Human activity discovery aims to cluster the activities performed by humans without any prior information on what defines each activity. Most methods presented in human activity recognition are supervised, where there are labeled inputs to train the system. In reality, it is difficult to label activities data because of its huge volume and the variety of human activities. This paper proposes an unsupervised framework to perform human activity discovery in 3D skeleton sequences. First, an approach for data pre-processing is presented. In this stage, important frames are selected based on kinetic energy. Next, the displacement of joints, statistical displacements, angles, and orientation features are extracted to represent the activities information. Since not all extracted features have useful information, the dimension of features is reduced using PCA. Most methods proposed for human activity discovery are not fully unsupervised. They use pre-segmented videos before categorizing activities. To deal with this, we have used a sliding time window to segment the time series of activities with some overlapping. Then, activities are discovered by our proposed Hybrid Particle swarm optimization (PSO) with Gaussian Mutation and K-means (HPGMK) algorithm to provide diverse solutions. PSO is used due to its straightforward idea and powerful global search capability which can identify the ideal solution in a few iterations. Finally, k-means is applied to the outcome centroids from each iteration of the PSO to overcome the slow convergence rate of PSO. The experiment results on five datasets show that the proposed framework has superior performance in discovering activities compared to the other state-of-the-art methods and has increased accuracy of at least 4% on average.