Program za pretraživanje međunarodne klasifikacije bolesti

undergraduate thesis

• Title (english): Software for searching of international classification of diseases
• Author: Marin Marjanović
• Mentor: Damir Kalpić (mentor)
• Committee member: Damir Kalpić (predsjednik povjerenstva)
• Committee member: Igor Mekterović (član povjerenstva)
• Committee member: Ivan Budišćak (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2016-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Kroz primjenu na stvarnom problemu, proučeni su i implementirani algoritmi za ispravljanje riječi, pretraživanje velike količine podataka i stvaranje sustava koji ima mogućnost učenja. Algoritam za ispravljanje riječi trebalo je prilagoditi specifičnostima hrvatskog jezika koje zbog promjene konteksta mogu značajno promijeniti izgled i duljinu riječi, što je posebno zahtjevno za računalnu obradu. Iako mi automatski poistovjećujemo riječi “pluča”, ”pluca”, “plućima” s riječi ”pluća” za računalo to nije ni malo jednostavan zadatak. Struktura koja je korištena za pretraživanje je neuronska mreža koja se sastoji od više slojeva od kojih se svaki sastoji on mnoštva čvorova (neurona) koji su povezani s čvorovima iz susjednih slojeva. Svaka ta veza ima svoju snagu koja algoritmu pretraživanja omogućava razlučivanje bitnih od manje bitnih rezultata. Te veze i povratne informacije korisnika služe kako bi sustav s vremenom mogao naučiti obrazac pisanja, odnosno izražavanja korisnika te im tako ponuditi samo bitne podatke.

Abstract (english)

Using real life example, algorithm for misspelled word correction, searching through large amount of data and creating system which is capable of learning were studied and implemented. Word correction algorithm needed extra work because of the specifics of Croatian language, where a simple change of context can significantly alter any given word. To Croats, it is clear that misspelled words like “pluča”, “pluca”, “plucima” are to be connected with anything involving lungs, which is English word for “pluća”, but explaining that to a computer is no easy task. Structure that was used for search algorithm is neural network which consists of multiple layers where each layer consists of nodes (neurons) that are connected to other nodes in neighbouring layers. Each connection has its own strength that enables search algorithm differentiating important from less important search results. Those connections are also used in combination with user feedback as a tool that network uses to learn users writing pattern, and often mistakes that they make and to present only the most important results to the user.

• Keywords: ASP.NET; MVC; pretraživanje; tražilica; ispravljanje pogrešaka; neuronska mreža; računalno učenje
• Keywords (english): ASP.NET; MVC; searching; search engine; misspelled word correction; neural network; computer learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:321763
• Study programme: Title: Computing; 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 resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 10:27:04