Open Science and Collaborations in Digital Humanities Part 2

Video in TIB AV-Portal: Open Science and Collaborations in Digital Humanities Part 2

Formal Metadata

Title
Open Science and Collaborations in Digital Humanities Part 2
Alternative Title
Quantitative and qualitative methods
Title of Series
Part Number
2
Number of Parts
4
Author
License
No Open Access License:
German copyright law applies. This film may be used for your own use but it may not be distributed via the internet or passed on to external parties.
Identifiers
Publisher
Release Date
2019
Language
English
Production Year
2019
Production Place
Dubrovnik, Croatia
Loading...
Axiom of choice Trajectory Complex (psychology) Group action Presentation of a group Source code Floppy disk Numbering scheme Software bug Different (Kate Ryan album) Core dump Kognitionswissenschaft Physical system Collaborationism Touchscreen Sampling (statistics) Electronic mailing list 10 (number) XING Quicksort Cycle (graph theory) Newton's law of universal gravitation Arithmetic progression Writing Reading (process) Arc (geometry) Web page Slide rule Civil engineering Similarity (geometry) Online help Rule of inference Frequency Latent heat Pi Term (mathematics) Radiology information system Percolation theory Operating system Authorization Data structure Lie group Computer-assisted translation Metropolitan area network Form (programming) Pairwise comparison Standard deviation Focus (optics) Quantum state Weight Schmelze <Betrieb> Expert system Volume (thermodynamics) Line (geometry) Extreme programming Grass (card game) System call Frame problem Word Uniform resource locator Software Mixed reality Universe (mathematics) ECos Interpreter (computing) Game theory Building State of matter Differential (mechanical device) Plotter Multiplication sign Direction (geometry) Decision theory View (database) Sheaf (mathematics) Combinational logic Computer font Mereology Likelihood function Dressing (medical) Mathematics Ontology Thermal fluctuations Data Encryption Standard Square number Cuboid Endliche Modelltheorie Extension (kinesiology) Thumbnail Scripting language Service Pack Main sequence Closed set Physicalism Stiction Commercial Orbital Transportation Services Measurement Oscillation Connected space Right angle Asynchronous Transfer Mode Row (database) Probability distribution Server (computing) Identifiability Link (knot theory) Virtual machine Hidden Markov model Heat transfer Repetition Attribute grammar Wave packet Revision control Force Robotics String (computer science) Linear programming Addition Dialect Shift operator Mathematical analysis Evolute Peer-to-peer Point cloud Near-ring Context awareness Execution unit Turtle graphics Expected value Medical imaging Avatar (2009 film) Roundness (object) Hypermedia Single-precision floating-point format Kerr-Lösung Error message Descriptive statistics Texture mapping Arm Digitizing Sound effect Bit Lace Dean number Arithmetic mean Wave Process (computing) Internet service provider Order (biology) Mathematical singularity Moving average File viewer Spacetime Point (geometry) Sine Ripping Transformation (genetics) Letterpress printing Translation (relic) Mass Code Computer icon Product (business) Gaussian elimination Intrusion detection system Hierarchy Queue (abstract data type) Representation (politics) Energy level Netzwerkdatenbanksystem Computer architecture MIDI Graph (mathematics) Matching (graph theory) Scaling (geometry) Information Artificial neural network Physical law Content (media) Cartesian coordinate system Magnetic-core memory Personal digital assistant Network topology Table (information) NP-hard Code Length 40 (number) Modal logic 1 (number) Set (mathematics) Insertion loss Semantics (computer science) Usability Formal language Web 2.0 Bit rate Semiconductor memory Diagram Position operator Area Email Simulation Algorithm Electric generator File format Data storage device Fitness function 19 (number) Data mining Vector space Uniformer Raum Hard disk drive Website Metric system Bounded variation Resultant Thomas Bayes Game controller Observational study Markup language Temporal logic Graph coloring Theory Power (physics) Twitter Causality Tunis Module (mathematics) Multiplication Projective plane Expression Horizon Analytic set Morley's categoricity theorem Einbettung <Mathematik> Formal grammar Speech synthesis Fuzzy logic Natural language Object (grammar) Abstraction Maß <Mathematik>
a will put ok i'm so a how work come back we're going to get into the next session now here i again on call that ativan cantet it methods but more i about to at dana formats and a taishan zz an units lang which structures is and they in frames of meaning and.
so there we are practical it the end again similar to the previous one but just building on on mm that about meaning change in woods so want to kind of go through layers of abstraction thought hope fl it'll mike cents once so get towards the end i'm the latter of obstruction is really you working with data but it has to be stored somewhere and.
all tem utley your radikal cations that consume unt a and so the letter of obstruction is really we don't use hard disks quite so much jenny mall that this is the technologies in the four mites an the standards that have come in the past there's anybody know at this is on the slide.
yen this is a core memory the this is about one kill about of memory stored in that an on teen fifty's at tell i store to put electrical impulses stir in each of the arise on the side innit would polar eyes a magnet that was ruck pte around the into section of the k. waals and that's hey stuart of bit yet so though we kind of doing that now.
now with natural anguish pos a sing but just not in physical hard way hua almost inpe is ical harder a and the whole point of this is to recognize that obsolete power corrupts ob sully pte ll he so at every layer of this obstruction in i'm going to guy three now you got standards and four mites and technologies changing all the way and the impact of these chain.
ms a is well what what is the impact of these changes on meaning you nurse a ted nelson's my here had had so the next lev a lot is the in coding leia a lot of he would be famille your these theories i will kind of start of with most hood which is really dots and dashes and ce you signals and sounds being submitter of a single.
or wires are we've book that up quite substantially these that ace out the next lat op from that his form an pts fall or forma pts i'm shore you're all familiar with our stand of power forma nts yet but we don't really consider these as meaning nw full changes we just use them as he to ll a-t. we don't consider what the impact richt a.
if for much changes are on the meaning of the tex that we use building on for mutts with n. get into sc aim is and on tala g.'s n n o you don't some sessions on the sme aaj nw tick way of an the knowledge graph and things like this this is a simple relational they'd a base.
but the principles of the same using pour mites an the your then building more structures more meaningful structures on top of that and ultimate they we end up with applications this is what we do all of this lie a or zz of representation for end when you working with texture really kind of working at the inn coating forma nts and sc aim is lays yeas a rib addi familiar with the ce.
technologies these form utz you know ths again they always change though always change and the best wh wait to prevent against that changes to recognize that the changes always going to has an standards help as well i originally those the u s pask eco cha it not sher of any what he sane.
and i ting serve tea to version but it was very constrained yet eight didn't allow very many i a known english characters that was extended i obviously to ascii with at wh welt yet an extended again or g. euna curried which is what we all work with today i'm not scheu if the dater that you end up.
but processing is uni kurd prop or when you work with or will you have to go through transformation processes they will affect carrick to changes a niche or per taishan zz so on top of that kurt ifs got that code point in the uni kurds standards nine seven three one is the little glyphs ff full of the snowman.
and on top of that we got funt changes with operating systems as well and are not really shoal what the impact of this is on meaning changed so i think i rendered the same code coy n't on the macin took a screen chote or the here out i'd this much be the mac cme act andrs to have more colours on an a come or meme which operating system that was as well but i had no choice over the i con a graph he or the auc.
can ugh roughy that was being used for the representation of this cope point and for may its tool looks like assn omagh n't but if this is a more meaningful erg live if it's a character for example and it's being presented in a slightly different font format and yet have control over that at the application lay a then i not really shore had a measure the meaning change of that so this is another lie urdd obstruction for.
an sidor i shun a and with some of the twitter what that we've been doing we were extracting a mole tea but a mower ge he and kind of a mailing those bakken for for monks the project group and riddick struct a the over hakan say there's to skin ton earned thumbs up https yep when a copy dis ch string that i add struct id extracted using partha nw.
from u.t. eth eight strings i pace today into my mac my ol application and i a mao that to jane and who open dud up in the marks of wh word application and replied to me markers of word thes interpreted the bit level coding and strict out the multi byatt u.t. eth a a mogi car ids.
this at they sickly stripped that these little squares on the side which are their skin coloration character's now that's just a bargh in a pace of core software in mars' softs out look software that was probably written in the eighty's to some auld texts standard but these days you could probably inter pert that as a form of racism if it stripping skin color off.
and you multi bought in kurd ing gns and so this is where it's kind of getting into a slippery tear it you between bugs in interpretation of maine ing and the social contexts that and part from the as so are going up again from the next level is up an to the fore that's at everybody is probably quite familiar with x m l you know.
this is not the same document but the job a script object no taishan is syria lies a shun of the same data but syria lies i shun is not really considered as a meaningful change it's more just a structural change i'm not really shore if serialisations have an impact on meaning change does anyone.
and have any examples where that has happened.
known i think the most meaningful change that a serialisation change will have is on the skill lls and the tools that you need to learn in order to work with the text to work with the day are so it's not so much an meaning change of the content itself but a meaningful change in that you need to learn a whole new set.
of tools and applications in or to work with the day i.
tay is svay that tab separated formats a gyan there's what's of lahr breeze and scripts that allow you to work with they a's and this is a a simpler data structure it's a tabular data structure you could use commas instead of tabs for example to split things out but again in it to learn a whole new set of tools and techniques and lahr breeze to load these up an it or eight through the.
structures and split them cirque it's not main ing china nj but it's a meaning full change in the methods and tools and technics titty ne to use to work with this tayde off and id eth will our idea has not really a form a it's more of a sc aim a.
and sss.
the way eat hw are.
it.
standard sss in scheme is help us learn a specific tool an tech naik and soft will arbor he to work more rapidly rift additional or data of this format so we're off i'm ill yer with the hypes chea mao stand id where i will to work with web pay just quite rapidly n if you working with web ah car vz being from ill you with the tags of aisle of ol.
in thank ched you know is one why of extracting diet or what you not going to be quite sir familiar with is how hr specific web page.
used that tag.
i weep sz is i guessed a contexts to consider on ump i.d.f. a game you've probably expose to ardi f m the knowledge graph its khan of difficult to get yeah head around for a most people that on computational in if thinking i'm.
so yeah anyway it i guess.
a who works with triple lls in terse ol forma.
you sheekey know you knows what triples in turtle forma.
