One of the most successful new internet companies, judging by the amount of traffic that they are getting, is Zillow, a real estate data company that specializes in the prices of housing. However, they have provided very interesting plots of home values in several metropolitan areas in the US. Finally, we can throw away the Boston housing dataset.
Zillow's method of producing real estate values is proprietary and has an unknown level of relation to reality.
My anecdote: I live on a street of identical townhouses, so I can use many comparable last sales to get a very good idea of my house value. But Zillow marks up my house above anything reasonable by about 20%.
This wasn't always the case: a few months ago, the value of my house jumped about 10% in one week.
Other Zillow users have told me similar stories, about how their house is Zestimated at far below or far above value. On average, it clearly does OK, and the heat maps to which you link are entirely plausible. But I'd have great reluctance toward anyone who uses their proprietary estimates for academic research. Their website protects the hard-data last sales from robot harvest, so you'll have to go elsewhere for that too.