prikaz prve stranice dokumenta Prepoznavanje neželjenih SMS poruka
  Download
PDF 738.48 KB
undergraduate thesis
Prepoznavanje neželjenih SMS poruka
2018. urn:nbn:hr:200:908716
Tubić, Eugen
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Software Engineering
Chair of Programming Languages and Systems

Cite this document

Tubić, E. (2018). Prepoznavanje neželjenih SMS poruka (Undergraduate thesis). Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:908716

Tubić, Eugen. "Prepoznavanje neželjenih SMS poruka." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2018. https://urn.nsk.hr/urn:nbn:hr:200:908716

Tubić, Eugen. "Prepoznavanje neželjenih SMS poruka." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2018. https://urn.nsk.hr/urn:nbn:hr:200:908716

Tubić, E. (2018). 'Prepoznavanje neželjenih SMS poruka', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 04 August 2021, https://urn.nsk.hr/urn:nbn:hr:200:908716

Tubić E. Prepoznavanje neželjenih SMS poruka [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2018 [cited 2021 August 04] Available at: https://urn.nsk.hr/urn:nbn:hr:200:908716

E. Tubić, "Prepoznavanje neželjenih SMS poruka", Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2018. Available at: https://urn.nsk.hr/urn:nbn:hr:200:908716

Islandora Bookmark - Please login to use this feature
Metadata
TitlePrepoznavanje neželjenih SMS poruka
Title (english)Identifying of unwanted text messages
AuthorEugen Tubić
MentorGoran Martinović (mentor)
MentorBruno Zorić (mentor)
Committee memberGoran Martinović (predsjednik povjerenstva)
Committee memberZdravko Krpić (član povjerenstva)
Committee memberDean Vučinić (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Software Engineering)
(Chair of Programming Languages and Systems)
Osijek
Defense date and country2018-10-01, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Program Engineering
Abstract
U ovome radu objašnjen je problem neželjenih SMS poruka i prikazan je program koji koristi pet klasifikatora za prepoznavanje neželjenih i pravih SMS poruka. Za klasificiranje poruka koristili su se sljedeći klasifikatori: kNN, SVM, naivni Bayes, šuma odluke i umjetne neuronske mreže. U teorijskom dijelu rada opisana je problematika neželjenih poruka i nekoliko metoda kojima se takva problematika rješava. U praktičnom dijelu rada osmišljen je program koji klasificira SMS poruke pomoću pet klasifikatora koristeći opisne značajke poruka. Podaci u skupu se prvo dijele na trening i testni skup poruka u kojima postoji približno jednak omjer između pravih i neželjenih SMS poruka, a nakon toga se svaka poruka obrađuje na temelju zadanih opisnih značajki. U sljedećem koraku se program istrenira na trening skupu koje su poruke prave a koje neželjene i onda na testnom skupu pokušava prepoznati klase poruka. Klasifikator šuma odluke se pokazao kao najtočniji klasifikator pri prepoznavanju poruka.
Abstract (english)
This thesis researches the problem of spam and it presents a program which uses five classifiers in order to decipher between spam and real messages. The following classifiers were used to classify text messages: kNN, SVM, Naive Bayes, Random Forest and artificial neural netowork.The theoretical part of the thesis describes the problem of spam and several methods used to solve this problem. For the practical part, a program which classifies text messages with regard to their descriptive attributes was created. The program does this by the means of five classifiers.The data in the data set are first divided into the training and testing categories of messages where there is an approximately equal number of real and spam messages, and afterwards each SMS is processed on the basis of predefined descriptive attributes. In the next step, the program learns, on the basis of a training set, which messages are real and which are spam, and then it tries to recognize the classes text messages belong to in a testing set.The research conducted showed the random forest classifier to have been the most accurate one in deciphering between different messages.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:908716
Study programmeTitle: Undergraduate University Study Programme in Computer Engineering
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2018-10-22 08:30:07