there.
and i cade sir.
so at it it it a jt ole it loomed it but if has serial ization is representing the same information and it's more difficult to util are as they in you kind of saw.
i'll owing off using your data based on it syria lies ation.
i'm these are also not human readable i.d.f. is not human readable this his id affix an outs very of about its there's a lot of attributes in the x m l or that link it through ge a the ontology ease that h. of the feel values us a post to about represent it's not human readable what they've done in digital humanities is create the texting kurd ing in ih should ia.
if which was much more focused aunt lane markup much more focused on the content rather than machine verbosity if that make sense and so are for example the tea i spec has a section for pros tags and you can see the tag ste used in the x m l here very lane there any true carrick just represent to claw was then.
s. tag pte represent to sentence so it's the kind of very lane tags used for syntax ma cop so eking get quite into the text wih that we're being overwhelmed by the measured art or and there is another it kind of sub more jewel in the tea i standard for.
talking out with her says so there is and a tay shuns here for the top averse so there's a sonnett or acqua train or ter set for different taught suv pros an and l. tag for each line of text so what i'm getting at here is that adi f. is used for mush anes it's very of about us and it's not really human readable but if you've writing scripts that will link.
ced together then it makes a lit of sense cause your tache you know a dent afire zte to on tala g.'s to relight paces of data yeah t i was much more are bout the the kind of a the interpretation a mark up of the content and as.
wh well yeah and.
and what happened agus with tea ivor the years is that it buik a it was an x m l by ste standard but it expanded to include also pts of ma jal's for all sorts of specialist literary interpretations and it's mostly english pastes thi ch as so are our a simple analytic mechanisms or this some linking in segmentation.
tag stick can be used and h. of these models with in the tea icon of work a little bit like ontology ge do in that you bara tag ce from one module an the other and you put them together into the mark up of the document that's interesting for you at tohn.
and a mm it's also been used i guess for sz by shull arrangements as well and some tools have been bilk on top of on top of or it and this is for i guess genetic mark up of tex where you can and a tight and it-i shuns yet so you can say the the box his oeuvre a side here actually.
really marking up with t i tags m i hand written correction and then a correction to the hand written correction and they can tyne the coal warden a nts of the position of these mocks on the pied age which sz she's useful for some kind of an owl asus mostly it's what the benefit of the research are going.
and frew the thinking process of the or third who was hand writing they ce things and i'm correcting they document as pot of the kind of hr evasion re process why corrections said certain things thw either or mocked in certain positions so it's more difficult to see have our mockup alike this could be use computationally but it's.
a very useful for the thinking and the kuala teddy process to guerre through the inter pert i shun of the dock human when you've got very specific tags and standards to help constrain the way that your interpret ing and digitizing texts thats i'm a pa i knock it lay got lutz the of i.
ites eighth it ning what yea i'm pts majdal pts i that that it this that that correct thick got the well they my that's that it's a sat un's unturned and i miss how thieu its it pull of whole the tick furyk cont you chun that what stage an dippers s. they limb aiding you think mmmm.
which is or i guess the deconstruction of the the group f. it that was ste involve din the production of texts.
and i so a with multi mode you look stand ids wh with ontology zit you come bar ub its him paces from which forms of annotation do you pih can choose and which units do you then pick to ma cock and to and a tight so a of going to gk try to go into a little bit more linguistics now would talk about.
measurement units forma nts and frames and some of the literary tech naik ste that have been used in digital the triche or and i was so ce matthew jock ers has recently really ce to book and this is his plot of a sentiment analysis for the fluctuation of that what a fuck shoe ation in sentiment in alice's for the plot trajectory for joyce's.
as putsch or of the artist and i think this is simple centamin alice asus just doing token matching it's not doing n e graham a pa zing but you can say that a grass like this is not very good him in its measured but it's not it difficult to interpret i what he's looking hear a here is coasts shies it a linney a progression of a naret.
to which helps to understand the up manna at the or for presents the vents to the read us so this is over the course of the story auc and and he was speaking i think to somebody from cern who suggested that he prove perform a four he a transfer my shun on that data which transforms it into or clot light this which looks.
this like it's more meaningful and gives you a sense that this some positive things happening at the stott there's a does oster in the middle of the story and then everything in the up happy it the n what a four he a transfer my shun also does is bring everything for multiple volume's into the same kind of narrative time sir firms ear itch one hundred server a godless of the length of the volume which the the.
korea ce diagram shows you this is lecht five thousand pages this is kind of bring it into a percent tolle form so that you can then stott to compare the comparison of the narratives a crossed multiple volume's yet so what this is done is transform the measurement and form out it for comparison at a heil evel of interpretation yeah i'm he did.
it this in iin third or he twenty fifty ain and in i pull twenty fifteen he was chris sar zte for and for a changing the main ing of the interpretation of the tex and so a what is done here is then clot the for ih transfer my shunning graeme the rolling main in blue a and the original oh sentiment a m.
and points in black behind the sane than the can see the four he a transfer my shun pay kes hear it about seventy but it's actually bain that paik is probably influence by the narrative time out round id eighty five it so it's actually shifted narrative time through this furry a transfer my shun and while it might seem at a heil evel.
