We propose a fuss-free gait analyzer based on a single three-axis accelerometer mounted on a cell phone for health care and presence services. It is not necessary for users not to wear sensors on any part of their bodies; all they need to do is to carry the cell phone. Our algorithm has two main functions; one is to extract feature vectors by analyzing sensor data in detail using wavelet packet decomposition. The other is to flexibly cluster personal gaits by combining a self-organizing algorithm with Bayesian theory. Not only does the three-axis accelerometer realize low cost personal devices, but we can track aging or situation changes through on-line learning. A prototype that implements the algorithm is constructed. Experiments on the prototype show that the algorithm can identify gaits such as walking, running, going up/down stairs, and walking fast with an accuracy of about 80[%].