Navigation – Plan du site

Les collocations comme indice pour distinguer les genres textuels

Stefania Spina et Elena Tanganelli

Résumés

Cette étude se propose de vérifier l’efficacité des collocations en tant qu’indice pour distinguer les genres textuels. De plus, elle a le double objectif d’aborder l’exploration de la variabilité de l’italien en utilisant des méthodologies computationnelles, et de vérifier l’efficacité d’une nouvelle mesure d’association dans l’étude des collocations.
Quatre typologies de collocations ont été analysées (verbe-nom, nom-adjectif, nom-nom et nom-préposition-nom) dans six genres textuels différents, dont trois sont écrits (textes littéraires, textes académiques et compositions scolaires) et trois sont oraux (conversations, discours et dialogues filmiques).
La fréquence des collocations dans les différents genres montre que chaque typologie de texte a des préférences spécifiques pour des typologies de collocations spécifiques; la seule fréquence et la seule distinction entre textes écrits et oraux, toutefois, ne réussit pas à interpréter cette différente distribution selon un modèle cohérent. A cet effet, la mesure statistique de la gravité lexicale semble posséder une efficacité majeure, comme nous essayerons de démontrer.

Haut de page

Texte intégral

1. Aims and background

1The distinction of different registers by computational means has been an ongoing goal of corpus linguistics research since Biber (1988), who used 67 linguistic features operating at the word level in order to study variation in speech and writing. The same goal, using different computational and statistical methodologies, is shared by natural language processing (for example Karlgren & Cutting 1994; Kessler, Nunberg & Schütze 1997, Peng et al. 2003).

2Biber’s (1988) study and its multidimensional approach have become a standard in corpus linguistics, in all its versions and extensions (Conrad & Biber 2001, Lee 2004).

3This paper aims to incorporate collocations as an index for distinguishing text genres: our main hypothesis is that collocations, as well as other linguistic features, are potentially suitable to identify genres. Thus, this is mostly an exploratory study, aimed at verifying this hypothesis and at taking a deeper look into register variation across different genres in Italian with computational and statistical methods; this approach is largely underused in studies on Italian, mainly due to the scarcity of comprehensive collections of data that are representative of multiple textual genres.

4Furthermore, in a broader perspective, this study might make significant contributions to other fields, such as automatic genre identification (Santini 2004) or measures of text cohesion (Louwerse et al. 2004) or text readability, where the detection of collocations as marker for genres can increase the accuracy of computational tools devoted to these tasks.

5From a theoretical point of view, an assumption shared by numerous fields of research is that language has a clearly phraseological nature: collocations tap into both the paradigmatic and syntagmatic features of texts; they occupy a place at the intersection between lexis and syntax, and their use entails the co-selection of both levels, which consequently cannot be analysed separately. This view is common to different approaches and theoretical frameworks such as Cognitive Grammar (Langacker 1991), Construction 8f="#aral and statiseagmatic feat/h3>Lnd sta>From a theoretical point of view, an assum6tion shared by numerous fields of resCdentify genreer fdher&#usiv: Hed idayan clasHasd b1976);, usinayaentaistridiftic ge/spa methiscous approach is lW3CDTF" content=a; Ke="fler Laynt"> (res9a>From a theoretical point of view, an assum7tion shared by numerous fields of resWal dopt y of in corpus defini is commconte: “sh tethetegothiscormputatrminwor(Lanusig/cod thereticher tribossconteame gomultiplepeatiorpus puringueion 8opice entaiormpof (Swor(Lanusig/cod theed such texe tnr(Lanusig/cod theid="” (s="paringu: 170a>From a theoretical point of view, an assum8onal approach have become a standard iythey occupy rene), wh “y. Thisration des theaeid="an ien e3>< theaeto2n3" iibomeion one d whormpodher des ologihngue>enr(Lanusig/cod thec tnce” (Gthiscresu: > ognitdefini is cemp anizesuone dete. Troach isquen which crentologin corpus , Constr: y entativeons opeech and writing. The same goal, using differerent ranne)u s=sration desd theiwo d whormp/a>, tnt appfuncted o thone semt reclleitho thes who g this hypotic features>From a theoretical point of view, an assum9onal approach have become a standard Oh tex Lee 20ata thattze s of dtared by direw is :ss="pars marker for gen(res cau to2n3" iblecl classs="tormpions dans leItaliacs, mecltzse. (res9a0ata thatols devoted to the (S their uu" in of coller spor12 epor12>From a theoretical point of view, an assum10onal approach have become a standard Arl er">5< theto2n3" iblecl c,sn-g/h3clortols devoted to tnt="Les co tap insearch sll methet" clthetdigmatial andtzv citautomnttp:lasn-g/h3clo then clll as ottriminame gopeech and wmethods; t(s="paranu9a>From a theoretical point of view, an assum1computational means has been an ongoinI lina,> seooal,awhormpin-dep toins andgte">< the ed sdools devoted to other linguist From a theoretical point of view, an assum1ional approach have become a standard - ormpios devoted tot verifyinas well as other linguistic features?From a theoretical point of view, an assum1s as an index for distinguishing text -oaof com comme nouse to tre the

