Skip to content

JCMoses/JCMoses.github.io

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 

Repository files navigation

Data Scientist

Technical Skills: R, Python, Shiny, Azure

Education

  • B.E., Civil Engineering | University of Sydney (2006)
  • B.Sc., Geology and Geophysics | University of Sydney (2006)

Projects

Sales Forecasting on Email Analytics Data

For this project, the requirement was to provide a useful forecast of the future sales of an online store, to enable management to better plan inventories, staff requirements, and forecasted revenue. Daily data was used, which enabled good forecasts around sales and other marketing events. It also enabled management to see which days were more effective for sales and marketing events. Forecasting was conducted using a variety of different machine learning (including deep) algorithms, combining the best into weighted ensembles for greater accuracy. Data pipelines and ML pipelines were created to make the analysis repeatable and able to be updated with new data. The analysis was undertaken in R with some Python also used for the deep learning algorithms.

Sales Forecasting Projection

Email Lead Scoring with Forecasting

This project involved classifying leads into categories based on a forecast likelihood of purchasing in the future. The initial business case was to separate the leads who are more likely to purchase from others, but with additional data analytics, it was found that a different strategy would nurture the less likely leads into those who were more likely to make purchases. This would reduce churn significantly, increasing sales. This project required analytical analysis of data, classification using various machine learning algorithms, and forecasting of future sales under different conditions. Data and ML pipelines were created to enable the analysis to be revisited with new data periodically. The Analysis was undertake using Python.

Expected Value Plot

Stock Market Edge Analysis

Traders require “edge” in order to consistently make money in the stock market. This project involved looking for different sources of edge and confirming / disproving their continued existence in the market. Different assets were analysed and back-tested against known sources of edge. Comparisons were made against benchmarks (SPY, 80/20 portfolio) and each edge was assessed for improvement over the benchmarks, max. drawdown size, and other statistics. This analysis was undertaken using the R language.

Construction Project Risk Management Dashboard

This project was developed in order to better manage project-based risks for construction industry projects for a consulting engineering office. The major requirements were to enable a fast, easy way of adding, adjusting, importing and exporting risk assessments for projects. These should be easily visualised, with any areas of concern identified at a glance. The overall concentration of risks in different areas should also be visualised. It was also important that all risks could be easily exported to excel, and also imported so that details can be easily shared and incorporated quickly into the dashboard. The dashboard and analysis was written in R and Shiny.

Construction Project Risk Dashboard - Data Page Construction Project Risk Dashboard - Radar Plot

Customer Segmentation Project

Code

Imported the dataset from SQLite and analysed it to differentiate different kinds of customers using R. The business question for this project was "Can we predict if a customer will make a new purchase within 90 days?" A data-driven approach led to better characterization of the different customers by analysing different kinds of variables to assist the classification model. The project was written up using R Markdown to allow for the work to be properly documented and repeatable. The results were presented in a Shiny app, which allowed the user to view characteristics of each customer, along with the likelihood of them making a new purchase within 90 days. A user can also select customers based on a strong likelihood of a new purchase in that timeframe to view their profile.

Customer Segmentation App

Sales Prediction for next 90 days

Code

The business requirement for this project was clear: Predict weekly sales for the next 90 days for each of the datasets. This enables the business to prepare / order sufficient stock to meet the expected requirements. The goal is to improve ordering to reduce waste by over-ordering, and also to meet demand by ordering enough. This project was written in R and R Markdown and a report has been generated in html format outlining the process.

Sales Predictions

Work Experience

Freelance Data Scientist @ Jeffrey C. Moses EPU

  • Focus on solving business problems using data, not data or modelling for it's own sake.
  • Undertaking data collection, processing for analysis, programming machine-learning models, and deployment through APIs and/or Shiny Apps.
  • Undertaking supervised and unsupervised modelling, with a focus on time-series data.

Technical Project Manager @ VCE - Vienna Consulting Engineers ZT

  • Technical project management in the Structural Health Monitoring department. This involved planning, designing, and installing monitoring systems (short-term and permanent) for measuring movement, acceleration, strain, temperature, weather factors, and other data using sensors installed on bridges, buildings, wind turbines, and other structures.
  • Engineering design in infrastructure projects, including railway design projects.
  • Internal quality management auditor to ISO9001 (Quality), 14001 (Environmental), and 45001 (Work Safety) standards.
  • Member of the Integrated Management System team, and Risk Management team for the company.
  • Undertook planning and design of the testing system for factory (FAT) and on site (SAT) acceptance testing processes.
  • Undertook planning of work processes, including deployment to site and site safety. Responsible for producing planning and safety documentation, including risk assessments.
  • Undertook tendering for proposals, resource planning, timeline and cost planning.

Indepenent / Freelancer @ Varied Clients

  • Part-time work on varied projects with varied clients, including Alinea Partners, Impact Hub Vienna, and others.
  • Data analysis, preparation of dashboards, and reporting
  • Undertook customer satisfaction surveys on behalf of clients
  • Undertook secret shopping for clients, with full reporting and analysis

Engineering Planner and Structural Engineer @ Waagner-Biro Stahlbau AG

  • Worked in the Steel and Glass Department, undertaking planning and design of engineering structures, tender planning and calculation, and design of steel temporary works.
  • Undertook structural design and planning of building facades, bridges, and steel building frames.
  • Worked closely with clients and architects to 'trial' design ideas and determine viability.
  • Undertook reporting and documentation of work.

Bridge Design Engineer @ AECOM Australia Ltd

  • Worked on planning and designing bridges and other structures. This included concept designs with iterative changes, to full structural design and analysis within projects and for project tenders.
  • Worked as an internal Quality Auditor to ISO9001 (Quality) and 14001 (Environment), as well as a project and project-team manager.
  • Undertook design and planning of prestessed concrete, as well as steel bridges and structures.
  • Undertook condition assessment of existing structures, including bridges and tunnels. This included reporting and mapping of faults.
  • Undertook reporting on all design works, including all project documentation.
  • Worked closely with clients and was involved with presentation and reporting of project documentation and progress.
  • Responsible for project budges and timelines.
  • Undertook site-based supervision of construction works.

Planning and Design Engineer @ Cansult Maunsell AECOM

  • Deployment to a project site office, Al Raha Beach Development, Abu Dhabi, UAE.
  • Responsible for the planning and structural design of 2 prestessed concrete overpass bridges.
  • Undertook site supervision for 11 prestressed concrete bridges over canals.
  • Undertook planning and structural design of concrete elements within 7 planned metro stations within the project development.
  • Undertook planning, strucutral design, and proof-engineer calculations for concrete building structures, as well as marine structures.

Planning and Design Engineer @ Maunsell (AECOM) Australia Ltd

  • Interdisciplinary work on varied projects within the design office.
  • Worked on structural design and planning of concrete structures, civil infrastructure planning, environmental planning and reporting, and traffic engineering.
  • undertook design of stormwater, freshwater, and sewer systems for greenfields housing developments. Also designed local roads and cycleways within the developments.
  • Undertook Traffic surveys, then data analysis of the results. These were then reported and presented, and conclusions used to plan future road infrastructure.
  • Undertook project management work, responsible for budgets and planning timelines.

About

Project Portfolio

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published