diplomsko delo
Miha Debenjak (Author), Tomaž Curk (Mentor)

Abstract

Spremljanje raziskovalnih tematik znanstvenih objav skupine raziskovalcev je zanimivo in hkrati pomembno za razumevanje razvoja nekega znanstvenega področja in raziskovalcev, ki delujejo na področju. Raziskovalci se ukvarjajo z različnimi področji, zato se tudi teme, o katerih pišejo v znanstvenih objavah, razlikujejo. Na podlagi besed, ki so uporabljene v znanstvenih člankih, lahko določimo teme, o katerih raziskovalci razpravljajo. V diplomski nalogi je opisano pridobivanje podatkov o člankih, njihova analiza in modeliranje tem člankov. Izvedena je bila tudi analiza o zastopanosti različnih tem skozi čas, kar nam pove o aktualnosti tem v določenem času. Zgrajen sistem smo uporabili za analizo publikacij Fakultete za računalništvo in informatiko Univerze v Ljubljani.

Keywords

modeliranje tem;model LDA;rudarjenje besedil;vizualizacija;razvoj tematik;obdelava naravnega jezika;računalništvo;računalništvo in informatika;visokošolski strokovni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [M. Debenjak]
UDC: 004.93(043.2)
COBISS: 1538354371 Link will open in a new window
Views: 756
Downloads: 207
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Tracking research topics
Secondary abstract: Following the research topics of the scientific publications of a group of researchers is interesting and at the same time important for understanding the development of a scientific field and researchers working in it. The researchers do not work in the same field, therefore the topics of their work differ. Topics of the articles can be identified on the basis of words used. The thesis describes the acquisition of data on articles, their analysis and modelling of topics that they discuss. In addition, an analysis was conducted on the representation of different topics over time, which shows most frequently discussed topics in certain time periods. This system was used for the analysis of publications of the Faculty of Computer and Information Science of the University of Ljubljana.
Secondary keywords: topic modeling;LDA model;text mining;visualization;topic development;natural language processing;computer science;computer and information science;diploma;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000470
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 27 str.
ID: 11226244