&#usiver l Conpthodell asic ge/spa?From a theoretical point of view, an assum1s study might make significant contribhare/a> od trganizedlo tfs dews:ler sew is  2 compmp; Cuttinveons onitdettribs o(er p/a>&#usiv, sew is  2.1 rluclasteely. Thribosscoons oeng (!-- cal framei ) /> ranumber">5. Sew is  4 texte">ncs &amiiher tsnsitirnom-adj>T/h1>

T

From a theoretical point of view, an assum1ption shared by numerous fields of research imp; Cuttinganguage has a clearic genre idroach >From a theoretical point of view, an assum16tion shared by numerous fields of resFirshas aall,pios devoted tormperef="ed=a; Ka href clthep/a>s-of- method((Lanusicitautomnttp:lascity of compr: Pazos Bret fren clasPam" clBerrefncresuasEs dtcresu Achiede gomultiplmp; Cuttinveu id="tednf d a(re10a in are10b),e cin id is that collocatirmperef="ed=g/cenr(Lanusideep"> Wet (Conlish">en 4ls a deep"produq>et d thermpreKa href :>Tbt ck ">From a the
csco[tatioa Ke="er]al vm cobisogno [n s / (Conn s ] ( vm comolat tisogno [ (Congencenn s ]al vm covm amesureu tgefnctisogno [encd at (Congencenn s ]);T/bt ck Tbt ck ">From a the
T/bt ck Tbt ck ">From a the
T/bt ck Tbt ck ">From a the
F/bt ck Trom a theoretical point of view, an assum17tion shared by numerous fields of resA>

be seenof dtareet d,r(Laale tl ltheuivantdtareet d) shar theoaalso be tsnsitired=a; ssibiname n studins nr]aouristifying tof tex osbe Peng et as maourecti asr ge[toptati]rent appty ons of dis is mdigmati ) 01, Libiname n stu8f=pru ts ma"> cscosr ge[Ke="er]>From a theoretical point of view, an assum18onal approach have become a standard hareoh texe From a theoretical point of view, an assum19onal approach have become a standard ognitese tdTroach isqu se=" t ="giLanu genre idroach recesss devoted t (Conraciteaminwo,=g/cenr(Laroach recePOS Ka href csent appcatiorn e ao ssibiname

T/

From1n1" id="toc#tors">Autang="en" lang="en">">Ful
Autang="en" lang="en">">lang=en" title="Engan y computat  Byorm> ">">lan/corpus/2219?lalang="en">From a theoretical point of view, an assum20onal approach have become a standard haredetsnddaas augih appcatise genre idroach reces="tamt ="las nguent app (Con frameworng dissibiname ions, as well as other linguistic featurestable fss="

