Difference between revisions of "User:Jdirkman"
From REU@MU
(→Week IX) |
m (1 revision) |
(No difference)
| |
Latest revision as of 06:25, 12 June 2015
Contents
Jesse Dirkman
Personal Info
Jesse Dirkman, an undergraduate student at Marquette University, is majoring in mathematics and sociology. She will be graduating in May 2013. Jesse will be working with Dr. Corliss in the GasDay lab, focusing on water demand prediction models.
Research Objectives and Milestones
Research Questions
- How can we predict [water usage?] What factors are important? What sort of model is a best fit?
- Data cleaning/data quality issues
- What could be the use of such predictions?
- How can we turn this into a business model/ something useful?
Milestones/Weekly Plan
Week 1:
- Orientation
- Read about forecasting
- MATLAB
- Machine Learning
- Develop Milestones/goals
Week 2:
- Learn MATLAB application to GasDay data
- Graphs from data (using MATLAB)
- Neural Networks
- PRESENTATION: What can we learn from WATER data?
- Water forecasting research
- Intro to paper, what are people doing,
Week 3:
- Begin water forecasting
- What is the water data we’re looking at?
- Look into H20 Score or similar businesses
- Where can we get data from?
- What do industrial and commercial customers want?
Week 4:
- Continue water forecasting research
- Water forecasting
- Work on paper
Week 5:
- Continue water forecasting research
- Water forecasting
- Work on paper
- Mini presentation (REU)
- GasDay Presentation
Week 6:
- Continue water forecasting research
- Water forecasting
- Work on paper
Week 7:
- Determine business plan/proposal
- Continue water forecasting research
- Water forecasting
- Work on paper
Week 8:
- Rough draft of paper
Week 9:
- Final edit of paper
- Forecasting and modeling
Week 10:
- Present a business plan/proposal utilizing water use forecasts
- Turn in paper
Weekly Log
Week I
Tuesday
- REU Orientation
- Campus Tour
- E-Limo Tour
Wednesday
- GasDay Orientation
- Attended GasDay Grad Student Presentations
- Library Resources Tour
Thursday
- MATLAB Tutorials
- Reading: Water Demand Forecasting Articles
- Talk: Good Research Practices
Friday
- Readings: Forecasting and Water Predictions
- Met with Dr. Corliss
- Created Milestones and Objectives
Week II
Monday
- GasDay Camp
- Met with Dr. Corliss, discussed neural networks
- Worked on presentation for Wednesday
Tuesday
- GasDay MatLab Tour
- Worked on presentation for Wednesday
- Worked on modelling gas data
- Read about water forecasting
Wednesday
- Worked on presentation
- Presented for GasDay lab
- Read Water Forecasting
Thursday
- Worked on preparing graphs for WaterTech meeting
- REU Working Lunch
- Read from Water Forecasting
- Read from Urban Water Demand Management and Planning
Friday
- GasDay Camp
- Met with WaterTech USA representatives re: water forecasting
Week III
Monday
- Read from Urban Water Demand Management and Planning
- Read from Water Forecasting
Tuesday
- Attended "Good Presentation vs. Good Paper" talk
- Read from Hybrid Water Demand Forecasting Model Associating Artificial Neural Network with Fourier Series.
- Met with Dr. Corliss
Wednesday
- Finished reading Hybrid Water Demand Forecasting Model Associating Artificial Neural Network with Fourier Series.
- Prepared WaterTech graphs
Thursday
- Attended MatLab Seminar
- REU Working Lunch
Friday
- Database Design workshop with Dr. Corliss
- Met with H2Oscore
Week IV
Monday
- H2Oscore staff meeting
- Data exploration with H2Oscore
- Met with Dr. Corliss and McGee Young
Tuesday
- Read A Model of Daily Municipal Water Use for Short-Term Forecasting
- Cleaning H2Oscore data
Wednesday
- MatLab coding and modelling of Monona Water Data
- Worked on REU Mini-Presentation
Thursday
- REU student lunch
- Cleaning H2Oscore data
Friday
- Met with Dr. Corliss
- Introduction to Bloomberg software
- Extracted daily temperature and precipitation readings from Bloomberg
Week V
Monday
- Created Mini Presentation
- Created graphs of Monona Water Usage
Tuesday
- Met with McGee Young
- Mini Presentations
Wednesday
- Meeting with Dr. Corliss
- Intro to LaTeX
Thursday
- Created model for fake water demand data, MatLab models
- Worked on LaTeX
- Met with Dr. Corliss
Friday
- Met with McGee Young
- Met with Dr. Corliss
- Attended Mike Ziwisky's thesis defense
- Attended Dr. Adam Welc's research presentation (Oracle Labs)
Week VI
Monday
- Attempted exporting water data files
- H2OScore weekly staff meeting
- Met with Dr. Corliss, discussed research poster format
- Met with McGee Young
Tuesday
- Worked on learning command prompts
- Read from Programming Perl (Wall, Christiansen and Schwartz)
- Perl tutorial
- Worked on creating/understanding unzip program
Wednesday Independence Day
Thursday
- REU Working Lunch
- Met with Dr. Corliss
- First steps in creating MySQL database
- Read from Handbook of Writing for the Mathematical Sciences (Nicholas J. Higham)
- Met with McGee Young
Friday
- Created main categories in MySQL database
Week VII
Monday
- H20Score Weekly Meeting
- Met with Dr. Corliss
- Installed MATLAB Database Toolbox, created MATLAB/MySQL connection
Tuesday
- Worked on MATLAB database program, insert function
- Met with Dr. Corliss
Wednesday
- Worked on programs to import .csv files into MySqL database (Weather, Whitewater, and Monona)
Thursday
- Finished program to import .csv files into MySQL database
- Created program to extract data from MySQL database
- Met with Dr. Corliss
Friday
- Created program to extract data from MySQL database
- Successfully inserted Monona data into MySQL database
WEEK VIII
Monday
- H2OScore Weekly Meeting
- Ran code to insert Whitewater data into MySQL
- MatLab troubleshooting
- Successfully inserted (a portion of) Whitewater data into database
Tuesday
- Continued inserting Whitewater data into MYSQL database
- Continued troubleshooting
- Read Negotiating everyday life: The role of energy and water consumption feedback
- Read A Model of Daily Municipal Water Use for Short-Term Forecasting
Wednesday
- Continued inserting Whitewater data into MySQL database
- Read Monitoring Gas and Water Consumption through ICTs for Improved User Awareness
- Worked on paper
- Attended GasDay Seminars
Thursday
- First graphs of water usage data
- Working lunch
- Preliminary poster work
- Worked on paper outline
Friday
- Worked on Poster
- Created graphs using one week of Whitewater data, analysis
- Met with McGee Young
- MatLab coding to automate process of filling in water usage, preparing for regression analysis
Week IX
Monday
- H2O Score Weekly Meeting
- Began filling water usage tables
- Worked on poster
Tuesday
- Worked on Poster
- Finished filling water usage tables
- Ran iterations of multiple linear regression
- Created graphs of water consumption by day of week
Wednesday
- Worked on Poster
- Created Presentation for GasDay
- Presented for GasDay
- Met with Dr. Corliss
Thursday
- Finish Poster
- Research for paper
- REU Working Lunch
- Met with Dr. Corliss
Friday
- Ran query for McGee Young
- Worked on research
