Pretražite po imenu i prezimenu autora, mentora, urednika, prevoditelja

Napredna pretraga

Pregled bibliografske jedinice broj: 1005329

Performance of Common Classifiers on node2vec Network Representations


Pozek, Mislav; Sikic, Lucija; Afric, Petar; Kurdija, Adrian Satja; Klemo, Vladimir; Delac, Goran; Silic, Marin
Performance of Common Classifiers on node2vec Network Representations // Proceedings of the International Conference on Computers in Technical Systems MIPRO 2019 Opatija
Opatija, Hrvatska, 2019. str. 1071-1076 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)


CROSBI ID: 1005329 Za ispravke kontaktirajte CROSBI podršku putem web obrasca

Naslov
Performance of Common Classifiers on node2vec Network Representations

Autori
Pozek, Mislav ; Sikic, Lucija ; Afric, Petar ; Kurdija, Adrian Satja ; Klemo, Vladimir ; Delac, Goran ; Silic, Marin

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the International Conference on Computers in Technical Systems MIPRO 2019 Opatija / - , 2019, 1071-1076

Skup
42nd International ICT Convention MIPRO 2019

Mjesto i datum
Opatija, Hrvatska, 20-24.05.2019

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
node2vec ; classification

Sažetak
In this paper we evaluate the performance of different multi-class classifiers on network graphs. Since the node embedding techniques have been widely used to represent and analyze networks structures, we decide to transform network data (nodes and links) into attributes which are descriptive and contain correct information of its structure. For this purpose, we use a state-of- the-art algorithmic framework node2vec, which has been shown to outperform other popular methods when applied to multi-label classification as it manages to efficiently learn a mapping of nodes to a low-dimensional space of features. Applying this framework, we generate a set of representations for nodes of multiple network data sets. Using the generated representations, we evaluate the performance of common classifiers. We perform cross-validation and parameter tuning to get the best possible model of each classifier type. To compare their performance, we computed Precision, Recall and F1-score for each model on each data set. Following that, the obtained results are analyzed and compared.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo



POVEZANOST RADA


Projekti:
HRZZ-IP-2018-01-6423 - Pouzdani kompozitni primjenski sustavi zasnovani na web uslugama (RELS) (Srbljić, Siniša, HRZZ - 2018-01) ( POIROT)

Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb


Citiraj ovu publikaciju:

Pozek, Mislav; Sikic, Lucija; Afric, Petar; Kurdija, Adrian Satja; Klemo, Vladimir; Delac, Goran; Silic, Marin
Performance of Common Classifiers on node2vec Network Representations // Proceedings of the International Conference on Computers in Technical Systems MIPRO 2019 Opatija
Opatija, Hrvatska, 2019. str. 1071-1076 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
Pozek, M., Sikic, L., Afric, P., Kurdija, A., Klemo, V., Delac, G. & Silic, M. (2019) Performance of Common Classifiers on node2vec Network Representations. U: Proceedings of the International Conference on Computers in Technical Systems MIPRO 2019 Opatija.
@article{article, year = {2019}, pages = {1071-1076}, keywords = {node2vec, classification}, title = {Performance of Common Classifiers on node2vec Network Representations}, keyword = {node2vec, classification}, publisherplace = {Opatija, Hrvatska} }
@article{article, year = {2019}, pages = {1071-1076}, keywords = {node2vec, classification}, title = {Performance of Common Classifiers on node2vec Network Representations}, keyword = {node2vec, classification}, publisherplace = {Opatija, Hrvatska} }




Contrast
Increase Font
Decrease Font
Dyslexic Font