Swarm intelligence optimization algorithms are a class of effective methods for solving complex optimization problems by simulating the collective behavior of biological groups in nature. This paper proposes a novel swarm intelligence optimization algorithm—the Bird-of-Paradise Optimization Algorithm (BPOA)—inspired by the foraging, courtship displays, and social behaviors of birds of paradise. The algorithm models the long-distance foraging behavior of birds of paradise as a global exploration mechanism, the courtship display behavior as a local exploitation mechanism, and combines this with social network information exchange strategies to achieve an efficient combination of global search and local optimization. This paper introduces mechanisms such as individual display intensity, adaptive exploration energy, neighbor information sharing, and multi-scale random perturbation into the algorithm model to improve the search efficiency and convergence performance in high-dimensional complex optimization problems. Through mathematical formulas and detailed process descriptions, this paper elucidates the core principles and innovative features of BPOA, providing a theoretical foundation for its further application in continuous optimization, combinatorial optimization, and constrained optimization problems.