so more useful to compare multiple texts across with the for it transform version of the the literary narrative it's actually kind a shifting things back and forth in time when you do that and at a lower level of interpretation when euro reading of all you more reading a text more closely perhaps at makes more sensing contextual terms to know that it.
happened towards the end of the vole the i'm rather than say the chap to be fall.
so are then kind of units ict a assume everybody's from ill year with word vectors yeah and.
i so another way of changqing up texts i thinking about the layers of abstraction building on top of this word to veh qc aim our thing about twenty fifty ain a a myth used skip grams sower building a top of individual word make tryst sees its now kind of taking a word an looking for the the.
yes pts the by grams next to it study to build up another a mm vector representation on top of that and then taking that to the next extreme is birch has a pre body had a play with birch you near but is a new architecture bi directional in code are representations it's a transform.
tech cer that is a century building a crossed the engram skip graham buyer graham kind of dean wh in that state instead of dropping out were instead of kind of looking for word contexts yet it drops out words in it sentence and then builds probabilistic models on top of that and it does it not just with every other oil every.
second wordy a stotz to build further on top of that and dropping out the next sentence and filling that with a random sentence off two words and building a probability distribution across whether a not the sentence that follows a you know should follow it probably stickley on not i'm and then you can kind of ths thiede it in dh seed texts any ol liu's this probabilistic the mitral.
ix to start to generate text but again all of this is theoretically by stoff of wood token i was ation and would frequencies i'm not show or i haven't looked into the day tells of the book architecture whether it's using for in ames or more fame's or any other grammatical structures in the way that it's representing texts contexts probable this tickly.
so earn i thing the id here is that we're just kind of building abstractions upon obstructions and starting to lose some of the more meaningfully grounder dh linguistic contexts.
so it getting back to the heil of or of worth fricken sees this is another has it trust example of from ted underwood's book distant horizons which is about distant ratings i'm and hear his plodded the frequency of kola pte earns in a random sample fiction and is noticed that asked hr eighteen hundred the a mm the frequency of.
the of color words is starting to increase and the spat had a bit of a hunch about whether or not it represented the decline in third person nour ation in fiction a was kind of shifting from furred person aeration to a more descriptive way of writing and he re look for another study and found one.
from stanford in twenty twell oaf from hoy xer and the cat which were looking at the rising frequency of conch it concrete ag jeh ck dues in nah pte iin century novels soeur colors names body cots not just colors but other relate to descriptive terms these to physical descriptions end he gether a not just their died or but the half he trust.
data and plotted a longer time period and then decomposed a down into as john are as so a can stott see use this is a correlation metric for his hunch in the previous diagram that the way that literature was changing oft ray teen hundred was studying to become more descriptive it and way found here is in the fiction you can sade that there.
it the frequency of the color words is rising a lot mort than the fricken see of the color words in by arb roughy which suggests that the their genre a and the form of right xing in fiction is changing ivor tohn and its is just with very high level word frequency an hour cease and kind of taking two studies and correlating them together to form an interpretation.
so it what we don't know at this is what the annotation unit was a i m how as he plug of these ease over sue not using were pages but he could be going dance of the paragraph level in terms of ease aggregations among pretty shore its word frequencies but u z using say the half he trust data in would frequencies by.
a pied age ch not really shaw if that would have too much impact on how wheeze both up the frequency and aiw assess across this but i'm another decision that you need to mike about kind analyzing you dater is which what is a dock human what's your definition of a dock human i'm is a going to be the entire pull because a going to be just a single tway to zit can earn to be say.
just a chapter of a bulcke the m. packed of these des cision zz on the park line that you bills for analysis is sss kind of difficult to measure in some weiss.
so i'm lang which structures.
talk a little bit about kind of internal i'm external mort all ce whose heard of soccer linguistics.
yeah ok who's have a so sz your linguistics.
an a-k. is sort of a rough kerr leyshon here between internal quantitative an ex turn ol called a qca qualitative yet it's not direct but that's for i'm trying to draw the connection to twente.
so psych a language model sss i'm cycling with ticks is really about the the psychology and the brain science of the interpretation of language yen it look so that how long term an short term memories are ff eh qt our understanding in interpretation of the words that we hear i'm it looks ert the way that words are code it a incur to.
it the phone aims and the more fame's that we hear and whether or not they have priming affects it can also look at the kinds of glitches that the human brain exhibit soeur this things i think called spooner ism sss asuni one heard of what spinner is amiss.
kay have gotten example so they kind of slips of the town yep he hugh say the dia auld queen cui said the quiett auld dane baksh lee done a segment switch between the de ann the queue on dane and coy n't yet so papal will express this wih that n e per script he form of language this is just the way that we naturally speight so.
so i'm there's word switching that happens so the rules of word for my shun can be art urdd as the words of rule formations a people exhibit these sorts of expressive errors which give a bit of an in saw into the way that the brain is actually doing language process some and another one is more frame switching so a tease apple sentence is i forgot.
but in about that all you might accidently say i'd forgot about an that yet so these little kind of examples that humans expressed give a bit of an in so i into the linguistic structure and this is the sorts of things that cycling with stick language modeling he's looking to interpret in and i guess a lot of this.
is difficult to grab want turn.
