This paper presents an expert monitoring algorithm approach to detect, locate and quantify stiffness variations in civil engineering structures. The algorithm is based on pattern recognition and techniques that emulate knowledge of experts. Time-frequency information about dynamics of structure is processed by using discrete wavelet transform, self-organizing maps, case-based reasoning and principal component analysis. Environmental excitation is considered as a viable option for monitoring the structural health at level III. The proposed algorithm can be adapted tor other type of structures with an appropriate analytical model.