Concerning environmental concerns, it has become crucial to invent and develop alternative construction materials that can replace ordinary Portland cement and use waste materials instead of natural aggregates in the production of concrete. An important advancement in this area is the utilization of recycled plastics to create geopolymer concrete, which helps reduce plastic waste and minimize the environmental impact of construction materials. Since plastics have become a significant environmental issue, it is imperative to find ways to recycle them for a sustainable future. Thus, this study was conducted involving the preparation of seventeen different mixtures of geopolymer concrete (GPC) in three stages. In the first stage, five GPC mixes were created with varying amounts of nano-silica (NS) ranging from 0 % to 4 %. The main objective of this stage was to determine the optimal NS content that could enhance the performance of the GPC matrix. In the second stage, recycled plastic aggregates (RPA) were introduced to replace a portion of the fine aggregates, ranging from 0.05 to 0.5 volume percentages. This aimed to assess the impact of RPAs on the fresh GPC and its mechanical properties. The third stage involved an analysis to improve the performance of the GPC compound incorporating RPAs. The optimal NS content identified in the first stage was utilized for all mixtures with RPA. Lastly, several modeling techniques, including artificial neural network, M5P tree, multiple linear regression, full quadratic, multi-expression programming, and linear regression, were employed to predict the compressive strength of the GPC compound. The results demonstrated that the addition of nano-silica increased the compressive strength of the GPC mixtures by 6.3 %, 13.4 %, 20.5 %, 21 %, and 21.9 % at curing ages of 3, 7, 28, 90, and 180 days, respectively, when using 3 % NS content compared to the control GPC mixture. However, the inclusion of NS had a negative impact on the workability of the GPC mixes. Furthermore, incorporating RPAs reduced both the workability and compressive strength of the GPC compound. In the end, the results obtained from the modeling analysis revealed that the compressive strength of GPC could be predicted with a high degree of accuracy by taking into account the mixture proportions of its constituents. Among the various techniques used, it was found that the artificial neural network model outperformed other methods in accurately forecasting the compressive strength of GPC. This indicates that the neural network model is a superior and reliable approach for estimating the strength characteristics of GPC, providing valuable insights for optimizing its composition and enhancing its overall performance.