Androfaza, Izzat Aji (2026) Implementasi Graph Transformation Engine dan Semantic Analyzer untuk Konversi Model Bpmn Berbasis Json menjadi Graph Database. Project Report. [s.n], [s.l.]. (Unpublished)
![[thumbnail of 5025221245-Project_Report (2).pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5025221245-Project_Report (2).pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy |
Abstract
Sistem Graph Transformation Engine dan Semantic Analyzer dikembangkan untuk mengonversi model BPMN berbasis JSON menjadi representasi graf di Neo4j serta memberikan analisis semantik menggunakan LLM Gemini. Sistem ini menerapkan proses parsing, transformasi elemen BPMN, pembentukan node–edge, penyimpanan metadata di MongoDB, hingga analisis struktur graf dan kompleksitas. Pengujian black-box menggunakan tiga file BPMN dengan ukuran kecil, sedang, dan besar menunjukkan bahwa konektivitas graph mencapai lebih dari 90%, transformasi struktur berjalan 100% sesuai jumlah elemen, seluruh query Cypher tereksekusi dengan benar, dan waktu eksekusi rata-rata berada pada 0,50 detik. Kinerja sistem tercatat stabil dengan waktu respon <2 detik untuk file kecil–menengah dan <5 detik untuk file besar. Selain itu, Semantic Analyzer berhasil memberikan rekomendasi penamaan dan tipe elemen secara konsisten untuk seluruh node, sehingga sistem dinyatakan mampu melakukan konversi BPMN dan analisis semantik secara akurat, terintegrasi, dan siap digunakan untuk analisis proses bisnis otomatis.
======================================================================================================================================
The Graph Transformation Engine and Semantic Analyzer system was developed to convert JSON-based BPMN models into graph representations in Neo4j and to perform semantic analysis using the Gemini LLM. The system implements BPMN parsing, element transformation, node–edge construction, metadata storage in MongoDB, and graph structure and complexity analysis. Black-box testing using three BPMN files of small, medium, and large sizes shows that graph connectivity exceeds 90%, structural transformation is 100% consistent with the number of elements, all Cypher queries are executed successfully, and the average execution time is 0.50 seconds. System performance remains stable, with response times of less than 2 seconds for small-to-medium files and less than 5 seconds for large files. In addition, the Semantic Analyzer consistently provides recommendations for element naming and types across all nodes, indicating that the system is capable of performing BPMN conversion and semantic analysis accurately, in an integrated manner, and is ready to be used for automated business process analysis.
| Item Type: | Monograph (Project Report) |
|---|---|
| Uncontrolled Keywords: | BPMN, Gemini, Graph Convert , Graph Database, LLM, Neo4j, Semantic Analyzer |
| Subjects: | T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.6 Management information systems |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Izzat Aji Androfaza |
| Date Deposited: | 15 Jan 2026 06:09 |
| Last Modified: | 15 Jan 2026 06:09 |
| URI: | http://repository.its.ac.id/id/eprint/129452 |
Actions (login required)
 |
View Item |