Application of Improved Sparrow Neural Network Algorithm Based on Chaotic Mapping for Waste Quantity Prediction

To improve the accuracy of waste generation predictions, a method incorporating chaotic mapping into the Sparrow Search Algorithm (SSA) is proposed. This enhanced algorithm is then integrated with neural networks for more precise predictions. The SSA typically encounters issues like slow convergence and getting stuck in local optima during optimization. By introducing Tent and Sine chaotic maps, the algorithm's randomness and diversity are increased, enhancing the initial population quality. Neural network backpropagation is then used to update positions. Through comparative analysis of BP, SSA, and chaotic SSA neural networks, more accurate predictions of waste generation are achieved.