so ce year language models they looking about the languages i'm selves the dialects a mm the regional influence as of a time of the literature and the verse as an the songs the get transmitted between them i'm the ethnography ove www a language use i'm and how that a fx the culture of the papal that use them so this is an example from a.
project from the digital humanities conference in twenty nine tain which was just a few months ago a on digital focal ris ticks and there are looking at the kind of nor than regions of cha alms and sewn mns an the wave these these kind of cultures were transmitted the into my shun was transmitted and pot of the and aiw asus that.
they did is that they were what the ecj accts songs they realized that the personel pronouns which are often used to stop words and dropped at of standard analytical park lines for sim pour cons of processes actually had a fairly market impact on the categorize ation of the epic songs in the region and those correlated to different.
ply ce as.
so it what have got in gri us south a stern he end finish chogm zz where identified by personel pry an air ons ers a feature in the language so are dropping stock words at is usually used as a way to expert eye pte were to simplify some of the aisle and allitt ical processing but when you're thinking about social contexts.
they're actually really it's really useful to know if someone's ix pursing it from the first person or the furred person and you can then draw conclusions from that as will so a speaking men more just generically about willing gris tick models m.
the way you chum thing zam and can have an impact on the way that your scripts and your results a interpreted serve stemming for example and of e c effects the meaning of words but it's a computational technic to simple fi your dater representation zz m.
you then have kind of moeti lng wool models to work with which is starting to become quite.
there's no standards that i'm aware of yet i'm what be wrong but the standards for reuse of date a model's across mall tooling whole software treuille zz is starting to kind of accrete but not show have too many standard jet you goes much if are many with mall i've used spice she which is a page pay that tro language processing ll arbor in their very good about.
other is ing through eason a izzie download sss their multi lingle models i'm but i gain i think well the veni got one multi language model the otherwise they've got individual language models anse a mm not to show or whether or not they have kind of providing old the deep linguistic features that hugh much be used to using in which case.
down muddying the french list here if it's just word embedding zz and but you need something else for it you going shun he to produce an additional model for the tops are in our suss that the want to do in french.
so is a gte massive fracture of the models that you need to learn to use inter a the so ft were tools that you bring together into you nuch rell and ridge processing pop luntz.
i'm so it well fd hot these ers some bei or the models that are available i'm as anybody use many of these.
none a german net is a german language approval unt a word net and word next used quite extensive lng pow i'm the lleyn chris tick data consortia mm has quite a lot of died or on their rip ah za trees that you can surf through on browse around i'll you need to check what the provenance of all of that dater is beef.
for you start to use it but i guess that's the challenge of doing the nhgri stick research how i'm that i just my a what i'm try say here is that there's a lot of ce slippery standards that this start in to come into general use but pot of your research pos process should be to a value ict what are the standards of via little for the sorts of.
ha sp's that you're planning to do and trite to adopt the standard so that others can reuse your researched later on or reuse the models that you produce later on.
um i guess back on linguistic production is i ther end this blow posts pte sta mentions of dial of by droll simon and we're talking out ak a guest's semantic spices and meaning alignments eye and what is done is use collaborative machine learning he call the collab or he but it was a variant of eid the sary ol machine learning which is to.
a sheen models one produce sheen example lls one critiquing examples and they buy ft kind of learn from h. other in this competitive whey i'm hayes core that collaborative rather than competitive a exactly how is implemented or i'm knock shoal ppb att what he's done his amuse is networks to crate i'm urgent language arcelor pts whicher these small vector.
on our of me a pick torre representation zz of what they zz woods main and his started to say in language osce a let's that simile a concepts a producing similar images so they grounded in alexa khan the an english lexicon but they don't have the semantic all english stick context of just caught the lex acquire nis starting to bring out even with these.
each generic method sss i'm similarities in the would form represent tay shuns that are produced.
there is a he nother.
another technique another near on network model that walk that's called hierarchical temper row memory and it's a company that i haven't written on this sly could new men top i'm but they've bilk they har arc ical temporal mm memory and the i o m near ole network model by ste on nearest.
science of third struck sz as of the near core ticks of the human br eyeing so they been looking at the way that than your on's a structure it in to call ofs for ched the call core tickle call oem zz and there's seven lie isn't these core tickle call ums and they've built this neural model to represent the way that in for my shun flows across and between these neural columns in.
and the near core ticks the human near cortex which is a slot lee more i'd vance ste way of representing neurons here i what they've end that up doing is they've i'm pull what i sold this technology to company co cortical dot oya o which is produced what it's called some antic thinker prints which is again another represent taishan ove the.
the sme attic spices of these word say can see or hear that example the got were mice's that the think a print has a bit of a cluster down on the side here rat does as well but they both got a cluster the top core memo was and so few doing bip level comparisons you start in to get a different rep it's not quite have ect a.
arson but it's a pick turrell bit level represent taishan ff of the kinds of ste for magic meaning changers an the sme antic relationships with woods of a town.
so it as lot shove sz man tick main ing changes i'm words i'm a lot of these i'd don't know the meaning of mice fell ff but i do know that these are the sought serve language interpretations that we just a struggling to get a gras spawn with computational and which ticks soeur are hamann him zz and translations for.
and pl.
