• Pendaftaran
  • Webmail
  • Perpustakaan
  • Portal Akademik
  • Universitas
  • Indonesia
    • English
Universitas Gadjah Mada Departemen Ilmu Komputer dan Elektronika
Fakultas Matematika dan Ilmu Pengetahuan Alam
  • Home
  • Departemen
    • Manajemen
    • Staf Administrasi & Teknis
    • Fasilitas
    • Kontak & Lokasi
  • Program Studi
    • S1 Ilmu Komputer
    • S1 Elektronika dan Instrumentasi
    • Magister Ilmu Komputer
    • Program Magister Elektronika dan Instrumentasi
    • Magister Kecerdasan Artifisial
    • Program Doktor Ilmu Komputer
  • Riset
    • Kelompok Riset
      • Elektronika dan Instrumentasi
      • Sistem Komputer dan Jaringan
      • Algoritma dan Komputasi
      • Rekayasa Perangkat Lunak dan Data
      • Sistem Cerdas
      • Lab Riset​ Sistem Tertanam dan Robotika​
    • Roadmap Penelitian dan Pengabdian kepada Masyarakat
    • Laboratorium Layanan
      • Lab Komputer Dasar
      • Lab Elektronika Dasar
      • Lab Instrumentasi Dasar
  • Pengajar & Publikasi
    • Daftar Dosen
  • Akademik & Mahasiswa
    • Pendaftaran
    • Program Pra Doktoral
    • SOP Panduan MBKM DIKE
    • Laporan Evaluasi Semester Program Studi
    • Prosedur
    • Mahasiswa
      • HIMAKOM
      • HMEI
      • OMAH TI
    • Beasiswa
    • Dokumen Akademik
  • Beranda
  • Semua Berita
  • Berita Umum
  • Riset
  • Dosen DIKE UGM dan Alumni MKOM Kembangkan Automatic Extraction of User Story Elements Berbasis NLP

Dosen DIKE UGM dan Alumni MKOM Kembangkan Automatic Extraction of User Story Elements Berbasis NLP

  • Riset
  • 31 Juli 2025, 01.18
  • Oleh: wimbono
  • 0

Yogyakarta, 28 Juli 2025 – Dr. Sri Mulyana, M.Kom., dosen Magister Kecerdasan Artifisial (MKA) dari Departemen Ilmu Komputer dan Elektronika (DIKE) FMIPA UGM, bersama alumni Program Studi Magister Komputer, Aszani, M.Kom., mengembangkan metode ekstraksi otomatis elemen user story menggunakan pendekatan Natural Language Processing (NLP). Penelitian ini mengangkat topik “Automatic Extraction of User Story Elements: NLP-Based Approach” dan menggabungkan teknik Named Entity Recognition (NER) dengan model tok2vec dan transformer.

Penelitian ini diawali dengan eksplorasi sistematis terhadap 50 artikel ilmiah dari Google Scholar yang terbit antara tahun 2019 hingga 2022. Semua artikel dipilih secara manual dengan fokus pada pengembangan perangkat lunak di bidang pendidikan. Proses labeling dilakukan untuk membedakan empat elemen utama dalam user story, yaitu aktor, tindakan, deskripsi, dan topik. Hasilnya, diperoleh dataset berlabel sebanyak 2.369 baris, mencakup ratusan entitas dari setiap kategori.

Hasil evaluasi kinerja menunjukkan perbedaan yang signifikan antara model tok2vec dan transformer dalam mengidentifikasi entitas di berbagai dokumen. Pada dokumen awal, model tok2vec mengidentifikasi 37 entitas, sementara model transformer mengenali 17 entitas. Meskipun jumlah entitas lebih sedikit, model transformer menghasilkan tingkat akurasi yang lebih tinggi yaitu 61,67%, dibandingkan model tok2vec yang hanya 42,26%. Sebaliknya, pada dokumen yang kedua, kedua model tidak berhasil mengidentifikasi entitas apa pun, hal ini menjadi keterbatasan dari kedua model tersebut dalam konteks tertentu. Sedangkan pada dokumen ketiga, model transformer mendeteksi 49 entitas dan model tok2vec mendeteksi 95 entitas. Meskipun jumlah entitas yang dihasilkan lebih rendah, model transformer mencapai akurasi yang lebih tinggi, yaitu 71,53%, sedangkan model tok2vec 65,73%. Demikian pula, pada dokumen keempat, kedua model mengidentifikasi 40 entitas, dan model transformer menunjukkan akurasi yang unggul, yaitu 75,00%, sedangkan model tok2vec 45,03%. Yang terakhir pada dokumen kelima, ternyata model transformer dapat mengidentifikasi 54 entitas dengan akurasi sempurna (100%), sementara model tok2vec mengenali 83 entitas dengan akurasi 62,29%.

Penelitian ini telah dipresentasikan pada forum ilmiah internasional International Conference on Advanced Machine Learning and Data Science (AMLDS) 2025 yang berlangsung di Tokyo, Jepang pada 19-21 Juli 2025. Partisipasi dalam konferensi ini menjadi bukti kontribusi DIKE UGM dalam bidang riset pemrosesan bahasa alami dan pengembangan perangkat lunak berbasis AI.

Riset ini juga berkontribusi dalam pencapaian Tujuan Pembangunan Berkelanjutan (SDGs), khususnya SDG 9: Industri, Inovasi, dan Infrastruktur. Melalui pemanfaatan NLP untuk meningkatkan pemahaman kebutuhan pengguna dalam pengembangan sistem, penelitian ini mendorong terciptanya ekosistem digital yang lebih cerdas.

