Abstract: Both Dali M_S6.4 earthquake in Yunnan Province and the L'Aquila M_W6.3 earthquake in Italy caused heavy losses of life and properties because they failed to make accurate hazard predictions. What might be done to inform the government and the public about the possible occurrence of an earthquake and its maximum magnitude? What is the trend of earthquake in the future? In this paper, based on the idea of weather forecast, the seismicity is divided into several regions according to the clustering. The characteristics of seismicity before the two main earthquakes are analyzed from the aspects of M-t sequence in each area, seismicity of 3-magnitude earthquakes, the trend of energy release and escalation in the past two years. The G-R relationship, and the similarities between the two earthquakes are compared and analyzed. The similarities are as follows: ① Both hypocenters are shallow. The source depth of Dali M_S6.4 earthquake was 8 km, and that of the L'Aquila M_W6.3 earthquake in Italy was 8.8 km; ② b values are similar. They are 0.59 and 0.61, respectively. The G-R relation extrapolation magnitudes are close to each other, and they are all about magnitude 6; ③ The seismic energy release in M-t sequence shows an ascending trend, and the seismic activities are similar. After the calm was broken, earthquakes of magnitude 3 or above were active, and earthquake swarms appeared. The release of earthquake energy was accelerated; ④ They all belong to foreshock-main shock-aftershock type sequence. The largest foreshock before the M_W6.3 earthquake in Laquila, Italy, and the largest foreshock before the main-shock in Dali, Yunnan Province, are 5.2 and 5.6, respectively. And the characteristics of foreshock swarm are obvious. At the same time, the paper discusses the rationality of dividing the region according to the clustering of earthquakes, and it is considered that it is easier to grasp the characteristics of regional seismicity using this method. Although Dali M_S6.4 earthquake in Yunnan and L'Aquila M_W6.3 earthquake in Italy occurred in different areas with great structural differences and belong to different types of dislocation, the good results were obtained using the partition G-R relation and M-t sequence analysis. The results show that the G-R relation and M-t sequence analysis can be used to judge the seismic risk, which provides a basis for the next generation of “seismic risk cloud map”.