Analiza i predviđanje ocjenjivanih filmova

master's thesis

• Title (english): Analysis and Prediction of Movie Preferences
• Author: Leon Fadljević
• Mentor: Mihaela Vranić (mentor)
• Committee member: Mihaela Vranić (predsjednik povjerenstva)
• Committee member: Damir Pintar (član povjerenstva)
• Committee member: Mirko Randić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-09-20, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Duboke neuronske mreže predstavljaju moćan alat koji postiže izvrsne rezultate na teškim zadatcima učenja poput prepoznavanja slika, prepoznavanja govora i obrade prirodnog jezika. Ovim se radom ispituje njihova sposobnost savladavanja izazova poput regresije, klasifikacije i mapiranja jednog uzorka vremenske sekvence u drugi (tzv. seq2seq problem) nad podatkovnim skupom MovieLens. Trenirani model čini bazu sustava preporuke koji na temelju ulaznih vrijednosti za određenog korisnika uspješno preporučuje najboljih $k$ stavki koje bi mogle biti u području interesa dotičnog korisnika.

Abstract (english)

Deep Neural Networks (DNNs) are powerful tools that have achieved outstanding performance on difficult learning tasks like image recognition, speech recognition and natural language processing. This work examines capability of DNNs to overcome challenges like regression, classification and mapping a sequence to another sequence (so called seq2seq problem) over MovieLens dataset. Trained model makes the basis of the recommender system which predicts, i.e. recommends top "k" next items that could be in field of interest for observed user based on given (input) values.

• Keywords: Duboko učenje; Strojno učenje; Duboke neuronske mreže; LSTM; Regresija; Klasifikacija; seq2seq; Sustavi preporuke; Apache Solr; Anaconda; Java; Python; MovieLens; DL4J razvojna cjelina
• Keywords (english): Deep learning; Machine learning; Deep neural networks; LSTM; Regression; Classification; seq2seq; Recommender systems; Recommendation; Apache Solr; Anaconda; Java; Python; MovieLens; DL4J framework
• Language: english
• URN:NBN: urn:nbn:hr:168:768645
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 20:10:56