With the rapid development of artificial intelligence, governments of different countries have been focusing on building smart cities. To build a smart city is a system construction process which not only requires a lot of human and material resources, but also takes a long period of time. Due to the lack of enough human and material resources, it is a key challenge for lots of small and medium-sized cities to develop the intelligent construction, compared with the large cities with abundant resources. Reusing the existing smart city system to assist the intelligent construction of the small and medium-sizes cities is a reasonable way to solve this challenge. Following this idea, we propose a model of Ant Colony Optimization Ridge Regression (ACO-RR), which is an smart city evaluation method based on the ridge regression. The model helps small and medium-sized cities to select and reuse the existing smart city systems according to their personalized characteristics from different successful stories. Furthermore, the proposed model tackles the limitation of ridge parameters' selection affecting the stability and generalization ability, because the parameters of the traditional ridge regression is manually random selected. To evaluate our model performance, we conduct experiments on real-world smart city data set. The experimental results demonstrate that our model outperforms the baseline methods, such as support vector machine and neural network.