of computer dictionaries
meanings for instance is a is part of different elements this as i said we
are on the introduction of this course uh we'll also uh you will see more
this more in detail in the next units because we also deal with with these
databases with this type of databases and nowadays also there is an
increasing an increasing number of databases this we know for sure already
we all know an amazing amount of databases um they provide computer
generated information about various lexical properties of a language and
for instance word frequency whether they are graphemes or morphemes work
occurrences so all these databases um are used in the in the development of
computer dictionaries which which is very interesting for instance uh here
you have some examples of this type of databases um we have also um lingua
for instance the spanish which is very very much used in in spain and these
are just an examples of databases that that can give a lot of information
and that can be used in the in the development of of electronic
dictionaries digital dictionaries um the other option is the most um the
first one the one that started everything which was just this is this is
what we did with our lexicological database in the first instance the first
thing we did was to transfer the information that we found on printed
dictionaries especially in my case it was an old language uh that doesn't
exist anymore such as all language uh old english sorry so we first emptied
we say bathyard we took all the information we found in the three most
important dictionaries of old english that have ever existed and we compute
we we included everything in a file maker database so it was hand work very
exhaustive and this is what is done at the in the in the first instance um
why because later on the search is more faster you can update once you have
put everything you find in a paper dictionary you include it in a computer
database everything all the information is easy easy to update and in fact
nowadays such as happens in wikipedia there are volunteers or people who
work for us as yeah as volunteers a academic volunteers or general public
that want to participate in this or emeritus researchers that participate
in these type of updates so a dictionary doesn't need to be printed every
year in fact it can be updated almost daily we have here some examples of
digital dictionaries but there are so so many these are just a few examples
um and then something very interesting is once all this information has
been transferred from paper dictionaries from lexicographical databases or
dictionaries once this information is transferred into a machine into a
computer database um then we can keep on updating these and including more
information based on uh computer tools so an example here is the old the
oxford english dictionary uh was based first on five million excel which
was first handwritten or in paper clips now a computer can easily process
corpora of several hundred millions of data currencies and new sources are
rapidly added day by day so we would need a hundred uh researchers uh work
in a hundred years to implement all this information in a so what they are
doing is developing um this analysis by making this automatic computer
tools can also do immediately this type of analysis when you include
certain um i don't know how to say rules or we will see these no certain
mandates such as in this lexicological database where you implement a
special structure meta linguistic structure and then all the iterations
that have to do with this structure are immediately analyzed as uh as as it
imagine for instance in spanish the difference in gender we can say okay
everything that finishes in a is feminine except of of course several
contexts where this is not the case um and implementing certain rules etc
but you can automate automatize this let's say so when the computer
receives specific information it already can classify this lexical item
into a certain category let's say and finally the last development is the
development all these developments are complementary okay the digital
integral lexical systems are integrated so in these digital digital systems
integrate various sources of information that compute that consists of a
computer accessible and expandable corpus a set of tools which allow the
search for certain items statistical analysis there are programs that allow
for for this type of facilities allowances that have these certain
allowances in 2007 we used filemaker and now there is a different interface
this was a type of database that that has been probably superseded by many
other important databases and computer programs that can include all these
facilities all these allowances like searching for an item analyzing the
frequency of occurrence um identifying Andnowords in number of occurrences
understanding no words certain amount of context contexts I will give you
in the course you will find a document with some of these nowadays tools
that are available for free accessible through internet and they are around
let's say 50 types of tools and these are only the ones that are for for
free so there are many other ones that are professional and that you can
pay for or that are still being developed and finally is elected but
steadily proceeding lexical analysis of the corpus by experienced
lexicographers here we are we have human post edition let's say as in
translation we need to have post edition processes and a good translation a
good machine translation will depend whether it needs a lot of
post-edition, which always needs humans because the problem of
disambiguation is the biggest problem in translation as you probably have
seen in the translation course of this degree. We always have in this type
of programs and utilities, we will always need at some point humans to
revise the process. So no matter how, especially because the amount of
possibilities, the amount of revisions will determine the quality of the
final product. So we will always need some type of process. We need humans
for the last processes. Obviously, it's like as I said in machine
translation. So we have seen lexicography and how connected it is to
lexicology, and now we will see lexicology in general, not only related to
lexicography. Lexicology could be more the theoretical framework and it's
