Josh Jones, PhD

Data Science


Machine Learning
Anomaly Detection


Ph.D. Computer Science, Georgia Institute of Technology
B.S. Computer Science, University of New Hampshire



Recent Projects

Large Real Estate Analytics Company
Outlier Detection on Real Estate Data
Utilized unsupervised machine learning approaches on residential property data to detect anomalous real-estate features and outlier. Contributed to OSS development by improving a core outlier detection module. Developed models to feed predictions about real-estate properties and their residents.

Online Music Download Service
Predicting Hit Songs
Developed multiple machine learning models, used time series analysis techniques, and performed statistical analyses to predict which songs will be hits. This prediction was more timely than the Billboard charts. Built an interactive visual dashboard powered by the prediction engine to inform users and decision makers of the most up-to-date trending music.

Technical Expertise

  • Python data stack: scikit-learn, pandas, numpy, Jupyter, others
  • Apache Spark
  • Databases: PostgreSQL
  • Unsupervised Learning: Clustering, Outlier Detection, Anomalies
  • Supervised Learning
  • Time Series Analysis
  • Visualization