Vaš brskalnik ne omogoča JavaScript!
JavaScript je nujen za pravilno delovanje teh spletnih strani. Omogočite JavaScript ali pa uporabite sodobnejši brskalnik.
Nacionalni portal odprte znanosti
Odprta znanost
DiKUL
slv
|
eng
Iskanje
Brskanje
Novo v RUL
Kaj je RUL
V številkah
Pomoč
Prijava
Analiza odziva potrošnikov na Brexit s pomočjo metode tekstovnega rudarjenja v programu R : delo diplomskega seminarja
ID
Kolenc, Jan
(
Avtor
),
ID
Redek, Tjaša
(
Mentor
)
Več o mentorju...
PDF - Predstavitvena datoteka,
prenos
(3,91 MB)
MD5: F00BD2DF0425189802D9DC1B8FB79A68
Galerija slik
Izvleček
Ozadje. Brexit je eden najpomembnejših dogodkov v zgodovini EU in VB. Gre za izstop Velike Britanije iz EU. Začel se je z referendumom, ki ga je razpisal takratni premier David Cameron. Uradni postopek izstopa je nadaljevala njegova naslednica Theresa May, ki pa je postopek vodila neuspešno in je zato njeno delo prevzel sedanji premier Boris Johnson. Po dogovoru med EU in VB je slednja zapustila zvezo 31. 1. 2020 in začela prehodno obdobje odcepitve od zveze. Cilj. Diplomski seminar analizira s pomočjo analize odzivov bralcev (komentatorjev) novic odnos le-teh do izstopa Velike Britanije iz Evropske unije. Cilj je poiskati odgovore na raziskovalno vprašanje: kakšen je odziv Slovencev (oziroma bralcev) na Brexit in kako se je dojemanje/odnos Slovencev do Brexita spreminjal skozi čas? Metodologija. Analiza temelji na različnih metodah tekstovnega rudarjenja v programu R s pomočjo tabele AFINN. Podatki so pridobljeni s spletnega portala RTV SLO. Uporabljenih je bilo 20470 komentarjev pod 440 objavljenimi članki na to temo. Vsebinsko povezavo komentarjev s temo Brexita sem preveril z latentno Dirichletovo alokacijo, ki je potrdila povezavo. Uporabil sem tudi analizo korelacije med besedami.. Rezultati. Izkazalo se je, da so Slovenci precej nevtralno, čeprav v povprečju rahlo pozitivno naravnani do Brexita in sicer približno +0,8 po metodi merjenja sentimenta s tabelo AFINN. Analiza sprememb sentimenta glede na časovnico pomembnih dogodkov (npr. Predčasne volitve 12.12.2019) kaže, da je pogosto le-ta nižji od povprečja, razen v štirih primerih. Sentiment je bil nadpovprečen na dan napovedi referenduma - 23. 6. 2016, nato na dan, ko je VB zapustila EU - 31. 1. 2020, poleg tega tudi na dan, ko je kabinet prevzel Boris Johnson, ter tudi na dan, ko so bile v VB predčasne volitve na pobudo Mayeve. Sentiment je pri ostalih izbranih dogodkih blizu 0 oz. je ponekod celo negativen. Med najbolj negativne dogodke po oceni Slovencev spadajo »Chequers plan«, dan, ko premierka postane Theresa May, aktivacija člena 50, Johnsonov dogovor za odstop in druge predčasne volitve. Zaključki. Slovenski potrošniki so do Brexita v povprečju precej nevtralni, se pa nekateri skoki v sentimentu pokažejo ob različnih z Brexitom povezanih dogodkih.
Jezik:
Slovenski jezik
Ključne besede:
Brexit
,
Analiza sentimenta
,
Podatkovno rudarjenje
,
R
Vrsta gradiva:
Delo diplomskega seminarja/zaključno seminarsko delo/naloga
Tipologija:
2.11 - Diplomsko delo
Organizacija:
FMF - Fakulteta za matematiko in fiziko
Leto izida:
2020
PID:
20.500.12556/RUL-117390
UDK:
004.4
COBISS.SI-ID:
58663683
Datum objave v RUL:
09.07.2020
Število ogledov:
889
Število prenosov:
260
Metapodatki:
Citiraj gradivo
Navadno besedilo
BibTeX
EndNote XML
EndNote/Refer
RIS
ABNT
ACM Ref
AMA
APA
Chicago 17th Author-Date
Harvard
IEEE
ISO 690
MLA
Vancouver
:
Kopiraj citat
Objavi na:
Sekundarni jezik
Jezik:
Angleški jezik
Naslov:
Sentiment analysis on consumers regarding Brexit by data mining methodology in R
Izvleček:
Background. Brexit is one of the key moments in the history of EU and UK. Brexit stands for departure of UK from the EU. The departure went into motion as former PM David Cameron announced a referendum. Formally Brexit began in the eyes of the EU when Theresa May invoked Article 50 of the Treaty on European Union. As May was unable to provide a good enough exit strategy, she was replaced by the current PM Boris Johnson. After a deal was struck between EU and UK the latter left the Union on 31. 1. 2020 and began the transition period leading to full departure in the future. Aim. The bachelor thesis addresses the problem of sentiment analysis on Slovenian consumers regarding Brexit. Its aim is to find answers to the research question: what is Slovenes response to Brexit and how did it change through time? Method. All of the following results are acquired by data mining methodology in R. Data was mined from the web portal of RTV SLO and processed by AFINN sentiment table. Slovene consumers were regularly responding on the web portal of RTV SLO regarding Brexit. I managed to acquire 20470 comments from 440 published articles on Brexit. Slovenes reacted to monumental events regarding Brexit in a way that increased the rate of comments on those dates. Reader’s comments on RTV SLO were a topic well related to Brexit. The latter was confirmed by latent Dirichlet allocation. More credibility and confirmation for the first statement was also provided by the correlation analysis in comments. Results. It turned out that Slovenes were in general indifferent about Brexit but in average slightly positive towards it. The positivity was measured to be +0.8 by sentiment analysis with the AFINN table. Sentiment analysis in time of important events provided with the following results. Sentiment on the dates chosen for analysis was lower than average in all but 4 cases. Sentiment was above average on the referendum announcement day 23. 6. 2016, on the day of UK's departure from EU 31.1. 2020, on the day when Boris Johnson took cabinet and when the first snap election took place. Sentiment on the dates of other chosen events was around 0 or in some cases also negative. For Slovenes the most negative events were the »Chequers plan« announcement day, the day Theresa May took office, Article 50 invocation, Johnson’s departure deal and the second snap election. Conclusions. To conclude, the Slovene consumers’ disposition is averagely positive towards Brexit but with some sentiment dynamics in relation to major events affecting Brexit.
Ključne besede:
Brexit
,
Sentiment analysis
,
Data mining
,
R
Podobna dela
Podobna dela v RUL:
Podobna dela v drugih slovenskih zbirkah:
Nazaj