Kenneth G. Wood

Intro

I am a seasoned data scientist who selects, trains and tunes machine learning algorithms (linear/logistic regression, classifiers, neural networks) to transform structured and unstructured data into actionable insights and 'stories'. The results and value I produce typically drive and shape business strategies and product roadmaps for cloud-based and SaaS software technology solutions. I am proficient with a number of data analysis and programming tools such as Python (including Pandas, NumPy, SciKit-Learn & TensorFlow), R, SQL, Excel and Tableau.

Work

An API where I train a large-language model (LLM) to convert an audio file into a set of meeting minutes (including action items) generated in Markdown.
USING GRADIO AND META-LLAMA-3.1-8B-INSTRUCT LLM TO CREATE MEETING MINUTES FROM AUDIO FILE

A machine learning project where I used AWS Sagemaker to perform principle component analysis (PCA) and trained a XGBoost classifier to predict the presence of heart disease.
PREDICTING HEART DISEASE USING PCA FOR DIMENSIONALITY REDUCTION AND XGBOOST CLASSIFICATION IN AWS SAGEMAKER AND SCI-KIT LEARN

A machine learning project where I used Python's 'TensorFlow' package to train a 100 node recurrent neural network to make predictions on time-series data.
TRAINING AN RNN IN TENSORFLOW TO PREDICT TIME-SERIES DATA

In this project, I leveraged a machine learning algorithm called Markov Chains to describe a sequence of possible marketing events in which the probability of each event depends only on the state attained in the previous event. In the context of channel attribution, Markov Chains provide a framework to model user journeys and how each channel factors into the users traveling from one channel to another to eventually purchase (or not). The underlying concept of Markov chains is that customer journey data can be processed to identify the probabilities of moving from one event to another in a network of potential marketing channel events and conversion events.
MARKETING CHANNEL ATTRIBUTION USING MARKOV CHAINS

A self-produced video that presents the key findings of a comprehensive marketing data analysis project that used SQL queries to extract relevant data from a multi-location database. A majority of the subsequent data analysis (i.e., number “crunching”) was performed in Excel (using the 'Solver' optimizer) and the results were presented using Tableau, a powerful data visualization platform.
INCREASING PROFITS BY HARNESSING DATA ANALYTICS

KEN'S GITHUB SITE

KEN'S RPUBs (R PROGRAMMING) SITE

Resume

Download a copy of Ken's current resume.
Resume

Social

• LinkedIn
• GitHub
• R
• E-mail

Elements

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';