Modeling of Crude Oil Types Classification Using the Naive Bayes Classifier Method


  • Harry Budiharjo Sulistyarso Universitas Pembangunan Nasional Veteran Yogyakarta
  • Dyah Ayu Irawati Universitas Pembangunan Nasional Veteran Yogyakarta
  • Joko Pamungkas Universitas Pembangunan Nasional Veteran Yogyakarta
  • Indah Widiyaningsih Universitas Pembangunan Nasional Veteran Yogyakarta



Crude oil, EOR, Naïve Bayes, Prediction, Viscosity


This research is part of previous research regarding the prediction of petroleum's physical properties to help the user get the prediction value of crude oil's physical properties from field test data, which was carried out from Enhanced Oil Recovery Research Laboratory, Petroleum Engineering UPN “Veteran” Yogyakarta. The field data that is measured in the laboratory that has been done is by adding biosurfactants and increasing the temperature. Various steps have been taken to reduce crude oil's viscosity value so that it can be diluted could flow. It is necessary to calculate the viscosity of crude oil in this process to determine the extent of the viscosity level as expected by adding biosurfactants or increasing the temperature that has been carried out in the EOR process. Naïve Bayes Classifier is used to classify oil data into three categories: light oil, medium oil, and heavy oil, based on the prediction of the viscosity value. The Naïve Bayes Classifier is a robust algorithm for performing machine learning-based predictive modeling that applies the Bayes theorem. This predictive modeling for the physical properties of crude oil was built using the Python programming language and the PyQt5 library to build desktop-based applications. The classification of oil has arrived at labeling the prediction results of crude oil's viscosity into three categories, namely light oil, medium oil, and heavy oil. The test results with testing data produce accurate data; the predicted value is within the specified range.


ASTM D445-11, Standard Test Method for Kinematic Viscosity of Transparent and Opaque Liquids (and Calculation of Dynamic Viscosity). Designation 71/1/97. ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States. pp.2-3.

Ayu Irawati D. et al., 2018. "Development of Android-based Rabbit Disease Expert System." International Journal of Engineering and Technology. 7(4.44). DOI: 10.14419/ijet.v7i4.44.26868. pp. 82-87

Nugroho A., Subanar. 2013. “Klasifikasi Naïve Bayes untuk Prediksi Kelahiran pada Data Ibu Hamil”. Jurnal Berkala MIPA Vol 23, No.3(2013) ISSN:0215-9309. Fakultas MIPA UGM. Pp 297-308.

Pressman, S. Roger. Software Engineering: A Practitioner's Approach, Eight Edition. 2015. McGraw Hill. The USA. p. 42

Saleh, A. 2015. Implementasi Metode Klasifikasi Naïve Bayes Dalam Memprediksi Besarnya Penggunaan Listrik Rumah Tangga. Citec Journal, Vol. 2, No. 3, ISSN: 2354-5771. pp 207-217.

Sarkar S dan Sriram Ram S. 2001. Bayesian Models for Early Warning Bank Failures. Management Science Journal. 47(11). pp 1457–1475.

Sulistyarso HB., Pamungkas, Joko, Gusmawarni, Sri Rahayu, Widyaningsih, Indah, Kurnia, Rafli Arie. 2020. “Application of Bio-surfactants as an Effort to Enhanced Oil Recovery (EOR) in Kawengan Oil Field," AIP Conference Proceedings 2245, 090018. pp. 1-5.

Vural MS dan Gök M. 2017. Criminal prediction using Naïve Bayes Theory. ACM Journal Neural Computing and Applications. 28(9) pp 2581–2592. DOI 10.1007/s00521-016-2205-z.