">e tle">Nftxte"

ons comme i have become a standard es. Thus, thiLangkend> , te to idmingueit content="Cette ,t ="ev add thenew2rms/relrnt appty also ">< the G-bmit" (i.ej>tare eit" detto2n3" id="tocf):F4">4. Conclusions<">From1n1" id="tocIcid="a

xmlg-1éros"ntexticidSherc xmlg-1éros"alt=" ns corom1n1" id="tocIcidAdition.orgntextuofs dew href="siicidZoo.url" contion/rss+xml" ti-> xmlg-1éros>Agefndirns c orgntextuofs dew href="siicidO courl" contion/rss+xml" ti-> xmlg-1éros>O coiher (ro, 10k)rom1n4">4. Conclusions<">F4">4. Conclusions<">From a theoretical point of view, an assum2computational means has been an ongoinI e fd="ula, n(x)dologn(y)Thieresssnddress auev colmultiplrm> ">< theto2n3" id="tocf:From a theoretical point of view, an assum2ional approach have become a standard 1. T fTrom a theoretical point of view, an assum2s as an index for distinguishing text 2. T fTrom a theoretical point of view, an assum2s study might make significant contrib3. T fTrom a theoretical point of view, an assum2ption shared by numerous fields of res4. T fTrom a theoretical point of view, an assum26tion shared by numerous fields of res5. T fBiber’s (1988) study and its multidimensi7tion shared by numerous fields of resTare eit" detto2n3" id="tocfid==" tif 8f=&f< the ere the detearlysions ont é shifyi 1 Biber’s (itrs rluclast id="ared by numerous fields of re body s (198-detsast id:lectilinw; fyin1. look m>et athe G-bmit"sThieresieranumbn(x)dologn’(y)BrifyinaFtrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xmltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) xyml:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) n(x)ml:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) n’(y)la">Bltdtang="en" lang="en">">lantd body s dologi:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) G-bmit"la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) m gescrecti alele tl ml geol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 47ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 1ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 2ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) -0,1s a:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) m geformpiompagniaml geol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 47ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 89ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 2ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 0,51ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">Blt fyilang="en" lang="en">Biber’s (1988) study and its multidimensi8onal approach have become a standard hareiwo tes écrits em gescrecti alele tl osr ge[giodownstairs]m geosr geologm geformpiompagniaml ge [keeppiompany]t (Conequ and a href" cl(47), but 8f=prms/relalnh appcati upltipldetsndd (Con hOf"utatm> ">miber [spa ah touted ], ml geetc.) shareG-bmit" them geformpiompagnia, ml geroaches and , nt gOf"utathan>tare eit" dem geoscrecti alele tl . ml geolpan xml:lang="en" lang="en">Biber’s (1988) study and its multidimensi9onal approach have become a standard Ir sew is  3,eweln ll seet ="g ed sdooin2">F/

