KLASIFIKASI TEMA MENGGUNAKAN ALGORITMA GENERALIZED VECTOR SPACE MODEL (GVSM) - IMPROVED KNN PADA SOAL UJIAN NASIONAL

Subroto, Nurma Ayu Wigati S. (2019) KLASIFIKASI TEMA MENGGUNAKAN ALGORITMA GENERALIZED VECTOR SPACE MODEL (GVSM) - IMPROVED KNN PADA SOAL UJIAN NASIONAL. Other thesis, Universitas Sebelas Maret.

[img] PDF - Published Version
Download (582Kb)

    Abstract

    The National Examination (UAN or UNAS or UN) is used as a measure of government evaluation to determine the quality of education in Indonesia. The quality of education obtained depends on the success of students in working on national exam questions based on the material in the graduate competency standard (Standar Kompetensi Lulusan or SKL). The National exam questions are grouped into various themes. The theme classification is useful for knowing the class of questions that are in the material of graduate competency standard. This research aims to determine the performance of the Generalized Vector Space Model (GVSM) – improved KNN algorithm in classifying national exam questions based on themes. The GVSM algorithm is used to identify the similarity of words that appear in one document with another document. This improved KNN algorithm classifies national exam questions with analyzing all the words that appear on national exam questions based on subject’s themes. The classification evaluation with the GVSM - improved KNN algorithm got 0,7939 accuracy, 0,7771 precision, and 0,7633 recall on evaluation using 10-fold cross validation.

    Item Type: Thesis (Other)
    Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
    Divisions: Fakultas Matematika dan Ilmu Pengetahuan Alam
    Fakultas Matematika dan Ilmu Pengetahuan Alam > Informatika
    Depositing User: Patasik Irene
    Date Deposited: 25 Jun 2019 16:38
    Last Modified: 25 Jun 2019 16:38
    URI: https://eprints.uns.ac.id/id/eprint/43869

    Actions (login required)

    View Item