arm and there's an example in gem n e where air was a bit of a maim where they started to print post cods and so much in the ice but a under the idea of i think i spider which is a literal tex translation so urfi i that's big german that if a trans like the words direct lay it's as i believe i spider that actually it's mm.
more of a color ck rhyl ism that you know the the linguistic would translation was quite funny to nade you speak his but complete they lost on of the cultures.
ago few more examples here i'm my english ease all wonder pig ch the gns your ing a she's really bad so are it fd yeah had had he represent those sorts of meaning changes with suh magic frames ball with the cortical learning i'll bei or them's it's the stew kind of grasping at these looks of things i'm this is an exe.
sample of karua i'm from robot home gr out called sma can mirrors what they've done with fished is take a but kavya larry of words from a previous re such project which were dame ped to represent vagueness in english anne's they a sen sz alee did would frequency analysis cross core pour a of so i think's.
and so mm literature of but some speeches from up utan to you night a oh and from speeches from nato to putin and they plotted the vagueness over time so die chronically they conch stay ease core pour op and they noticed that over time putin's language of creeley more mall of a and nato's language as getting more more cancri.
right but hell you interpret that i guess is related similarly to the x. a saw as we do earlier which is at that word frequency level here so what they've done is due would frequency in ol asus to see that the lexicon of vague terms is increasing over the speech time i'm but again that's just a the word low will there's nour kind of sin tactic.
understanding that's going on there.
i then at the are kind a graphic level we start thing to ceased with things happen on lined so in the same were a show adjusts nw or mend font representation of meaning change this came up on twitter i'm a few months ago or ths who why is the safe but no mark as of dick selver fuzz eh nw to ba the picture a vending machine so papal who don't know what fla pedia ce pts are starting to lose what it and what this i.
icahn mains as floppy disks and so we've got this so shoe interpretation occurring because our obsolete technology is disappearing from the mines of society and sir or their nother tricky thing tug rev on to.
n n even trickier one would be double speak so this says on curran econo mutely in active jiu to being ofa it and early were time an opportunity as a result of mutt previous employ is human rees a lse is redundant see elimination initially.
which she's basically ick ein on it a money plebe caused company was far in people and i got fired so are first you sentences mean the saying thing i have a you would deconstruct gen represent those meaning gns in order to detect the same thing in different texts bees a pretty complicate or problem an.
first raising problems back.
but what's happening with all of these layers of attraction in the evolution of these techniques and meth as and technologies is that we start in to used combined methods and so we start with them graham frequencies for girth through to occur locations your the the likelihood of coaker a nts as that occurring were go where in betting zz arse from twenty six taints earth incas a two.
t. fifteen aaliyah i.
and then some of theirs kind of hm near core tick or grid sell algorithms much be other methods to use for an alternate represent tie shun to combine i'm and this what.
this example is called dis in tangling a trinity as another presentation from the des jeux of hugh mary's contra praa nz in you tracked a few months ago or and that we using most three methods here.
two and lawyers an crunch the dutch use paper corpus i'm to look for network relationships and main ing relationships between the words mulder nw a-t. civilization and europe here unser with h or ver is diaye connick segments they've plotter dh out how each of these words related to h. parts of the.
pay preserve a time and so that was a combination of these three different methods that a used i'm so urdd yet i guess.
the more methods that you can produce shun the more i open you can be about haj our you're using know is in your interpretation the that are sown.
an hour pos an to jane.
a.
a.
it's md act.
just kind how debt you'd han her k. a anu histoire in ste are very very interested in concepts owes a taw i ain it's that that ten perel aspects which is is something that that we spend a lot of time thinking about there is even believe it are not a society for the history of caen sets which has a very big international.
friends every year so a it it's a big tale at and say it's oh he the tyne and the meanings of was in khan sets can changed and their lots of differ in fact it's working alongside each other pte effect this and say difference ocean and cultural contacts pte pts and within i'm between different languages a some of that matti mm was just talking about that.
that's the kind of work that they ce groupies dealing with they these very difficult a multi lng erg rule contacts pts and it's not a are the sort of extended t. hundred yea period of time that weeping talking about earlier this morn he but and language changes dairy very rapid lee nw eh ol a mm partly because of social media in the exposure to lots a different context in cultures dh.
there was the a wonderful project and funded by the dick ing into data scheme a few years ago which ould looked ab the spread oath new words on social media and found that the very interesting national oh and specific lee cultural variations into how word spread in north americ are aware they cane from and which ones will.
was excess for the which ones wha n't i a m dh and it's particularly difficult to i'm pih coal of their ce because ste when a new meaning emerges you don't lose the old one they comb rig zz a haste for long nd perriers of time in you may and i'll with those meanings ole weights ca existing or some of the may stott to four will aaj out of yeas ihg ch and the.
it the rate it which that happens varies why aled laid and it maybe a generational thing nd they example we going to use a bit later on his saic which people my or a dj which still never he use to mean ger it but younger people in the uk rid use thaw up to mean something that's really get a say a a sears some.
https never get they ce new meaning and others use that will the time and stop sully's the ope meanings.
so it's much easier to identify neil ijaz them's there is new words the thirst to carra nts of the term than ace to trace those shifts in meaning between words oh you've time a there is at ever really good project at the yoon of recce chipped law sky called the historical thesaurus of english which is been looking.