less, as I said, less concrete, less specific. And we can study many topics
that have to do with the study of language. For instance, in lexicology, we
study the lexicon. We can study gender. Gender is a key category for
lexicology because it is crucial to understand the nature of lexical
entries. For instance, lexical entries in certain languages will depend on
whether they have a linguistic gender or not in their language. Every
language defines, marks gender in a different way. Here you have an
example. For instance, it is possible for certain nouns to be of different
genders according to the sex of the speaker in Garifuna, which is an Arawak
family spoken in Belize, Honduras, Nicaragua and Guatemala. So the study of
gender is very interesting, and it is something cross-cultural in the sense
that there is no equivalence many times between the how we mark gender or
not. That's why it's, it's so interesting for lexicologists. And it is also
very important, as I said, because it has, it affects the shape of the
lexical entries. And this question concerns obviously lexicologists and, as
you see here, it's also of interest for other disciplines such as
psycholinguistics. Also social linguistics, anthropological linguistics,
etc. An important concept we need to have in mind in this course is the
concept of the idea of conceptualization. You have seen this already in
semantics, probably in detail. Here we are just going to see how this is
connected to the development of ontologies in this case. Because
conceptualization, there are many definitions. This is one of the
definitions. You can find other definitions that fit you better. You are
free to find other definitions, of course, because they will also be
dependent on the theory. The theoretical framework of the person who
creates this definition, who establishes a definition. Conceptualization in
this case, for instance, from what we have included in the book is the
process that leads to the extraction and generalization of relevant
information from one's experience. Here I want to focus on the idea of
experience. Conceptualization has to do with experience. What we
experience? Out of what we experience, we create concepts. We design
concepts in our mind. And these concepts we associate with certain
references. There is a discussion going on whether this is an individual
experience or whether this is a collective experience. But in the designing
of ontologies, we need to see it as a collective process. Otherwise, it's
impossible to design an ontology. And in the definition of rules, in the
compilation of dictionaries also, we need to see conceptualization as
something that we are wired to do. And also, we need to agree on the fact
that a specific concept is going to be used. Which means that depending on
our reality, we have a different experience. Depending on this experience,
we will have different concepts of life. And these concepts are culturally
based because a specific culture is surrounded by specific experiences and
therefore, different concepts. For instance, you will see several examples
in the book. You can think of different ones also. The idea of, for
instance, they use the word for a specific type of rain. But this is a type
of rain that is very frequent in the north of Spain. But not in Andalusia.
So this word will be more used and is created in, let's say, in the north
Spanish language. Or the idea that in France, for instance, they have
several names, many, many types of names for bread. Because they have many
types of breads. Which when in Spain until very recently we had the Pueblo
Hueco and Sobado. That was all. Now we are getting into this world of
bread. But this is quite recent in our culture. In our cultural history. So
now it's when we start to see other types of bread, the Masa Madre, etc.
But when you go to the north of Europe, not only France, you can also see
that in Belgium, in Holland, in Sweden, Denmark, the amount of names they
have, the amount of terms they use to name the concept of bread is amazing.
In this case, they share a specific conceptualization and we don't. So it
has to do with... It becomes language-specific. But first, this culture or
this group of speakers have agreed on these concepts. So conceptualization
is going to directly affect the shaping of specific languages. Another
important beginning, another important distinction that you need to know
about, which you probably have seen already, is the difference between
semaseological approach, the semaseological approach to the study of
language, and the onomaseological approach to the study of language. Both
are complementary. They are not... contrasting opposites. They are
complementary. But it is true that one will be used more in certain
disciplines and the other one will be used more in other disciplines. The
semaseological approach is the one that starts from the form, from the
name, from naming, and considers different meanings of the word and
determines the interrelations with them. It discovers semantic
relationships between different words. So it goes from the term to the
concept. So this is specifically useful in the designing of terminologies.
However, the onomaseological approach starts from the object, from the
concept, and then it analyzes different words that are correlated to this
concept. So this approach helps to discover how meaning is formed and
considers properties and peculiarities. So it goes from the concept, from
the meaning to the word. And this is more... This approach is used in the
designing of ontologies. The use of a metalanguage, as I said, is very
important for several reasons. Because we have different languages and, for
instance, you have seen in semantics that the generative language,
generative grammarians, use one type of metalanguage because this
metalanguage is supposed to represent the underlying structure of all
languages in the world. Functional linguistics or role and reference
grammar use other types of metalanguage which they apply to represent the
underlying structures also of all the languages in the world. So the use of
a metalanguage is cross-linguistic, let's say. It's used to represent
categories, relations and procedures. Here we have just an example which is
from choral representative language. It's a database... It's an ontology.