 

Author: Sri Mulyana, Aszani
Editor: Marina
#SDGs9 

Tags: SDGs SDGs 9: Industri Inovasi dan Infrastruktur

Programs

  • Undergraduate Programs in Computer Science
    • + Regular Program
    • + International Program
  • Undergraduate Programs in Electronics & Instrumentation
    • + Regular Program
    • + International Program
  • Postgraduates
    • + Magister in Computer Science
    • + Magister in Artificial Intelligence
    • + Magister in Electronics & Instrumentation
  • + Doctorate in Computer Science

Prospective Students, welcome!


Berita UGM

  • Childhood Allergy Cases Rising in Indonesia, UGM Pediatrician Stresses Accurate Diagnosis and Prevention
    Cases of childhood allergies continue to rise in Indonesia, requiring proper diagnosis and treatment to prevent adverse effects on children’s growth and development. Data from the World Allergy Organization (WAO) indicate that the global prevalence of allergies ranges from 10% to 40% of the population. In Indonesia, data from the Indonesian Pediatric Society (IDAI) show […]
  • Arjuna UGM Wins Six Awards at Formula SAE Japan Electric Vehicle Competition
    The silent roar of Indonesia’s electric car stunned audiences at the Aichi Sky Expo, Tokoname, Nagoya. Universitas Gadjah Mada’s (UGM) Arjuna EV Team achieved a monumental milestone at the Formula Society of Automotive Engineers (FSAE) Japan 2025, held from Sep. 8 to 13.  Amid intense competition against 84 teams from six technologically advanced countries, Arjuna […]
  • 610 UGM Students Awarded LPDP Scholarships for 2025, Driving Nation-Building Efforts
    A total of 610 students from Universitas Gadjah Mada (UGM) have been awarded scholarships by the Indonesia Endowment Fund for Education (LPDP) for the odd semester of 2025. At the Welcoming Awardee event held on Sunday (Sep. 28) at the ASLC Hall of the Faculty of Animal Science (Animal Science UGM), 270 awardees attended in […]
  • Biology UGM Shares Best Practices in Waste Management through Interdisciplinary Collaboration
    The Faculty of Biology, Universitas Gadjah Mada (Biology UGM), through its Organic Waste Management Task Force, accepted an invitation from the Faculty of Mathematics and Natural Sciences (FMIPA UGM) to share experiences in waste management. The session, held on Saturday (Sep. 27) at the Herman Yohannes Auditorium, was attended by lecturers, department heads, and around […]
  • KPU Correct to Revoke Candidate Document Restrictions, UGM Political Communication Expert Affirms
    The General Elections Commission (KPU) officially revoked KPU RI Decree No. 731 of 2025 concerning the Determination of Presidential and Vice Presidential Candidate Requirements Documents as Restricted Public Information on Tuesday, Sep. 16.  The decree had previously classified 16 documents as restricted, including copies of national ID cards, diplomas, police clearance certificates (SKCK), and others. […]
  • RSA UGM Doctor Shares Health and Nutrition Tips for Successful Blood Donation
    Blood donation is a form of care that can save many lives. However, several requirements must be met for someone to pass blood donor screening. Through the TikTalk RSA UGM Podcast on Friday (Sep. 26), Universitas Gadjah Mada Academic Hospital (RSA UGM) featured its Head of the Blood Transfusion Unit and clinical pathologist, Titien Budhiaty, […]
  • KKN-PPM UGM Revitalize Uvenja Weir in Donggala, Central Sulawesi
    After more than 14 years in disrepair, the Uvenja Weir in Dusun II, Limboro Village, Donggala Regency, Central Sulawesi Province, is now fully functional again. The revitalization was successfully carried out through a collaborative effort by Universitas Gadjah Mada (UGM) Community Service Program (KKN-PPM) students as part of initiatives aimed at strengthening food security. Addien […]
  • UGM Study Reveals Link Between Knife Sharpness and Chicken Meat Quality
    A team of researchers from the Faculty of Veterinary Medicine, Universitas Gadjah Mada (FKH UGM) is conducting a study on the effect of knife sharpness on the quality of chicken meat.  The research also highlights the role of certified halal slaughterers (juru sembelih halal or juleha) in ensuring both the quality of meat and animal […]
  • UGM’s Sekolah Wartawan Explores Evidence-Based Policy Briefs to Strengthen Media Literacy
    Universitas Gadjah Mada (UGM) reaffirmed its commitment to strengthening public policy literacy through the Sekolah Wartawan program. In a session held on Monday (Sep. 29), the program featured Alfath Bagus Panuntun El Nur Indonesia, who delivered a presentation on the theme Developing Evidence-Based Policy Briefs.  He emphasized the role of policy briefs as a bridge […]
  • 2,143 Athletes Compete in Hamengku Buwono X Cup Badminton Championship
    A total of 2,143 athletes from 433 clubs across Indonesia are competing in the 2025 Hamengku Buwono X Cup National Badminton Circuit C Championship, which officially commenced on Monday (Sep. 29) at the Universitas Gadjah Mada (UGM) Sports Hall. The national-level championship was inaugurated by the Governor of the Special Region of Yogyakarta, Sri Sultan […]
Universitas Gadjah Mada

Departemen Ilmu Komputer dan Elektronika

Fakultas Matematika dan Ilmu Pengetahuan Alam

Universitas Gadjah Mada

Gedung C, Lantai 4

Sekip Utara Bulaksumur Yogyakarta 55281

Introduction

Mahasiswa

©

KEBIJAKAN PRIVASI/PRIVACY POLICY