Metode Penilaian Keserupaan Semantik dan Struktural Rancangan Perangkat Lunak Berdasarkan Artifak Universal Modeling Language (UML)

Fauzan, Reza (2021) Metode Penilaian Keserupaan Semantik dan Struktural Rancangan Perangkat Lunak Berdasarkan Artifak Universal Modeling Language (UML). Doctoral thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
05111860010001-Disertation.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (5MB) | Request a copy
[img] Text
view_usp=sharing - Accepted Version
Restricted to Repository staff only

Download (74kB) | Request a copy

Abstract

Penilaian keserupaan diagram UML memiliki beberapa permasalahan utama. Permasalahannya terdapat pada penggunaan kembali diagram UML, deteksi kloning diagram UML, dan penilaian keserupaan diagram UML. Penelitian ini berfokus pada penilaian keserupaan diagram UML. Dalam penilaian keserupaan, keserupaan memiliki dua karakteristik, yaitu keserupaan struktural dan keserupaan semantik. Penilaian otomatis saat ini hanya melihat keserupaan semantik saja, sedangkan keserupaan semantik tidak dapat melihat struktur dari diagram secara keseluruhan. Penelitian ini telah membuat metode penilaian keserupaan antara dua diagram UML sejenis dengan melihat aspek semantik dan struktural diagram. Diagram UML yang dihitung adalah diagram kasus penggunaan (use case diagram), diagram kelas (class diagram), diagram urutan (sequence diagram) dan diagram keadaan (state diagram). Penilaian keserupaan semantik dilakukan dengan membandingkan informasi leksikal dari dua diagram yang dihitung. Penilaian keserupaan struktural dilakukan dengan mengkonversi diagram menjadi sebuah graf, kemudian dihitung menggunakan Greedy Graph Edit Distance. Metode penilaian keserupaan ini dapat berkinerja seandal (reliability) seorang pakar dalam menentukan keserupaan antara dua rancangan perangkat lunak sejenis untuk keperluan penilaian keserupaan. Pada diagram kelas dan diagram urutan, kesepakatan antara metode yang diusulkan dan pakar terdapat pada kesepakatan substansial. Pada diagram kasus penggunaan dan diagram keberadaan, kesepakatan antara metode yang diusulkan dan pakar terdapat pada kesepakatan hampir sempurna. ==================================================================================================== The UML diagram conformity assessment has several major problems. The problem lies in the reuse of UML diagrams, detection of cloning UML diagrams, and assessment of UML diagram similarity. This study focuses on assessing the similarity of UML diagrams. In the assessment of similarity, similarity has two characteristics, namely structural similarity and semantic similarity. Currently automatic assessment looks at semantic similarity only, while semantic similarity cannot see the structure of the diagram as a whole. This study has made a method of assessing the similarity between two UML diagrams by looking at the semantic and structural aspects of the diagram. The calculated UML diagrams are use case diagrams, class diagrams, sequence diagrams and state diagrams. Semantic similarity assessment is done by comparing the lexical information from the two calculated diagrams. Structural similarity assessment is done by converting the diagram into a graph, then it is calculated using the Greedy Graph Edit Distance. This similarity assessment method can perform as reliably as an expert in determining the similarity between two similar software designs for similarity assessment purposes. In class diagrams and sequence diagrams, agreement between the proposed method and the expert is substantial. On the use case diagram and the existence diagram, the agreement between the proposed method and the expert is almost perfect agreement.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: penilaian keserupaan desain perangkat lunak, penilaian keserupaan diagram UML, penilaian otomatis, penilaian semantik, penilaian struktural, automatic assessment, semantic assessment, software design similarity assessment, structural assessment, UML diagram similarity assessment
Subjects: Q Science > QA Mathematics > QA76 Computer software
Q Science > QA Mathematics > QA76.758 Software engineering
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55001-(S3) PhD Thesis
Depositing User: Reza Fauzan
Date Deposited: 15 Jun 2021 09:28
Last Modified: 15 Jun 2021 09:28
URI: https://repository.its.ac.id/id/eprint/84246

Actions (login required)

View Item View Item