It's an ontology-based database, lexicological database developed by our
Chancellor Ricardo Mairal and Periñal Pascual and others some years ago.
Here in this link you have the access to a very interesting document for
extra information about how he wrote with Pamela Faber, a very important
linguist in this world of lexical functional... lexicomatic functional
linguistics about the importance of developing a correct metalanguage for
the design of ontologies. So... and databases in general. Metalanguage is
always necessary to build up an ontology, especially when it is aimed at
several languages. Several linguistic models I have given examples have
used different metalanguages. The ideal situation would have been that they
all use the same metalanguage but this is difficult. First of all because
they have... Everybody wants to... to... to improve and to propose a
different metalanguage because they think their metalanguage is going to be
easier or better. And also because their own conception of the structure of
language is different. And they will use symbols that are better able to
represent their own conception of the language. Okay? In my case, for
instance, in my thesis, I applied the metalanguage used by the role and
reference grammar school together with the lexamatic function and lexamatic
school developed by Ricardo Mairán and others. But it is perfectly okay to
use another metalanguage but of course, depending on the purposes, in my
case, for my purposes, the theory, the linguistic, the framework, the
theoretical framework about the conception of language for me was better
for my purposes of representing Old English in this case. For other
linguists maybe generative metalanguage, generative grammar metalanguage is
better. So it depends. It depends first on your theoretical standard. So
there is not a common agreement and there is still not a universal
metalanguage which makes it also in a way difficult for developers and for
lexicologists because nowadays there are thousands of ontologies around the
world that are being developed and it could be interesting if there was
some agreement to create standardization, standard metalanguage for all
types of ontologies. By now, this has not happened yet. This has to do with
language-dependent and language-independent representation.
Language-dependent representation is an example, a very clear example is
dictionaries. Okay? The organization of words takes many forms. Knowledge
representation takes the form of language. However, language-independent
knowledge representation an example would be mathematical annotation,
parsers and this is what we would use in the creation of lexicological
databases, ontologies where we add several mandates such as hierarchical
organization, concepts, sub-concepts, et cetera. Knowledge networks are
developed through the knowledge representation which is
language-independent. The truth is that for instance here as you can see in
the previous example from Corel someone x1 say x2 for say somebody x3
related to a question x4 here they are using English as for the structure
of say as a lexical item in their conceptual representation language. So,
they are using language to represent a concept which is everything related
to the concept of say. Okay? So, the same happens for functional
linguistics for role and reference grammar for cognitive linguistics it is
very difficult to escape the use of language. So, in general there is a
combination of both language itself and non-language items such as numbers
or logical notation coming from philosophy philosophical logics et cetera.
And finally some the book has in this unit some ideas which are a kind of
warm up theoretical ideas and reflections about well the different
approaches to meaning interpretation. We have for instance the idea of
cognitivism in ontologies. Ontologies are influenced by cognitivism in the
fact that cognitivism was the first area in linguistics that introduced the
concept of categorization. Categories are very important for ontologies.
So, this stems from cognitivism. Categorization has to do with how we
structure the world into different categories and these categories have
different types of categories. They can be sort of an item relational items
individual collective functional concepts the structure of the categories
also included relationships all these ideas of how we categorize the world
stem from the world of cognitivism. Cognitivism is not only based on the
last trends which are metaphor and metonymy structures that we apply to
sorry the mechanisms of metaphor and metonymy applied to how we express our
language in figurative language etc. This is let's say one of the recent
developments of cognitivism but the most important one is the idea of
categorization okay then we have the idea of predication predication
started with few more in the 70s and this came into every relationship
every word has a frame a frame structure this also influenced role and
reference grammar a lot and functional grammar from the Netherlands school
from Dick and followers and it turned it ended up in the development of a
resources such as frame net or verb net or word net and finally the
philosophical perspective which is more stance philosophical stance is that
well they some authors such as Nirenburg and Raskin that will be mentioned
a lot in the book argue that the real distinction between ontologies and
natural languages is that languages are used by people and ontologies are
constructed for computers so it's just a philosophical perspective let's
say that is a reflection about what ontologies are for okay there is a
question I see there is a question of an ontology we will see different
examples is something a bit different from a lexical database a lexical
database makes use well a lexical database is part of an ontology in fact
or an ontology can use a lexical database to be developed so you are saying
that WordNet is a lexical database exactly and fangrant KB is an ontology
exactly that's the point yes so yes they are you will see when you enter
that they are different in this you have on the coursebook on the virtual
course a different and interesting talk from Antonio Pareja who is also a
computational linguist and he has a talk about the use of methodology and
the construction of ontologies that can be interesting for you if you want
to clarify your ideas about what an ontology is and how you can develop it
any of you could design an ontology I mean you could design the parts of an
ontology and then create some small examples of what an ontology is and he
describes this in that talk which is in the virtual course so if you want
more information about please go to the website and more information on
ontology in the much for your attention and thank you for watching today
thank you watching this see you bye ...З 謝謝 З ... ... a professor in
Ghent University. They have a very interesting research group there that
are specifically focused on the development of terminological databases for
the companies. I don't know if you know the company Yandenul, for instance.
