A study on the role of language intensity in a corpus of Dutch-English online hotel reviews.

Evaluations in online reviews written in different languages
of 37
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Evaluations in online reviews written in different languages  A study on the role of language intensity in a corpus of Dutch-English online hotel reviews.  Nele Gehling s4119533Milou Jansen s4239407Intercultural Research EnglishE.B. e Groot ! Grou" 3 Ra#ou $ni%ersit& Ni'(egen 1  Introduction )nline re%ie*s are in%alua#le to an& sel+,res"ecting co("an& or institute. -he& o not onl& allo* an organiation to +ollo* custo(er e/"eriences *ith its ser%ices or "roucts #ut also ena#le the organiation to tae ti(el& an targete (easures *hen its re"utation is at stae. s a result o+ consu(er,generate content share on online #ran co((unities #logs an co("arison sites organiations no longer ha%e +ull control o%er their i(age an re"utation. -oa& consu(ers largel& eter(ine an organiations re"utation.Research sho*s that co("arison sites an electronic custo(er re%ie*s e/ert a signi+icant i("act on consu(er #eha%ior Bonner  6ille(sen 2012. 6hile +or centuries traitional *or,o+,(outh has #een (ost in+luential in regulating consu(ers o"inions an "urchase  #eha%ior Nielsen 2009 clai(s that electronic *or,o+,(outh e6)M is a close secon in in+luential "o*er. Inee so(e scholars argue that e6)M o++ers aitional a%antages that cannot #e "ro%ie #& traitional 6)M. 8ue to the s"ee ease o+ use i((eiate an *ie accessi#ilit& o+ e6)M an ue to the a#sence o+ +ace,to,+ace co((unication the a%antageso+ e6)M (a& *ell out*eigh those o+ traitional 6)M eso%ec a(ic  :u(er(an 2007.;onse<uentl& it is crucial +or an organiation to closel& (onitor its re"utation in e6)M. =houl its re"utation #e negati%e an the organiation is a*are o+ this it can still tr& to e("lo& ae<uate *e# care or other re(eial actions. Not sur"risingl& se%eral so+t*are  "rogra(s so,calle #usiness intelligence tools ha%e no* #een e%elo"e that ena#le an organiation to (anage its re"utation online. ""l&ing senti(ent anal&sis or o"inion (ining these so+t*are "rogra(s auto(aticall& collect the ata that are rele%ant +or (a""ing the co("an& i(age >ang  ee 200?. =enti(ent anal&sis e/a(ines the *a& language users%er#all& e/"ress their +eelings an o"inions. In "articular senti(ent anal&sis +ocuses on the use o+ e%aluati%e language an on the %alence o+ the e%aluati%e language use i.e. "ositi%e or negati%e. =o+t*are "rogra(s a""l&ing senti(ent anal&ses to ate ha%e #een #ase on English ata e/clusi%el&. In +act in case these so+t*are "rogra(s alrea& allo* +or the "rocessing o+ ata in other languages the& onl& o"erate #ase on an auto(atic translation (oule that is 2  co("ara#le to Google -ranslate or Ba#el+ish. ccoringl& intercultural i++erences in the %alence an *oring o+ e%aluations are not at all accounte +or in the so+t*are "rogra(s c+. =hanahan Gre+enstette @u an E%ans 2004 an 8enee 200? #ut also see =chul 6o(ser,:acer an Manl 2010 +or a i++erent a""roach. Aurther(ore the (a'orit& o+ the so+t*are "rogra(s inclue %er& li(ite lists o+ e%aluati%e *ors *ith a set o+ no (ore than 50 ite(s ranging +ro( #a goo e/cellent (agni+icent etc. 6hen these *ors occur in a re%ie* the "rogra( e/tra"olates that the re(aining te/t in the re%ie* is (are #& the sa(e %alence or "olarit& >ang ee  Caith&anathan 2002. >ut i++erentl& onl& the *ors that clearl& articulate an o"inion in English *ill #e translate into another language an #ase on that translation conclusions *ill #e ra*n *ith regar to the +ull e%aluation re"resente in the