Model za kategoriziranje patenata baziran na sadržaju te dubokom učenju

undergraduate thesis

• Title (english): Content-based categorization model for patents based on deep learning
• Author: Lucija Knežević
• Mentor: Mario Brčić (mentor)
• Committee member: Mario Brčić (predsjednik povjerenstva)
• Committee member: Agneza Krajna (član povjerenstva)
• Committee member: Kristijan Poje (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-04, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Patenti su glavno sredstvo zaštite intelektualnog vlasništva te ih je prilikom prijave potrebno svrstati u pripadajuću unaprijed definiranu kategoriju. S obzirom na velik broj patentnih prijava javlja se problem nedovoljnih kapaciteta ljudstva za prihvatom i obradom istih. Kao rješenje se nameće primjena algoritama za kategoriziranje patentnih prijava čija je zadaća odrediti kategoriju pod koju se patent svrstava s obzirom na službenu nomenklaturu izdavača patenata. U ovom je radu razrađena problematika kategoriziranja patenata te je opisan implementiran model temeljen na metodama prirodne obrade jezika (engl. Natural Language Processing, NLP). Ponuđeni NLP model temeljem riječi sažetka opisa patenta određuje kategoriju u koju se isti svrstava. U konačnici, s ciljem zaštite izvornosti, kriptirani se sažetak patentne prijave upisuje na javni blok-lanac Solana.

Abstract (english)

Patents are the primary means of intellectual property protection and must be classified in the pre-defined category. Given many patent applications, a problem of the insufficient human capacity to receive and process them emerges. As a solution, the application of algorithms for categorizing patent applications is imposed. The task is to determine the category under which the patent is classified concerning the official nomenclature of patent issuers. In this paper, the issue of categorizing patents is elaborated on, and the implemented model based on methods of natural language processing (NLP) is described. The offered NLP model, based on the words of the abstract of the patent description, determines the category into which it is assigned. Finally, to protect the authenticity, the encrypted abstract of the patent application is written on the public blockchain Solana.

• Keywords: patent; izvornost podataka; blok-lanac; obrada prirodnog jezika; transformer
• Keywords (english): patent; data provenance; blockchain; natural language processing; transformer
• Language: croatian
• URN:NBN: urn:nbn:hr:168:431544
• 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: 2023-01-15 10:31:18