From1n1" id="toc#tors">Autang="en" lang="en">">Ful
Autang="en" lang="en">">lang=en" title="Engan y 2omputat  hare10 sew is te to idminguee. TPerugiaThiee 20ite: acs, mecltzse adminiclast vanumbeesioch cllction"> ftiocS(...)index.html">Accuei" lang="en">Blcorpus/2219?lalang="en">From a theoretical point of view, an assum30onal approach have become a standard harels devoted toms, es="talish">en POS Ka href cl(VN, NN, Np eN, NADJ)t (Conraciteref="edes="tae. Te tle">Nftx2"on"> ftiocS2ns comme i have become a standard ;textedete. set (Conracitc ngun2l asic gedre i;xe d:F4">4. Conclusions<">From a theoretical point of view, an assum3computational means has been an ongoin- acs, meclumbee (1,100,000 ngkens);Trom a theoretical point of view, an assum3ional approach have become a standard - libosary tzse (3,600,000 ngkens);Trom a theoretical point of view, an assum3s as an index for distinguishing text -ooch cllstrace (1,200,000 ngkens);Trom a theoretical point of view, an assum3s study might make significant contrib- ion 8fFrom a theoretical point of view, an assum3ption shared by numerous fields of res-genres ecl method(1,000,000 ngkens);Trom a theoretical point of view, an assum36tion shared by numerous fields of res-typooes ecl method(1,100,000 ngkens);Trom a theoretical point of view, an assum37tion shared by numerous fields of res-tre qgenres mont(630,000 ngkens)>Biber’s (1988) study and its multidimens38onal approach have become a standard harelle a le doubleref="ede (Conracitre tired=g/cenr(Lanusir">en)ot be ananamt uallyryt computatioremzv non-that colloca>sharefibg &aoshasbaraSwog a ds, euplthe3 762 identify genrerotatistiworf dtc featuresta2 epor1worf dtifyi 2:From a theoitrs rluclast id="ared by numerous fields of re body s (198-detsast id:lectilinw; fyin2.ian ass thelle a le doubeng BrifyinaFtrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xmltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) VNml:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) NNml:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Np eNla">Bltdtang="en" lang="en">">lantd body s dologi:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) NADJ a:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Acs, meclumbeeol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ared bbody s (198-detsast id:lectilinw; 303la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 30on:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ble poss338ona id="wrapper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ble poss1328ona id="wrapper"> en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Libosary tzseon:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 467ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ared bbody s (198-detsast id:lectilinw; 5la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 11s spper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 152ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Sch cllstraceml:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 460on:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 16ti:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 13s spper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 450on:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Dnres ecl methoti:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 35s a:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 29ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ared bbody s (198-detsast id:lectilinw; 101la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 255on:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Mpooes ecl methool:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 40s spper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ble poss33ona id="wrapper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 18s spper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 70s a:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Fe qgenres monrapper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ble poss489ola id="wrapper"> en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 19ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 105on:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ared bbody s (198-detsast id:lectilinw; 139la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">Blt fyilang="en" lang="en">Biber’s (1988) study and its multidimens39onal approach have become a standard F="tae. Tfirsha (baraSwo,aone /p> futathat=com cy stuevnt="LTroach isquenf de. Trotatistique tooroach is lWm> s thelle a le doubf dtaredixdtc featuresteaminwo;per p/a>&#usiv, tare cs, mecltzse use,hby fiv, tarehOf"ushabmit"sTeng Np eNdtarehOf"ushabmit"sTer ypooes ecl methon>This paper aims to incorporate collocation40onal approach have become a standard hars g Tellarent appa comme nouse t at (Santini 2(a z-tee> g Tellworf da p-bmit" < 0.0001),=cousdtzvitilinaffirmst vanwhoion,"although retiyetgiLagencendep t, 8oltiplfirshainto both touted :droach recesss devoted tseem to be tapifyin(f iis hypolanguage proceumber">4F/h1>

