This weekend I am taking stock of where I am - and juggling projects.
Yesterday I had a kick-off meeting with Julian House, a charity working with the homeless and housing issues based in Bath. I will be working with them on a pro-bono basis, to see whether they can improve the way they use data. This will be a learning experience for me, as I will be using Talend and working with Salesforce, both for the first time. Homelessness is an important issue here in the South West, and while I might not have money to donate, I can help with my time and knowledge. I will blog about some of the aspects of the project as we go along.
I am also working on a Kaggle competition, predicting whether a company should send direct mail to customers. This has a lot of data - the training set is over 140,000 records, with 1934 fields. This weekend I intend to use Talend to convert the string fields to integer, and then run principle components analysis, to see whether I can reduce the size of the dataset... more on that later.