Yandenul is one of the most important companies in the world. It's
Belgian-based. It's a Belgian company for engineers doing, for instance,
the different canals of the world. I don't know the names now, oh my God.
Well, Canal de Suez, Canal de... I don't remember, sorry, for this pure
ignorance. But okay, they are the ones who do such important constructions.
Also, they have been working in Dubai doing... I don't remember what I know
about the company, but let's say that the biggest constructions in the
world, probably Yandenul will be there. So they handle a lot of
information, a lot of document, a lot of instructions, a lot of manuals,
user manuals. What does terminology do? To handle all this knowledge and
all this language into a good gestion del conocimiento, into a good... You
will see here, into a good database for understanding the terms. If you
have a user manual, for instance, and a specific term and a specific
concept has many, many different uses in the same manual and there are
errors, this is totally difficult to understand. And if you want to
computerize this knowledge to give a machine, for instance, a specific
computer machine or even a building machine or an engineering program, you
want to give specific computer instructions to this program you need to
standardize certain terms so this is basically what terminology does it is
focused on first understandability common understanding of documents
translatability so they create one only possible translation for instance
you have an example here store manager or store administrator they say okay
this is going to be just your nerve do magazine so they take the document
and they make it more simple using one just single translation a
maintenance retrievability has to do with disambiguation this is very
important both in translation and in terminology consistency so using
always the same term for the same concept for instance um input key enter
key line etc for one single concept they will they will use only one single
consistent use so that it is easy to understand okay and also the same for
consistent spelling they will use a main spelling and they will say okay
the main spelling is a record number and there there are different
variations that will appear here there but we are going to substitute all
of them for this main term for instance so as you see it is very much more
um practical oriented going to the focusing on on doing on creating simple
language easy to use a language um to be able to manage language easily by
translators users workers or specific or of a specific company etc um and
then we have another you another part of terminology which is uh only
applied to um to the the companies to to uh yes to companies which is a
more uh also applicable for instance for translators for translators of
engineering translators or science translate translators translators use
need a term bank these term banks are also done by terminographers and
terminologies and in order to do this we need three elements a word
processor spreadsheet database management system or and a terminology
management system and this ends up in the creation of multilingual term
banks which are useful as i said also if you want to dedicate to your life
to translation for instance you have here some links um that are for the
first one is a very famous because it is used by the european institutions
so it has a lot of european commission jargon language which is a whole
word a whole language by itself and these are some other term banks that
you can see as examples of how a term bank is is designed so for
terminology terminological work how would would we do terminology work we
compile a corpus we structure the terms we establish a list of preferred
terms set in the canonical form and then the variants and we include their
translation so if you ever want to create a term bank these are the steps
you need to to follow a turn back is for instance a term bank for uh terms
what is the difference sorry that i didn't mention this the difference
between a word and a term is that the term is a word that refers to
specific language knowledge knowledge area uh for instance medical
terminology economic terminology european terminology okay so that's the
difference between a term bank and a dictionary a dictionary um will focus
on general language in general we can also have specific language in
general so that can natürlich dictionaries, but the way of compiling the
terms in a dictionary is different from a term bank because a term bank
doesn't include all the lexicological information that a dictionary will
include. It's going to focus more on the form. This is the word, this is
the meaning, these are the variants, this is how you translate them, that's
it. So these would be the steps for terminological work and well this is
it. I don't know if you have any doubts. If you don't have doubts now, you
can also... There is a question? Okay. I'm going to stop if you don't mind
here the recording. And then we can continue a bit with some of the
questions, okay?