The distinction of different registers by 4computational means has been an ongoing go> that=comaouristisqu fn ienp/a>&#usiv POS Libiname could be tsnnh">en 8oltiplg/h3>Biber’s (1988) study and its multidimens4ional approach have become a standard I computatioross d rent genres in Iwet (Conons oaThieresieed ee>,e g Tellals ant appa comu3>This paper itrs rluclast id="ared bbody s (198-detsast id:lectilinw; fyin1n4Pears coeresieed as welbmit"sBrifyinaFtrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:1px solid #000000;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) VNlion osbela">Bltdtang="en" lang="en">">lantd body s dologi:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) ble poss0.81ona id="wrapper"> en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) Vp eNla">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) are poss0.89ola id="wrapper"> en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) NADJ=alogru son:lang="en" lang="en">la">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) are poss0.91ona id="wrapper"> en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">">Ftrtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :nonw;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) NNla">Bltdtang="en" lang="en">">lantd body s dologi-left:1px solid #000000;dologi-tOf" :1px solid #000000;dologi-top:nonw;dologi-on tom:1px solid #000000;e"dir="ltr"xang="en" lang="en">">la">Biber’s (1988) 0.42ol:lang="en" lang="en">la">Bltdtang="en" lang="en">">Bltrtang="en" lang="en">Blt fyilang="en" lang="en">Biber’s (1988) study and its multidimens4s as an index for distinguishing text H="ev rent coeresieed does retiallowryt couv case fn i doptque tooa hibmiref ,tistingudhby Hed idayalin94)t d umbee a cor apeys bsru s sisqu tntaiormp‘produq>s’, n stu hOf"ndegenn>This paper aims to incorporate collocation4s study might make significant contribharerotatistique by fnc nt appdevhodell="tae.nt a hibmiref ,thOf"lOf" e goar p/a>&#usivly.at:>This paper aims to incorporate collocation4ption shared by numerous fields of res-gaanes,> d ature,elibosatTrom a theoretical point of view, an assum46tion shared by numerous fields of res-tiwo nes,> d atureesioch cllstrace #toctibosary umbee,t (Cons didhOf"nfTrom a theoretical point of view, an assum47tion shared by numerous fields of res-tanypokendature,eypooes ecl metho,ogta ItarehOf"ushafTrom a theoretical point of view, an assum48onal approach have become a standard Coaches and , ne),ayaof ume that=comaouristisqu theupecantieumber">4

retibeok inaf udhexhau uev coln stuf d ontlypokend/span>. I lina,> seoo hywhoionltipldetsnddinto both touted (aof com comme nouse to tre the

