Besemah Language Translation Machine Model Based on Machine Learning

Strategic Insights -

Research at Universitas Bina Darma in Indonesia aims to develop a translation machine for the Besemah language, which is spoken by a tribe in South Sumatra Province. The study utilizes machine learning and neural networks to create a model that can translate the Besemah language into Indonesian and vice versa. The research found that the model achieved a high accuracy rate of 84.69% for translating Besemah to Indonesian and 84.92% for translating Indonesian to Besemah.

Key Takeaways:

  • The Besemah language, spoken by the Besemah tribe in South Sumatra Province, Indonesia, is at risk of becoming extinct as the number of speakers decreases over time.
  • Machine learning and neural networks were used to develop a translation machine for the Besemah language.
  • The study aimed to create a model that can translate the Besemah language into Indonesian and vice versa.
  • The research utilized the Recurrent Neural Network (RNN) approach to develop the translation machine.
  • The model achieved high accuracy rates for translating Besemah to Indonesian (84.69%) and Indonesian to Besemah (84.92%).
  • The study used experimental research in machine learning to validate the performance of the translation model.
  • The research was conducted at Universitas Bina Darma in Indonesia.

Statistics:

  • The val_accuracy value for the Besemah-Indonesian translation was 0.8469.
  • The val_accuracy value for the Indonesian-Besemah translation was 0.8492.
  • 100 epochs were used in the translation trial.
  • The batch size for the translation trial was 64.
  • The validation split for the translation trial was 0.2.

Sources:

  • Besemah Language Translation Machine Model Based on Machine Learning with Recurrent Neural Network (RNN) Model Algorithm. Jurnal Teknologi Informatika & Komputer, 2025,11(1):394-408.
  • DOI: 10.37012/jtik.v11i1.2614
  • Universitas Bina Darma
  • Universitas Mohammad Husni Thamrin (publisher of Jurnal Teknologi Informatika & Komputer)
  • https://doi-org.sdpl.idm.oclc.org/10.37012/jtik.v11i1.2614