This study is about monitoring the vibration signatures of a Compressed Ignition (CI) engine tested with fish oil biodiesel as a fuel. For analyzing the signatures, data mining plays a vital role. Machine learning has been extensively applied to the engine for the fault diagnosis. Vibration signals are captured using shear accelerometer by varying the load at a constant speed of 1800rpm. Statistical features were extracted from the captured rough vibration signals and feature selection was carried out. The selected features were then classified using Tree classifiers like J48 and Hoeffiding tree. The classification accuracy has been discussed. Comparatively, Hoeffiding tree maximum classification accuracy with 97% © 2019 IEEE.