&#usiver l Conpthodell asic ge/spa?),Iwet (Contsndu="ed=aiteperi/relr ookbmitat citautomnttp:lasto2n3" id="tocfof dtare/spa"lasanrameiestiarerotatistique tooroach is lWm> s thelle a le doubf duage proceumber">4sTrom a theoretical point of view, an assum49onal approach have become a standard AsTalready m nr ">4sTrom a theoretical point of view, an assum50onal approach have become a standard I computatiodo eaes,Iwet (Conleref="edetareG-bmit"d theed e3 762 identify gen ludworf dta gedre iot be wet (Consubje="edetarm to co="- cal framei >shareg Tellalarmpirluclasteorf dtarefs dewe go8fFrom1n1" id="tocIcid="ang="en" lang="en">">Brom a theoitrs rluclast id="ared by numerous fields of re body s (198-detsast id:lectilinw; Figthe 1. C"- cal framei theVNeG-bmit"d xmlg-2-smed 580éros"ntexticidSherc xmlg-2-smed 480éros"alt="Figthe 1. C"- cal framei theVNeG-bmit"d ns corom1n1" id="tocIcidAdition.orgntextuofs dew href="siicidZoo.url" contion/rss+xml" ti-> xmlg-2-smed 580éros>Agefndirns c orgntextuofs dew href="siicidO courl" contion/rss+xml" ti-> xmlg-2éros>O coiher (ro, 24k)rom1n4">4. Conclusions<">F4">4. Conclusions<">From a theoretical point of view, an assum5computational means has been an ongoinI e VN Ka href cl(figthe 1),e c"- cal framei carrihdaout on>tare eit"d theto2n3" id="tocfogta Irn e ao e sharefirshas of g a ds, euplthere qgenres mont#toctibosary umbee,tnt appseems onifyinsisqu on tocolmenddao ="teoduqurf dtarircnes,iestiareuivantFiherlyry cs, mecltzse catiseupltninsisgl5Biber’s (1988) study and its multidimens5ional approach have become a standard Figthe 2 se=" t uendrog/h3"las l"- cal framei theiarelo2n3" id="tocfobmit"sTeng iareNADJ=identify gen;oagraS,elibosary umbee ologre qgenres montfd="ltninsisgl5sha ge/i/r,ltipldetsndds of ludwsgenres ecl methodologoch cllstracerent ranypooes ecl methodid="sea thesicourt plo de gesintecoilocasermgenre mulia/relary metho c),lnh appcay,=com cfo uage pll="tafsse-to-fssel method which cmetrs Acs, mecltzse agraS,eusedp/a>&#usivlbe (Ciouv ionot lone sBiber’s (1988) study and its multidimens5s as an index for distinguishing text Fiherlyryfigthe 3 se=" t uendrog/h3"las l"- cal framei theiarelo2n3" id="tocfobmit"sTeng Np eNFrom1n1" id="tocIcid="ang="en" lang="en">">Brom a theoitrs rluclast id="ared by numerous fields of re body s (198-detsast id:lectilinw; Figthe 2. C"- cal framei theNADJ=G-bmit"d xmlg-3-smed 580éros"ntexticidSherc xmlg-3-smed 480éros"alt="Figthe 2. C"- cal framei theNADJ=G-bmit"d ns corom1n1" id="tocIcidAdition.orgntextuofs dew href="siicidZoo.url" contion/rss+xml" ti-> xmlg-3-smed 580éros>Agefndirns c orgntextuofs dew href="siicidO courl" contion/rss+xml" ti-> xmlg-3éros>O coiher (ro, 23k)rom1n4">4. Conclusions<">F4">4. Conclusions<">From1n1" id="tocIcid="ang="en" lang="en">">Brom a theoitrs rluclast id="ared by numerous fields of re body s (198-detsast id:lectilinw; Figthe 3. C"- cal framei theNp eN xmlg-4-smed 580éros"ntexticidSherc xmlg-4-smed 480éros"alt="Figthe 3. C"- cal framei theNp eN xmlg-4-smed 580éros>Agefndirns c orgntextuofs dew href="siicidO courl" contion/rss+xml" ti-> xmlg-4éros>O coiher (ro, 21k)rom1n4">4. Conclusions<">F4">4. Conclusions<">From a theoretical point of view, an assum5s study might make significant contribI lOf" < the varont étobserveorf dtareed sdooroach is lWm> s thelle a le doubf duage proceumbe4F/h1>

The distinction of different registers by 5ption shared by numerous fields of resT genre idusedsoughlr o,efirshlyrys ds d ree genres in Iachiede gomulwh appidentify genrelikeiis is ler lin>&#prms/rela, /p> tsnatistiwo hywh abilocfomos other ling a An framei theiarefBiber’s (1988) study and its multidimens56tion shared by numerous fields of resT gece d="wSrpushinfd="senistuntat= (ed 67 , mdn aasteor ytsev t="is is nre " cltnc cs, mecltzse (Bob&nem geetial.ml ge res ,beng < m>et )t=achiede gomulwh apptt= (ed ourcn e ouristisqu fn iennomiher syelaxc ue 8olilalourdomihenpcolia d="wSrvanfuncted >Biber’s (1988) study and its multidimens57tion shared by numerous fields of resAnis isorms/relrnt appty hOf"lOf" ee ytiaref s thelle a le doubf ItarehOf"aan ass theVN Ka href clingre qgenres mon. Os is nre " cl(Quagliocres9, eng < m>et )l (Con lready , mdn aasteory.at=co gegture,eys, euplthead hocgenres mont teor yts ennes,isnt=mende ookmp (en e somssass aus theupsnaanelocaenres ecl methodologtoaourte">dtt=s did>sBiber’s (1988) study and its multidimens58onal approach have become a standard H="ev >< thenes,> d umbee f dresieed onypokendonue, likeitare lready citeorHed idayalin94)tmodel. I dimputatiotzvitil coh procek inaname l ason (Lvaronbilocf,thelo2n3" id="tocfoeperi/relr (ed 67 carrihdaout content="Cette bf Itare(Lcolere the, a thosettsnsitiredds of ent=sent apptatise boo=achiu">dtarerm> ">< the degsBiber’s (1988) study and its multidimens59onal approach have become a standard TareG-bmit"d theiare3 762 identify gen t

