Decided to try to apply the data mining techniques learnt from Intelligent systems course on publicly available economic data. The two sources for data that I will start of with are:
- World Bank (http://data.worldbank.org/indicator?display=graph)
- Human Development Reports (http://hdr.undp.org/en/statistics/)
- Google public data (http://www.google.com/publicdata/directory)
I will run a mix of supervised and unsupervised techniques. When conducting supervised analysis I will look for relationships between economic indications to provide inference on discussion topics such as:
- The value of high equality in an economy
- Benefits of non-livestock or livestock agriculture
- Gains through geographic decentralization of population
- Estimates on sustainable housing price ranges
- The value of debt
- Productivity of information technology
- Cost and benefits of lax/tight immigration policies
- Cost and benefits free-market/regulated/centralized economic governance
Techniques used for quantitative analysis will be varied a dependent on subsequent success. To start with I plan on using the raw data sources in conjunction with some simplistic python and Java scripts. If that turns out to be ineffective I will work with MatLab, Weka and Netica. Google and the World Bank also have numerous interfaces for exploring data. This will be an ongoing project so I will make these posts help myself keep track of progress.
