Survey results

This research was part of a PhD research named "Regularization methods that include experts’ domain knowledge for feature selection in a linear regression model" and research project "Methodological Framework For Efficient Energy Management By Intelligent Data Analytics" (Croatian Science Foundation project, IP-2016-06-8350).

For the purpose of including experts' domain knowledge in a feature selection method proposed in PhD thesis, four experts were examined during April 2022.  The experts were people with experience in energy efficiency and construction. They were asked to express their opinion on the influence of each input feature on the output feature.

input features:  25 characteristics of buildings such as the use and purpose of the building, structural and energy characteristics, heating and cooling characteristics, geographical and other (input characteristics).

output feature: annual energy cost of public sector buildings in the Republic of Croatia.

The expert's opinion is represented by an ordinal scale with 4 possible categories:  

1 - No influence, 

2 - Weak influence, 

3 - Medium influence, 

4 - Strong influence.

The repository consists of:

Description_of_features.xlsx - table that contains names and descriptions of features used in the research.

Survey_results.xslx - table that contains the results of experts' opinions on the influence of each input feature on the output feature.