Data scientist with extensive skills in NLP and time series
Data Scientist2-5 Year Experience
Statistical Analysis Expert2-5 Year Experience
Data Science fellow
• Launched an interactive web-app that applies a data-driven approach to rank the most likely successful oil and gas industry participants based on the descriptions users provide for their upcoming projects.
• Developed a customized neural-net ensemble model that actively learns weights from each sub-model, which reduced over-fitting and increased f-score by 8% compared to neural net and random forest sub-models.
• Improved accuracy of ranking order of successful contractors from 4.6% (educated guess based on imbalance in historical data) to 67% (TFIDF fed to NLP/RF models) using text mining on notes from government regulators.
Full Professional Proficiency