Interview with Data Scientist Jukka Viikari  about Property Data Reliability

Interview with Data Scientist Jukka Viikari about Property Data Reliability

We’ve interviewed our Data Scientist, an expert in analysing data, Jukka Viikari, to get a better view of the problems he faces every day with real-estate data, and the solutions he implements to solve these issues.  

What kind of problems do you face with data?

One problem is missing data. Sometimes our customers have limited knowledge of the physical attributes of the house/apartment or they don’t know the exact location. That kind of situations are easier to cope with because we know what we don’t know. Our confidence indicator will take that into account and we can also use other sources to fill missing fields.

A much harder situation is when a customer has incorrect information about its properties. Usually, we assume that customer data is correct, but this is not always the case. The reliability of property valuation using statistical methods depends crucially on the quality of data. Hence incorrect data will lead to incorrect results.

Can you give me an example?

For instance, the floor area of a house is usually called just floor area. Sometimes that means net floor area, sometimes gross floor area or gross building area. In Denmark, for example, public spaces are counted in the floor area.

What kind of problems is this causing?

For example, if a bank is lending money to a customer, this can easily lead to a situation where different parties have different floor areas for a house/collateral, because the collateral applicant may have an incentive to do so. The difference can be remarkable and typically lender has the overestimated floor area. Therefore, it can lead to a situation where the total market value is overestimated and therefore, creating a credit loss.

How can you solve this? How can you fill the missing and wrong information?

We can fill the missing information and verify the validity of the data.

Jukka, with his expertise and knowledge, is finding innovative solutions to face property data so that anyone who is interested in investing in properties, is well informed, and we are very grateful for his brilliant work!

If you would like to know more about this subject or simply ask questions, don’t hesitate to drop a message at jukka@skenariolabs.com.