Virtual real estate investments are becoming an increasingly important investment instrument in terms of investing in virtual land assets in the Metaverse. In addition, when investment decisions in the Metaverse are analysed, it is observed that there are highly similar market movements with the real world. This situation can provide insight into real-world prediction when using traditional methods such as multiple linear regression (MLR) on very large databases. This paper analyses the correlation between the two worlds by obtaining statistical results using data mining algorithms.Purpose of the study is to identify the potential high-yielding ones among Istanbul neighborhoods with analyzes made for virtual lands, which are seen as a virtual real estate investment instrument, and to determine the relationship between them by comparative analysis of virtual land values and real estate values. In this context, Metaverse platform (overthereality.ai) where the world map is divided into virtual lands, was accessed with open access software to get the data of virtual lands traded in primary and secondary markets within the borders of Istanbul. At the same time, square meters prices data of residential units for sale were also accessed from the real estate site (sahibinden.com) with the highest sales volume in Turkey, and these data were converted into land values and statistically analyzed, together with virtual land values, specifically for Istanbul neighborhoods.