Skripsi

PENERAPAN KNOWLEDGE-BASED TEXT SIMILARITY UNTUK MENINGKATKAN AKURASI KLASIFIKASI KESESUAIAN PENULIS PADA DATA BIBLIOGRAFI

---

Penilaian

---

In research using data in the form of text, a problem theme can be drawn called Author Name Disambiguation (AND). The theme can include a case of synonymy and polysemy that occurs when classifying authors in bibliographic data. This research will then use Deep Neural Network (DNN) as a classifier used to classify authors. Before the classification process is carried out, the data must enter the pre-processing stage. For feature extraction, the level of similarity will be calculated after combining data using cosine similarity for the author name, author list, and venue attributes as well as absolute reduction for the year attribute. Specifically for the title attribute, the knowledge-based text similarity method will be used to calculate the level of similarity which is carried out using two approaches and wordnet as the ontology. Furthermore, the results will be evaluated using performance measurement as a reference for its success. Accuracy obtained reached 99% for both knowledge-based text similarity approaches, namely path and wu palmer.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2107002207	T52713	T527132021	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T527132021

Publisher

Inderalaya : Fakultas Ilmu komputer, Prodi Sistem Komputer., 2021

Collation

xiv, 62 hlm,: ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

004.07

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Komputer
Pemrosesan Data-Ilmu Komputer

Specific Detail Info

-

Statement of Responsibility

MURZ

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment