This paper highlights the challenges associated with unbalanced data by examining how balanced data affects Artificial Neural Network (ANN) classification on Rodent Tuber’s Liquid Chromatography Mass Spectrometry data. Chemical disparity in the dataset drives a reexamination of traditional algorithms and their preference for balanced data. We present ANN as an alternative, discussing its benefits, such as fault tolerance and flexible computation. The study assesses ANN performance using balancing data strategies: Synthetic Minority Over-sampling Technique (SMOTE), Adaptive Synthetic Sampling (ADASYN), Random Under-Sampling (RUS), and NearMiss. ADASYN achieved the best outcome and ANN still cannot provide optimal results with balanced data. So, additional studies and continuous improvements are needed.