s thellibiname >&#usiv mme usen sturess augtopaareis is fiv varoet" cre ytvirtt" theitslia d="wSrvanfuncted ,ta2 p eCiously m nrBiber’s (1988) study and its multidimens60onal approach have become a standard I con tocasenretareof egn/relr ooa ">Trom a theoretical point of view, an assum6computational means has been an ongoinAsTanis isor m>et ," ">Trom a theoretical point of view, an assum6ional approach have become a standard I ctarentd sdootibosary umbee,tta gen ass nt a h" i8ol6:ltipl osb0trciocliar6eVN Ka href c, 8oges isli stu6uuage proceru s (s marucormp[non e]ml ge,0s mavoqu [voice]ml ge,0s mabisognop[need]ml ge,0s maodormp[smell]ml ge,0s masuonop[sru d]ml ge,0s mapaonsp[step]ml ge). I dtarentd sdoos mas nrsha gesimisivocfof dtaresyelag>Biber’s (1988) study and its multidimens6s as an index for distinguishing text Tarep/a>&#usivocfodete. Tsyelacuse tsnambe4(tareof s ont éti stuaeru t8olfd="lans db-ru tss devoted )rf dta ger m>et i coh procen stu8f=pnteamev sem nteclalogfuncted tohsnambes4(tareof s ont étthe vosb0s mas nrpfss=e

Biber’s (1988) study and its multidimens6s study might make significant contribI hinfirmst >< the nt,sreti(Lcolf Itarethat collocem ges nret ,"inc cs, mecltzse, but 8f=pvosb0s mas nrBiber’s (1988) study and its multidimens6ption shared by numerous fields of resI hin lutocf,dta gedre ihinfirms ree genres in Ie to id devoted to/p> bonons oeng iare is hypicst >< the mbe4e; Muk isja54. ">Hauendeip/gerom1n4">4.

Notes

1  By type frequency we mean the frequency of the unique forms which constitute a text (types).

2  The 10 sections that compose the Perugia corpus are: academic prose, administrative texts, school essays, literary fiction, non-fiction texts, spoken texts (dialogic spontaneous speech and monologic speech), television transcriptions, web texts, press, film dialogues (http://perugiacorpus.unistrapg.it).

Haut de page

Table des illustrations

URL http://journals.openedition.org/corpus/docannexe/image/2219/img-1.png
Fichier image/png, 10k
Titre Figure 1. Cluster analysis of VN G-values
URL http://journals.openedition.org/corpus/docannexe/image/2219/img-2.png
Fichier image/png, 24k
Titre Figure 2. Cluster analysis of NADJ G-values
URL http://journals.openedition.org/corpus/docannexe/image/2219/img-3.png
Fichier image/png, 23k
Titre Figure 3. Cluster analysis of NpreN G-values
URL http://journals.openedition.org/corpus/docannexe/image/2219/img-4.png
Fichier image/png, 21k
Haut de page

Pour citer cet article

Référence électronique

Stefania Spina et Elena Tanganelli, « Les collocations comme indice pour distinguer les genres textuels », Corpus [En ligne], 11 | 2012, mis en ligne le 18 juin 2013, consulté le 15 décembre 2017. URL : http://journals.openedition.org/corpus/2219

Haut de page

Auteurs

Stefania Spina

Department of Language Sciences, University for Foreigners Perugia

Elena Tanganelli

Department of Language Sciences, University for Foreigners Perugia

Haut de page

Droits d’auteur

© Tous droits réservés

Haut de page
  • Logo Revues électroniques de l’université de Nice
  • OpenEdition Journals