RStudio addins - first attemptRecently I found RStudio began to provide addin mechanism. The examples looked simple, and the addin API easy to use. I immediately started to try writing one by myself. It will be a good practice project for writing R package, and I can implement some features ... [Read more...]

SummaryThis is my second post on topic of Data Cleaning. Cleaning addresses format turned out to have a substantial positive impact on Geocoding performance. Deep understandings of address format standard is needed to deal with all kinds of special cases. [Read more...]

SummaryData cleaning is a cumbersome but important task for Data Science project in reality. This is a discussion on my practice of data cleaning for NYC Taxi Trip data. There are lots of domain knowledge, common sense and business thinking involved. [Read more...]

SummaryI discussed all the problem I met, approaches I tried, and improvement I achieved in the Geocoding task. There are many subtle details, some open questions and areas can be improved. The final working script and complete workflow are hosted in github. [Read more...]

SummaryThis post discussed the background, approaches, windows and linux environment setup for my Geocoding task. See more details about the script and workflow in next post. [Read more...]

SummaryThis is a write-up for my volunteer Data Science project for the American Red Cross. The project used public data to help Red Cross directing limited resources to homes that more vulnerable to fire risk and loss. My work in the project:Discovered the hidden information in NFIRS dataset, obtained ... [Read more...]

SummaryThis personal project was inspired by my own experience in housing market. I was wondering what I can achieve with public data and Data Science methods. Built upon extensive research of domain knowledge, my model is very simple yet powerful. It used the data with most information, followed the trends ... [Read more...]

SummaryThis is a simple python script that can measure python program running time in fine levels. It’s simpler than a full profiler, easier to use than other currently available similar scripts. [Read more...]