Because of the omnipresence of microplastics in soil, water, sediment, and even food items, researchers gained interest in identifying and classifying them to study the hazards they pose. The classification of microplastics based on spectroscopic techniques like Raman, Fourier Transform Infrared (FTIR) spectroscopy, Laser-Induced Breakdown Spectroscopy (LIBS), and others can be done to detect source and estimate the human exposure. Often these spectroscopic data are complex, making it difficult to extract the minute information for meaningful interpretation. In such situations, chemometric tools (a combination and mathematical and statistical methods to extract maximum information from the raw data) can be used to facilitate robust analysis to get comprehensive results. In this chapter, we are discussing various statistical and machine learning tools combined with spectroscopic data for microplastics classification. Among the statistical tools, Principal Component Analysis (PCA) is widely used to classify microplastics based on Raman/FTIR data, where the co-variance among the data points are extracted to implement the classification. For complex data like LIBS, machine learning tools like Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Network (ANN) are preferred, which provides better identification accuracy. Also, preprocessing of the data (such as smoothening, baselining, and normalizing) before coupling to suitable classification algorithm significantly reduces the analytical time and aids in successful classification. In addition, implementation of image processing with spectroscopic analysis enhances the identification accuracy further.