Abstract: Based on 16 municipal panel data of Anhui Province from 2015 to 2020, this paper uses the hierarchical analysis—entropy weight method to construct a regional innovation evaluation system, characterizes the level of high-tech industry agglomeration by location entropy, and then empirically analyzes the interaction between regional innovation system and high-tech industry agglomeration. The results show that: the differences in the trends of the concentration of high-tech industries and the differences in the comprehensive scores of regional innovation systems in Anhui Province are generally consistent, and there is a Granger causality relationship between the two; the spatial distribution characteristic is a decreasing trend in the north-south circle with Hefei and Wuhu as the centre; there is a significant "snowball effect" in the development of high-tech industry agglomeration in Anhui Province; the regional innovation system has a more obvious role in promoting the level of high-tech industry agglomeration than a single indicator; foreign direct investment, the investment in scientific research and the level of economic development all have significant positive effects on high-tech industry agglomeration; but due to the siphoning effect of Jiangsu Province, Zhejiang Province and Shanghai on high-level scientific and technological innovation talents in Anhui Province and the tendency towards the heterogeneous structure of scientific and technological innovation talents, the investment in scientific research personnel cannot play a positive role in promoting the high-tech industry agglomeration for the time being.

Key words: hierarchical analysis method, entropy weight method, regional innovation system, locational entropy, high-tech industry agglomeration