The growing environmental concerns and increasing volumes of plastic waste necessitate efficient and accurate methods for plastic classification in waste recycling. Laser-Induced Breakdown Spectroscopy (LIBS) has emerged as a powerful analytical technique for rapid material identification. In this study, we utilized LIBS spectral data to develop and optimize machine learning (ML) models for accurate classification of various plastic types. Initially, Pearson correlation analysis was employed to evaluate spectral similarities, followed by principal component analysis (PCA), which demonstrated effective clustering of different plastics. An artificial neural network (ANN) was then used to assess classification accuracy. Our findings indicate that the presence of contaminants can influence classification performance. Overall, the integration of LIBS with ML techniques offers a promising approach for automated, real-time plastic sorting, significantly improving the efficiency and sustainability of the waste recycling process.