but mutha ping these changes than identifying the points at which meaning star sits to shift i and they found of very in varied complicated picture this ear it's for the words there is lie mentioned an it finds forty sever and different meanings and usage is if thought term with a in its core purse.
and waialae t. differing some of them are to do you with mental health he she would expect physical health around in phiri or a-t. a md and jacking under ston day ing some of their is in meanings that wouldn't mean anything to oth any mu all because stay php in lost their he star ical meaning but all of these.
sista long side each other come pluck a sing what ever an owl us this at a text that your going to do a a erm he and in study the people who might the most yi use if they ce rizzo orser actually he store cool novelists be hears they want to write using words that have appropriate for the period that that's that ing their research in stowe.
so you a i and this is kind of broken al to the universe t. space innes become something they gets use much more wind lee in popular culture i he us very interesting the even nd and people writing fiction a starting to get interested in howl meanings change and chide to make sure that the diallo qt at they using is a peru prix att for the period that they working in.
i and another project at the allen cheering ince to sheet in the uk with you saying the web bakalli for at the uk web ow kaif and again they started out being in trusted and tracking changes of caen sets oh you've a time using that his star cool ft the the sorest as their way to trine identify this.
and gee might see from the description of the project up there that they a parm dunned that approach fairly quick clay because it was just tier you complicated to do even with a fifteen year period if day cer so they start to to look at puh leh are see a raul the of than getting into those morse suck tulle difference ers zz so i were out his and negative all a positive concepts role the than how whe huh.
this they squirts changed where is it big new used what context this it a pairing in.
a some of you looked at they ste early air it's the end graham view at that six ever the uk web bol kaif and they says to show really that we had all that can complicated mass than on difficult pts in ta up writ i'm gr all fs earlier when we win a king at the and grams but this is sparely straight forward an its to do you with tracing the first and car.
urdd ths of a term in they score pursed and a colleague it the push ally brie jason web burke came up with am steampunk as the grid example that would be that would work well for data from this period it was thirst i'm cloy endz by the site stiction writer am his pte sightings jeter i connor base first night in nineteen eighty seven so it's a very.
he knew were and it a peers thes in ninety ninety six which is the first date that we've got in our koper ce hair an naira ony stick steen instance as off were pts ch in more than eight hundred an sixty six thousand rei cords sary were mm and all of thurs p much ole of their is our on was kate of alee gita that was his name.
or my stall of to their ease are on universe t. websites so are there are sixteen examples at this new turman their repairing on university websites i many she get through to the lace peer it it really start to take often t. thousand and seven an peaks to woods the and of or piera to she'd expect is its starts to become really mainstream and fiction and midi are in people he easing it mo common.
late a mmmm so that you can thi that some things interesting this started to and to people so i camp yul or ease that base point and it became really popeil are at this point what u. conn pts a is there all only sixteen occurrence as of their sin nineteen ninety fall ife nineteen a key six will you know i'm their ce there's more going ole nad but you can.
the sense of a trend gns but you car really rely on the fake us at and maj ing a.
a and supports this twat none wh or con read o.e.s. this is that were md saic that i rush showing you a only air with its forty seven differ meaning thes a ameen you can see that their look you're ation zinn a how often it peers that you have atb salute lee narrow idea air which versions of fat were ge you're looking at here.
seo it that's really not telling you anything a toll of the vande this word to peers wuffli this numb burst times but you can't meet anything meaningful into that linger stickley.
i i'm e.u. can start tick dig down a little bit more we didn't took about they surly a but is a way of penn ish ratings some a fat end mean graham obscuring of wh war scaring on that you can identify parts of spay ch a bite these are the different pas a speech that google ang graham viewer will let you look cats oh i ticked tis i'd verbs prior nouns.
and so on again you during a hell they ft of so i'd id that this is an eye git to www or a per noun but vt still it get she bei some potter the way to getting to that complex tea as language and it's a it's west mention hey that something and most of your probably use to excluding stop words pte for men at any kind of tice tune and alice us that your dearing first some he mn i see.
for said a ched the stop poor exe might be exactly want you knew want to look kept to identify he ste the orth earth a particular taxed because it maybe somebody that that uses the particular basic gr i'm a formulation lot so a choice that you make to strict stop wh words out because slot simplifies the processes my actually complicate the work of or search or who's interested him.
per psych slee that asked but diff the text so orde of those to cision zz really stopped ply into this so a ish few i've the pots a speech call if a cation tugay got and graham www year mm this is school the word to tack ole which is the example they you gay of its being a very straightforward example to disambiguate so you got tekel is a verb unaids tackle.
is a noun.
but tackle as their furred of course doesn't just have one meaning it can mean tackling from buddy a in a football match or it can mean tackling a jol pp on i'm sure there are verus other in tepper taishan zz of the hot to flour say even the example it's given it's very simple way a if getting three samad its linguistic complexity really is in dearing quite da its.
i i mean this i dear about using in apart prix at language for the pier it that you interested in studying a something it's been picked up the by at lot over searches this is an american academic coupe ben schmidt to hear you has spends a loss of time studying the scripts fa television perry grams to see how many.
