Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Media Informasi Teknologi
Yudha, Dandun Kusuma
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Comparative Analysis Scoring System Auto Essay (simple-o) Based Algoritma Generalized Latent Semantic Analysis (GLSA) Laplacian Eigenmaps Embedding (LEM) and Hybrid Indexing Yudha, Dandun Kusuma; Ratna, Anak Agung Putri
Jurnal Media Informasi Teknologi Vol. 1 No. 2 (2024): Oktober 2024
Publisher : Digital Innovation

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69616/mit.v1i2.190

Abstract

This research discusses a comparison of two algorithms for an automatic essay assessment system (Simple-O), namely generalized latent semantic analysis (GLSA), Laplacian Eigenmaps embedding (LEM) and hybrid indexing. The two algorithms are compared to find out how the two algorithms work, processing speed, and assessment results. Comparison of how it works is done by comparing the pseudocode of each algorithm. The processing speed is calculated to find out a faster algorithm for assessing essays. The GLSA hybrid indexing algorithm is a development of the LEM algorithm. The fundamental difference between the two algorithms is in the treatment of nouns and words other than nouns. This research used a sample of eight questions completed by 48 students (384 data). From the research results, GLSA LEM has a total processing time of 46.51454 seconds, which is faster than GLSA hybrid indexing. Meanwhile, the average processing time for GLSA LEM and GLSA hybrid indexing to assess one answer is 6-6.6 seconds. The assessment results from GLSA LEM and GLSA hybrid indexing have the highest similarity level of 95.83% and the lowest 16.67%. Of the eight questions tested, five of them had a similarity level of more than 83.33%.
Co-Authors Ratna, Anak Agung Putri