Derek Owens-Oas

I developed features, a learning algorithm, and web app visualization for topics and connections in an online social network. The R and Python implementations are available. Blog posts, Facebook comments and messages, Twitter tweets, and courtroom transcriptions are among the communication modes analyzed.

Here is a link to the video showing this application:

https://drive.google.com/file/d/1-Goo7OjKdGs9cvYxDfAu58GUuzDNSQg3/view?usp=sharing

A statistical report and web application to evaluate water pipe quality in DC, Dallas, and Howard County. I wrote the code by applying machine learning algorithms to estimate the probability of each pipe breaking in the next three years, along with visualizing results on an interactive map.

I developed an interactive map giving estimates of water quality between sensor locations. I used machine learning, optimized linear predictive modeling, and spatial statistics along with writing a technical report which detailed the patterns of algae blooms.

I set up a free, simple, full-stack server for hosting a web app. The link is to an online, interactive box-plot visualization that enables exploring health procedure costs. Insurance claims data are used to show how expenses are distributed to the insurer, provider, and patient.

A second bar-graph version allows the user to mouse-over various procedural choices for treating osteoarthritis. Here is the link:

https://drive.google.com/file/d/10gVQWka51w0RA5wmO4_BPIeEt3nt-ZRr/view?usp=sharing

A healthcare provider can view the patient outcomes to guide the choice going forward.

I designed, implemented, and authored a publication which applies a statistical learning algorithm to political blog post data. A latent group that provides commentary on sensational crime is identified. The results are published in the Journal of Classification.

I contributed to and applied a dynamic programming algorithm to an election day mega-thread on Reddit and courtroom transcriptions. This method is a credit attribution method like those used in web advertising.

I created a WordPress web blog where I've written posts sharing my experience during my Ph.D. program in statistics and as an early-career data scientist and quantitative consultant. These include graphics I've created and discussion of industries of interest.

I used a tutorial to visualize CO2 emissions data by countries in years. One graphic shows the amounts on a world map, and the other is a time series plot. It's possible to subset portions geographically and to mouse over and get specific observational values.

SureTint Technologies software LaRu enables beauty salons to record customer hair formulas.

I continued developing application which converts formulas from one product line to another. Data are on AWS, code is Python, and a statistical model was used.

Features developed include a filter to ensure products conform to manufacturer recommendations.