te/t. -hus the "rogra( oes not tae into consieration the istincti%e *ors conce"ts inter"retations an "erce"tions that (a& signi+icantl& in+luence the %er#al e/"ressions in an e++ects o+ a i++erent language in a i++erent culture. It see(s that this t&"e o+ research co("letel& ignores the <uestion o+ *hether cultural i++erences (a& cause i++erent ieas a#out ho* to %er#all& e/"ress an inter"ret e%aluations. -he "resent stu& *ill aress this research ga"D it *ill #e the +irst stu& to in%estigate *hether the e%aluation o+ a gi%en ser%ice in one language anor cultural area is co("ara#le to an e%aluation o+ the sa(e ser%ice in another languageculture. More s"eci+icall& it *ill e/a(ine *hether language intensit& in relation to "olarit& has the sa(e iscursi%e +unction across English *ithin e%aluati%e hotel re%ie*s on the -ri"a%isor.co(.co( *e#site. s such this stu& sees to contri#ute to the e%elo"(ent o+ (ultilingual o"inion (ining so+t*are. Its results *ill lea to the +or(ulation o+ guielines that are i("ortant *hen using this t&"e o+ so+t*are +or (ulti"le language "airs hereF NE -hese guielines *ill re<uire the a""lication o+ algorith(s that account +or the t&"ical i++erences #et*een languageculture "airs. In other *ors is the 8utch "hrase #est een "ri(a hotellet'e reall& co("ara#le to Google -ranslates English translation <uite a goo little hotel In 8utch #est is the su"erlati%e o+ goo #ut *hen use as an a%er# it has a (itigating (eaning i.e. as a hege. -he sa(e hols +or "ri(a *hich in collo<uial language use re+ers to (eanings such as goe te oen )H an %alt (ee not too #a rather than e/cellent. lso the i(inuti%e in hotel,let'e has a "ositi%e connotation an oes not onl& escri#e the sie o+ the hotel. In short these *ors re<uire user inter"retations that e%iate +ro( the (eanings suggeste in ictionaries or  #& their +ace %alue. -hrough co(#ining se%eral (itigating ele(ents the 8utch te/t "roucer  3  thus a""ears to <uali+& the gi%en hotel as (eiocre or al(ost satis+actor&. 6ors that loo liesu"erlati%es (oreo%er actuall& see( to contri#ute to a (oerate re%ie* o+ the hotel. et this nee not al*a&s #e the case. Man& 8utch "eo"le "re+er to use unerstate(ents in their e%aluationsF the& e/"ress their "ositi%e 'ug(ents #& (eans o+ (itigating *ors. =uch t&"icaluse o+ language can #e regare as a characteristic o+ the 8utch culture. Not #a *oul #e the highest "ossi#le co("li(ent in this culture n so the "hrase #est een "ri(a hotellet'e coul *ell (ean that the 8utch *riter is actuall& reall& enthusiastic a#out the gi%en hotel.In short si("l& aing the ictionar& (eaning o+ ini%iual *ors use in an e%aluati%e re%ie* ! i.e. as current so+t*are "rogra(s ten to o ! "oses se%eral ra*#acs. Airst *or (eaning or %alue can onl& #e eter(ine in,conte/t an secon through the cultural inter"retation o+ this conte/t. ;learl& it is unliel& that co("uter "rogra(s *ill #e a#le to recognie the su#tleties in language use an translate the( in a si("le algorith( that onl& (a"s *ors *ith "ositi%e an negati%e %alence. n i+ the (a""ing o+ these su#tleties +or a single 8utch re%ie* is i++icult alrea& then certainl& it *ill also #e i++icult +or re%ie*s in the other languages. anguage an culture o not nee to coincie. )n the contrar& English is s"oen #& a larger nu(#er o+ non,nati%e than nati%e s"eaers an not all nati%e s"eaers srcinate in the British or (erican culture. =ince :o+stee 2001 culture has #een e+ine as the collecti%e  "rogra((ing o+ the (in share a(ong (e(#ers o+ a "articular grou". 