| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

Show document Help

Title:SISTEM PRIPOROČANJA DOKUMENTOV IN ANALIZA KVALITETE VSEBINSKEGA PRIPOROČANJA PRI RAZLIČNIH OBDELAVAH VHODNEGA BESEDILA
Authors:ID Borovič, Mladen (Author)
ID Ojsteršek, Milan (Mentor) More about this mentor... New window
Files:.pdf MAG_Borovic_Mladen_2012.pdf (2,13 MB)
MD5: 73320B56ACD2D02B773C1841C4CEEF8B
PID: 20.500.12556/dkum/87c83065-537e-4500-ade4-bedf363d639d
 
Language:Slovenian
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FERI - Faculty of Electrical Engineering and Computer Science
Abstract:V magistrskem delu obravnavamo načrtovanje in razvoj sistemov za vsebinsko priporočanje pomensko sorodnih dokumentov. V teoretičnem delu najprej podamo osnovne značilnosti priporočilnih sistemov. Ker v priporočilnem sistemu obdelujemo besedila v slovenskem jeziku, najprej podamo pregled nekaterih obdelav vhodnega besedila (lematizacija, izločanje pogostih besed in pomensko označevanje). Nato opišemo funkcijo razvrščanja BM25 in pristop z latentno semantično analizo. Sledi podroben opis razvoja priporočilnega sistema, ki je tudi praktični izdelek tega magistrskega dela. V nadaljevanju predstavimo in analiziramo uspešnost vsebinskega priporočanja pri različnih obdelavah vhodnega besedila. Na koncu podamo še nekaj potencialnih izboljšav v smislu pomenskega gručenja, klasifikacije, hibridnega pristopa pri razvrščanju dokumentov in uporabe razvitega sistema v drugih aplikacijah.
Keywords:priporočilni sistemi, vsebinsko priporočanje dokumentov, funkcija razvrščanja BM25, latentna semantična analiza, beleženje uporabniških aktivnosti, pomensko označevanje, statistične metode, procesiranje naravnega jezika, jezikovne tehnologije
Place of publishing:Maribor
Publisher:[M. Borovič]
Year of publishing:2012
PID:20.500.12556/DKUM-37811 New window
UDC:004.777:004.93(043.2)
COBISS.SI-ID:16586262 New window
NUK URN:URN:SI:UM:DK:BIR68NK6
Publication date in DKUM:30.11.2012
Views:3299
Downloads:281
Metadata:XML DC-XML DC-RDF
Categories:KTFMB - FERI
:
Copy citation
  
Average score:(3 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share


Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Secondary language

Language:English
Title:A DOCUMENT RECOMMENDATION SYSTEM AND CONTENT-BASED RECOMMENDATION QUALITY ANALYSIS WITH DIFFERENT TEXT PROCESSING METHODS APPLIED
Abstract:In this thesis we discuss the planning and development of content-based recommender systems for semantically related documents. The theoretic part describes the basic features of recommender systems. Due to processing of Slovene language, we provide an overview in the field of text processing methods (lemmatisation, stop word lists and semantic tagging). Then we describe the ranking function BM25 and the latent semantic analysis method. Further on, we describe in detail, the development of a content-based recommender system for documents, which is also the practical result of this thesis. We then analyse the success rate of the recommendation with different text processing methods applied. Finally, we describe some of the potential enhancements such as semantic clustering, user classification, a hybrid approach in ranking documents as well as using the developed system with other applications.
Keywords:recommender systems, content-based document recommendation, ranking function BM25, latent semantic analysis, user-action tracking, semantic tagging, statistical methods, natural language processing, language technologies


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica