For a couple years now, I have participated in the Data Science club at UCSB by taking part in the Project Group where I engaged in predictive modeling competitions held by kaggle.com. There, I've worked with others in teams of 3-4 to create predictive models by using powerful machine learning algorithms such as Linear Regression, Support Vector Machines, Decision Trees, Neural Networks, and more.
I've created a variety of Twitter bots (such as @get_analyzed and @get_mimicked) using the Twitter REST API, many of which utilize machine learning techniques such as sentiment analysis, markov chains, personality insights (IBM Bluemix), and more. For most of these Twitter bots, I've also implemented MySQL/PostgreSQL databases in the cloud using Amazon Web Services from which the scripts retrieve information vital to the bots' functions. These scripts automatically update their respective databases as more data needs to be stored.
Using various web scraping libraries in Python, I've written several web-crawlers that hop from site to site scraping and storing email addresses, tables, pieces of text, and/or datasets. For each of these crawlers, I've implemented a MySQL database into which the script stores all of the data in structured form.
Connect with me on one of my other sites!
Or shoot me an email at firstname.lastname@example.org