Anam, Moch. Irsyadul (2020) Sistem koreksi jawaban uraian singkat otomatis menggunakan metode Cosine Similarity dan Query Ekspansion. Undergraduate thesis, Universitas Islam Negeri Maulana Malik Ibrahim.
|
Text (fulltext)
16650119.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview |
Abstract
Indonesia:
Penilaian merupakan merupakan hal pokok dalam dunia pendidikan. Melalui penilaian guru dapat mengetahui seberapa efektif pengajaran mereka. Pada umumnya guru akan menggunakan berbagai instrumen untuk menilai dan mengevaluasi siswa, seperti kuis, ujian, dan pekerjaan rumah. Pertanyaan terbuka dapat mendorong siswa untuk mengekspresikan kemampuan mereka dalam mengintegrasikan, mensintesis, mendesain, dan mengomunikasikan ide-ide jawaban mereka dalam bahasa alami. Namun, untuk menilai dan menganalisis jawaban – jawaban siswa mungkin agak sulit dan memakan waktu yang lama. Dari permasalahan tersebut sistem koreksi jawaban uraian singkat diperlukan untuk membantu para guru dalam mengoreksi evaluasi siswa dengan waktu yang singkat. Berbagai penelitian mengenai sistem koreksi jawaban uraian telah dikembangkan dengan memanfaatkan berbagai metode. Salah satu metode yang banyak digunakan dalam proses penilaian jawaban uraian singkat yaitu Cosine Similarity.. Pada penelitian ini sistem koreksi jawaban uraian singkat dikembangkan menggunakan metode Cosine Similariity dan Query Ekspansion. Proses koreksi terdiri dari empat tahapan, yaitu preprocessing (terdiri dari case folding, tokenization, stopwords, stemming), pencarian sinonim, penghitungan kemiripan menggunakan cosine similarity, dan pengklasifikasian nilai. Peneliti menambahkan metode query ekspansion untuk mengatasi jawaban siswa yang berbeda teks akan tetapi memiliki makna yang sama dengan kunci jawaban, serta dapat mendapatkan hasil penilaian yang akurat. Hasil pengujian sistem menggunakan jawaban dari 50 siswa dimana setiap siswa mempunyai empat jawaban dari empat soal berbeda memperoleh nilai akurasi penilain lebih tinggi dibandingkan sistem yang hanya menggunakan cosine similarity tanpa query ekspansion .
English:
Assessment is the main thing in the world of education. Through assessment, teachers can find out how effective their teaching is. In general, teachers will use various instruments to assess and evaluate students, such as quizzes, exams, and homework. Open-ended questions can encourage students to express their ability to integrate, synthesize, design, and communicate their answer ideas in natural language. However, assessing and analyzing student answers might be rather difficult and time-consuming. From these problems, a short description answer correction system is needed to help teachers correct student evaluations in a short amount of time. Various studies on the description of answer correction systems have been developed using various methods. One method that is widely used in the process of evaluating short description answers is Cosine Similarity. In this study, the short description answer correction system was developed using the Cosine Similarity and Expansion Query methods. The correction process consists of four stages, namely preprocessing (consisting of case folding, tokenization, stopwords, stemming), searching for synonyms, calculating similarities using cosine similarity, and classifying values. The researcher added the expansion query method to deal with students' answers that differed in the text but have the same meaning as the answer to the answer and can get accurate assessment results. The results of testing the system using the answers of 50 students in which each student has four answers to four different questions obtained a higher accuracy rating than a system that only uses cosine similarity without query ekspansion.
Item Type: | Thesis (Undergraduate) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Supervisor: | Syauqi, A’la and Holle, Khadijah Fahmi Hayati | |||||||||
Contributors: |
|
|||||||||
Keywords: | Koreksi Jawaban Uraian Singkat; Preprocessing; Cosine Similarity; Query Ekspansion; Automatic Correction of Short Description Answers | |||||||||
Subjects: | 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080107 Natural Language Processing | |||||||||
Departement: | Fakultas Sains dan Teknologi > Jurusan Teknik Informatika | |||||||||
Depositing User: | Moch. Irsyadul Anam | |||||||||
Date Deposited: | 20 Jul 2020 09:13 | |||||||||
Last Modified: | 05 May 2023 08:23 | |||||||||
URI: | http://etheses.uin-malang.ac.id/id/eprint/19696 |
Downloads
Downloads per month over past year
Actions (login required)
![]() |
View Item |