many and akron isms that are in then how a many if you note the and the british television pray graham downton abyei be yat now could its us the pt iv early twentieth century periods drama and he's gone to all of the scripts the the stary zz that come or ing the years nineteen twell ft and i nw teen twenty wam and.
sound thirty four phrases that would not have been used at the time the been used in the scrip pts at some of these are wonderful am things that you think sal md very period specific stahl ce lug chun which is a very pts kind of formal old fashions anding fr eighth but that was not in yea ce up the tyne are things like dress fittings.
of waugh time marriage ch a want grams children realistic prospects there's that will things the dippin used by the orth as that no abid ia the time its fall ers we know you from our analysis if digitized and can tempering material was ever of yea eased a and you ca is on to find an other twenty six for aces that.
which if been pos a play he used in the nineteen tens but i knee really rarely a and them much much mall common now so again night that sss disproportion eight yeas pron eardley perry a dh and there's of things like like lee out come or spittle costs pts ol fl emits over ruled charge the basics they ce kinds of things.
so there i think this is a really interesting way a of what hmm tension humiliating script try to ce for not having dumb ever search praa palade mm but it it's its and you know and interesting a prey pts to using these meth its to tell us something about the way people communique sz it and how we should think about and rei percent that a nd un's that the final exam.
pl of or search using the source of math urdd zz is and english lit to chip profess a in the uk called more pte eenie he he www who's done some amazing work all nother novel cloud atlas by the orth a david mit show which is also mate into a thill mm he's just published a perk ah matt which she course close reading with computed is which is providing.
using these quanta to tiff meth it's that are often ease to the could a really large cool pursed to looking great d. telus tingle taxed which he would also be able to work with kuala to to ff lady but china i dents fi what quanta to if meth its can get you when you're working with a much smaller taxed and and this particular and not ol has i'm so des multiple.
oh fiction oh carried to his country be sing different chap to ste tear it and one of varies is written in and the style of a knowing teen ths century by ogre fate so that was what he want to to lick cat how a pert pitt is the language that the orth or his used here to reconstructs they ste a power in pts nineteenth century metairie a lier and it tense out that it's very very good in the.
the whole tapped are other any ther ie words that would know have been used during the period which is from eighteen fifty once and eyeing teen tad and they i all ce spill h. which dates for nineteen thirty fall ih t. night who which its from nineteen forty saic's and lazy are oit which wasn't stelae to to be used until nineteen sixty's said the orth as i.
annie got lat wrong three times php but i think more interesting from air he manatee sp a specht ith is that nd the oath the his only re used racist in colonial terms compared to how off i'm they were to been used in the nineteenth century so the author is high don't id a send sss of literature from that.
that pirie a it as you thing lots of cologna ill empire a racist relais to ter mm zz and is used that a a's and they are appropriate for the a h bay he's use the mm fall more than can temper he writes its would if done so there are that's an interesting different way if looking at its not jest about turn the carra nz but about the tight per.
if turman help often you using them and what that means to studying language it's well i so nd sit its research is being yn ce to a huge scale but oast i to get different insights into much smaller texts and to complement calder to to for set ched processes as well and again that me ving but quiz and forwards between the two yade.
is something that we deer in des jeux of he manatees a lot it's the all so there's a just some examples of how the techniques that marty biz to a king a bow i'm we ste try to get to that complexity that butts an vall www dh hear the pos they witty for the to last fit moves you a curse surer i bore we want to do was builds.
on what we do to early on using the him graham your again ono the wood lot of problems with it but this time we want to introduce sz word senses so or if you go to hage t. don't i a c don't uk it's the his historical thesaurus yet and who.
what we let you to do is kind of form groups different groups from last time if you can please in we want you to kind a mingle with each other.
and pick some woods look them up in the is stark with the sore us look at the senses of the use of the would over time and then try to find trend gns to those affects in the half he trust and gran view off how he hear it he.
they're your it the is ers that's the nut wins i yet infect that probably is a necessary point affront peris an.
it does at make sense.
yen you you ok it so out this morning we looked for trends in the happy trust end graham view on yet now what we'd like you to do is pick a couple of woods looked the mop and they hissed is stark of the source look at the scents says of the wooden hell have changed in the the source of a.
khaw i'm and then try to find those trends in the wood frequencies you you not i'd it in ridding and i.t. tear aqel which an at one point with applied fry much t women ce health and and women's mental health problems him particular and then maeve dh mo broadly into mainstream.
aim psych kyat tray and then us oath there has this meaning as something being he star icky funny say we're and there is a varied and tyne bang owned id schiff's than the meaning if that turn said their be other at other words that you're aware oaf ck or he search of things in see how off i'm they might you've changed all had they pts varied lol arched schiff's i'm in meaning and see if you can actually i des nd.
fide up from the tools and made available for this kind of in al assess.
a shoe give you got tem fifty minutes i'm rearrange tables yet mean go with people you haven't mingled with be fall you and will come around and talk with you as as a goes on here yet there you hulled it.
Loading...
Feedback

Timings

 1328 ms - page object

Version

AV-Portal 3.19.2 (70adb5fbc8bbcafb435210ef7d62ffee973cf172)
hidden