Pengukuran Keserupaan Semantik Menggunakan Pemrosesan Bahasa Alami Dan Keserupaan Struktural Menggunakan Graph Edit Distance-Greedy Pada Statechart Diagram

Munawaroh, Hidayatul (2021) Pengukuran Keserupaan Semantik Menggunakan Pemrosesan Bahasa Alami Dan Keserupaan Struktural Menggunakan Graph Edit Distance-Greedy Pada Statechart Diagram. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 05111950010033-Master_Thesis.pdf] Text
05111950010033-Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 April 2023.

Download (2MB) | Request a copy

Abstract

E-learning merupakan media pembelajaran secara elektronik untuk mendukung penyampaian dan penilaian materi pembelajaran. Model e-learning ini memungkinkan jumlah peserta didik yang jauh lebih banyak dibandingkan dengan kelas konvensional, akibatnya mendorong kebutuhan adanya otomatisasi proses evaluasi. Sistem penilaian otomatis yang ada pada e-learning meliputi jawaban benar/salah, pencocokan, pilihan ganda, daftar pilihan jawaban, dan isian teks. Dalam penerapannya, beberapa mata pelajaran seperti Rekayasa Perangkat Lunak (RPL) soal ujian yang diberikan terkadang melibatkan jawaban uraian dalam bentuk diagram Unified Modelling Language (UML). Sehingga perlu ada pengembangan terkait penilaian jawaban tersebut.
Penelitian ini mengajukan metode untuk mengukur keserupaan diagram UML khususnya statechart diagram. Pengukuran yang dilakukan antara diagram jawaban dan diagram kunci jawaban yang selanjutnya dapat digunakan sebagai dasar atau pertimbangan pengampu mata pelajaran untuk memberikan nilai. Pengukuran dilakukan dengan memperhatikan dua aspek, yaitu semantik dan struktural. Pengukuran keserupaan semantik dilakukan dengan membandingkan informasi leksikal dari kedua diagram yang akan diukur. Pengukuran keserupaan struktural dilakukan dengan memodelkan diagram dalam bentuk graf kemudian dihitung menggunakan metode Graph Edit Distance (GED) dengan tambahan algoritma Greedy.
Hasil pengujian menunjukkan bahwa metode pengukuran keserupaan yang diajukan dapat diandalkan seperti seorang pengajar mata pelajaran RPL dalam menilai jawaban statechart diagram siswa. Hasil kesepakatan antara pakar dan metode yang diperoleh adalah “almost perfect agreement” dengan nilai kesepakatan 0,85.

=====================================================================================================

E-learning is an electronic learning media to support the delivery and assessment of learning materials. This e-learning model allows a greater number of students compared to conventional classes, consequently driving the need for automation of the evaluation process. The automatic evaluation system that exists in e-learning includes true/false answers, matching, multiple-choice, answer choice lists, and text entries. In its application, some subjects such as Software Engineering (SE) the exam questions sometimes involve answers Unified Modeling Language (UML) diagrams. So there needs to be development related to the evaluation of these answers.
This study proposes a method for measuring the similarity of UML diagram, especially statechart diagram. Measurements made between the answer diagram and the key answer diagram which can then be used as a basis or consideration for lecturer to provide grades. Measurements based on two aspects, semantic and structural. The measurement of semantic similarity is done by comparing the lexical information from the two diagrams to be measured. Structural similarity measurement is done by modeling the diagram into graphs and then calculated using the Graph Edit Distance (GED) method with the addition of the Greedy algorithm.
The result obtained shows that the proposed method is reliable with lecturers of Sofware Engineering subject assessment result. Lecturers assess the statechart diagram answers from students. The result of the agreement between the experts and the method obtained was “almost perfect agreement” with an agreement value of 0.85.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Statechart diagram similarity, semantic similarity, structural similarity, Graph Edit Distance (GED), greedy, Keserupaan statechart diagram, keserupaan semantik, keserupaan struktural, Graph Edit Distance (GED), greedy.
Subjects: T Technology > T Technology (General) > T58.62 Decision support systems
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.88815 Semantic Web
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55101-(S2) Master Thesis
Depositing User: Hidayatul Munawaroh
Date Deposited: 01 Mar 2021 05:00
Last Modified: 01 Mar 2021 05:00
URI: http://repository.its.ac.id/id/eprint/83010

Actions (login required)

View Item View Item