KLASIFIKASI HOMONIM-SINONIM PADA BIBLIOGRAFI DIGITAL LIBRARY DENGAN MENGGUNAKAN METODE MACHINE LEARNING
Total View This Week0
Total View This Week0
Institusion
Universitas Sriwijaya
Author
VARINDO OCKTA KENEDDI PUTRA (STUDENT ID : 09011181520010)
Firdaus Firdaus (LECTURER ID : 0221017801)
Firdaus Firdaus (LECTURER ID : 0221017801)
Subject
T1-995 Technology (General)
Datestamp
2020-07-10 04:52:45
Abstract :
The process of finding an author's name for a publication becomes more complicated because of the case of Homonyms and Synonyms. Although many researchers in the world have classified Homonym data or Synonym data from a Digital Library Bibliography dataset, in this study the classification is done at once for Homonym data, Synonym data, Homonym-Synonym data, and Non Homonym-Synonym data. The classification in this study was carried out using three classifiers, namely Support Vector Machine (SVM), Decision Tree, and Deep Neural Network (DNN). With the DBLP dataset and the three Classifiers, the classification is done in twenty one trials (model / scenario). From all experiments, each classifier has a best model / scenario that is judged by the good performance value generated. Among the sixteen models / scenario scenarios conducted using DNN, there are two best models / scenarios that produce the highest accuracy values. These two models / scenarios are the best models / scenarios for all experiments of all Classifiers.
The process of finding an author's name for a publication becomes more complicated because of the case of Homonyms and Synonyms. Although many researchers in the world have classified Homonym data or Synonym data from a Digital Library Bibliography dataset, in this study the classification is done at once for Homonym data, Synonym data, Homonym-Synonym data, and Non Homonym-Synonym data. The classification in this study was carried out using three classifiers, namely Support Vector Machine (SVM), Decision Tree, and Deep Neural Network (DNN). With the DBLP dataset and the three Classifiers, the classification is done in twenty one trials (model / scenario). From all experiments, each classifier has a best model / scenario that is judged by the good performance value generated. Among the sixteen models / scenario scenarios conducted using DNN, there are two best models / scenarios that produce the highest accuracy values. These two models / scenarios are the best models / scenarios for all experiments of all Classifiers.
File :
RAMA_56201_09011181520010.pdf
RAMA_56201_09011181520010_TURNITIN.pdf
RAMA_56201_09011181520010_0221017801_01_front_ref.pdf
RAMA_56201_09011181520010_0221017801_02.pdf
RAMA_56201_09011181520010_0221017801_03.pdf
RAMA_56201_09011181520010_0221017801_04.pdf
RAMA_56201_09011181520010_0221017801_05.pdf
RAMA_56201_09011181520010_0221017801_06_ref.pdf
RAMA_56201_09011181520010_0221017801_07_Lamp.pdf
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