8ra*ing on this e+inition it is "lausi#le that the collecti%e "rogra((ing o+ the (in o+ 8utch "eo"le e%iates +ro( the collecti%e "rogra((ing o+ the (in o+ Belgians or Ger(ans or ="aniars or ;anaians see ;laes an Gerritsen 2012. -his cultural istincti%eness is liel& to #e re+lecte in the *a& "eo"le co((unicate. Most "ro#a#l& i++erences in cultural nor(s an %alues are also re+lecte in the *a& e%aluations are %er#all& e/"resse. Gi%en the i++erences :o+stee 2001 an -ro("enaars an :a("en,-urner 1997 in "articular +oun #et*een the 8utch an British cultures it can #e h&"othesie that the *oring an "erce"tions o+ e%aluations *ill also %ar& across these cultures. Gu&unst an -ingtoo(e& 19?? *ere the +irst to tr& to relate a grou"s %er#al st&le to the grou"s cultural i(ensions as outline in the *or o+ :o+stee 19?1D 2001 an :all 1972e.g. ini%iualis(collecti%is( or conte/t sensiti%it&. Gu&unst an -ingtoo(e& 19?? 4   "oint out that 6esterners +or instance "re+er a irect an o"en co((unication st&le *hereas sians lie an inirect an o"a<ue st&le #etter. Moreo%er 6esterners ha%e #een o#ser%e to  #e (ore ini%iualistic than sians. ccoringl& Gu&unst an -ingtoo(e& 19?? inicate that a "re+erence +or a irect co((unication st&le is consistent *ith a "re+erence +or an ini%iualistic cultural orientation. lthough the authors o not (ae e/"licit clai(s a#out causalit& the& o suggest that i+ a culture is characterie #& "re+erences +or s"eci+ic nor(s an %alues it also +eatures a corres"oning "re+erence +or a s"eci+ic %er#al st&le. In line *ith their suggestion the authors relate +our i++erent %er#al st&les inirectirectD succinctela#orateD instru(entala++ecti%eD "ersonalconte/tual to the %arious cultural i(ensions o+ :o+stee 19?1D 2001 an :all 1972. :o*e%er the& o not "ro%ie irect e("irical e%ience +or the relationshi" #et*een %er#al st&le an cultural orientation. ;onsiering the co("le/it& o+ the genre o+ the "rouct re%ie* *hich assu(es that the *riter %er#all& e/"resses hisher "ersonal 'ug(ent o+ a "rouct or ser%ice an *hich is "u#licl& rea an consulte it (a& not #e eas& to connect this te/t t&"e to the %er#al st&les that ha%e  #een istinguishe #& Gu&unst an -ingtoo(e& 19??. Because hotel re%ie*s ai( to +or(ulate 'ug(ents an "ri(aril& entail e%aluations the language the& inclue can #e e+ine as e%aluati%e. Cer#al st&les that are t&"ical +or e%aluati%e language use are not co%ere in Gu&unst an -ingtoo(e& 19??s *or. In +act the& i("l& that %er#al st&les are t&"ical o+ the general use o+ language in a "articular culture regarless o+ *hether o#'ecti%e or su#'ecti%e language use is concerne. =i(ilar to Gu&unst an -ingtoo(e& 19?? Johnson Hulesa lc ;ho en =ha%itt 2005 argue that there is a relationshi" #et*een res"onse st&le an :o+stees i(ensions 19?1D2001. Research has "ro%en that so(e sur%e& "artici"ants ha%e a stronger tenenc& than others to select the e/tre(es o+ an ans*er scale. -his is also no*n as the e/tre(e res"onse st&le. Johnson et al. 2005 sho* that it is "ossi#le to correlate a "re+erence +or the e/tre(e res"onse st&le *ith high scores on :o+stees 2001 i(ensions o+ (asculinit& an  "o*er istance. -he (ore a "erson tens to rel& on the e/tre(es o+ a scale the greater the chance that heshe re"resents a culture that is highl& (asculine an in +a%or o+ high "o*er istance. Naturall& the *a& "eo"le res"on in a sur%e& an the "re+erence the& ha%e +or scale e/tre(es is not ientical to an acti%e res"onse st&le or the st&le "eo"le a""l& *hen *riting a hotel re%ie*. Ne%ertheless it can #e e/"ecte #ase on Johnson et al. 2005 that cultures characterie #& high scores on the "o*er istance an (asculinit& i(ensions *ill also use 5
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks