izpis_h1_title_alt

Avtomatska ekstrakcija podatkov o zaposlenih s spletišč podjetij
ID Koplan, Matej (Author), ID Žitnik, Slavko (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (3,86 MB)
MD5: CC58ED01445CC5C3BC7A9C26732FEE53

Abstract
V tem delu se ukvarjamo s problemom ekstrakcije seznama oseb s poljubnega spletišča. V ta namen implementiramo spletnega pajka za identifikacijo potencialnih podstrani z osebami in ekstraktor podatkov, ki s poljubne spletne strani izvleče podatke o osebah. Pokažemo, da osnovne metode, kot so primerjava imena s seznamom imen, ne dosežejo sprejemljive natančnosti. Pokažemo, da je analiza strukture seznama in prenos odkritega znanja ključna metoda za izboljšavo rezultatov do stopnje, kjer dosežemo sprejemljiv nivo natančnosti. S pomočjo tega pristopa smo izboljšali F1 mero za 50 % na razvojni in za 35 % na skriti testni množici.

Language:Slovenian
Keywords:splet, ekstrakcija podatkov, avtomatska ekstrakcija podatkov s spleta, fokusirani spletni pajki, strukturirani podatki, nestrukturirani podatki
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FRI - Faculty of Computer and Information Science
Year:2021
PID:20.500.12556/RUL-132403 This link opens in a new window
COBISS.SI-ID:83603971 This link opens in a new window
Publication date in RUL:25.10.2021
Views:602
Downloads:92
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Automatic extraction of employee data from corporate websites
Abstract:
In this work we tackle the problem of extracting lists of people from corporate websites. For this purpose we implement a web crawler to identify possible subpages with people and a data extractor, which is designed to work on any website. We show that basic methods, such as matching names from a list, don't reach acceptable accuracy. We show that analysing the structure and transfrering the discovered knowledge of a list is crucial in reaching the required level of accuracy. Using this approach we have improved the score of our final results by 50 % in the development and by 35 % in the hidden test set.

Keywords:web, data extraction, automatic web data extraction, focused webcrawlers, structured data, unstructured data

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back