The recurrence properties of random walks can be characterized by Pólya number, i.e., the probability that the walker has returned to the origin at least once. In this paper, we consider recurrence properties for a general 1D random walk on a line, in which at each time step the walker can move to the left or right with probabilities l and r, or remain at the same position with probability o (l + r + o = 1). We calculate Pólya number P of this model and find a simple expression for P as, P = 1 − ∆, where ∆ is the absolute difference of l and r (∆ = |l − r|). We prove this rigorous expression by the method of creative telescoping, and our result suggests that the walk is recurrent if and only if the left-moving probability l equals to the right-moving probability r.PACS numbers: 05.40.Fb, 05.60.Cd, 05.40.Jc Random walk is related to the diffusion models and is a fundamental topic in discussions of Markov processes. Several properties of (classical) random walks, including dispersal distributions, first-passage times and encounter rates, have been extensively studied. The theory of random walk has been applied to computer science, physics, ecology, economics, and a number of other fields as a fundamental model for random processes in time [1][2][3][4].An interesting question for random walks is whether the walker eventually returns to the starting point, which can be characterized by Pólya number, i.e., the probability that the walker has returned to the origin at least once during the time evolution. The concept of Pólya number was proposed by George Pólya, who is a mathematician and first discussed the recurrence property in classical random walks on infinite lattices in 1921 [5,6]. Pólya pointed out if the number equals one, then the walk is called recurrent, otherwise the walk is transient because the walker has a nonzero probability to escape [7]. As a consequence, Pólya showed that for one and two dimensional infinite lattices the walks are recurrent, while for three dimension or higher dimensions the walks are transient and a unique Pólya number is calculated for them [8]. Recently, M.Štefaňák et al. extend the concept of Pólya number to characterize the recurrence properties of quantum walks [9][10][11]. They point out that the recurrence behavior of quantum walks is not solely determined by the dimensionality of the structure, but also depend on the topology of the walk, choice of coin operators, and the initial coin state, etc [9][10][11]. This suggests the Pólya number of random walks or quantum walks may depends on a variety of ingredients including the structural dimensionality and model parameters.In this paper, we consider recurrence properties for a general one-dimensional random walk. The walk starts at x = 0 on a line and at each time step the walker moves one unit towards the left or right with probabilities l and r, or remain at the same position with proba- * Electronic address: xuxinping@suda.edu.cn bility o (l + r + o = 1). This general random walk model has some useful application in phy...