Difference between revisions of "User:Lauraschultz"
From REU@MU
Lauraschultz (Talk | contribs) (Created page with "== Weekly Log == === Week 1 (5/29/2018 - 6/4/2018) === * Attend REU orientation * Background reading on history of the project, previous research, API tools, and different clu...") |
Lauraschultz (Talk | contribs) |
||
| Line 3: | Line 3: | ||
* Attend REU orientation | * Attend REU orientation | ||
* Background reading on history of the project, previous research, API tools, and different clustering methods. | * Background reading on history of the project, previous research, API tools, and different clustering methods. | ||
| + | * Met to discuss background of project and go over previous work. | ||
* Do preliminary exploration of crime data. | * Do preliminary exploration of crime data. | ||
=== Week 2 (6/5/2018 - 6/11/2018) === | === Week 2 (6/5/2018 - 6/11/2018) === | ||
* Continue researching clustering methods. | * Continue researching clustering methods. | ||
| − | * | + | * Create presentation on the different families of clustering methods and the algorithms that fall within each. For each algorithm, discuss its steps and the points at which human interaction is required (potential biases). Next, perform the different clustering algorithms on the sample Milwaukee crime data that has geocodes available. |
| + | |||
| + | === Week 3 (6/12/2018 - 6/18/2018) === | ||
| + | * Add entries to Python dictionary so that court address data can be sanitized. | ||
| + | * Begin exploring agglomerative clustering, seperate data by type of crime, time, location. | ||
| + | |||
| + | === Week 4 (6/19/2018 - 6/25/2018) === | ||
| + | * Learn how to use the Google Maps API, get neighborhood data about each point. | ||
| + | * Use neighborhood data to create map of Milwaukee's neighborhoods and show which have a high proportion of non-guilty verdicts. | ||
| + | * Use verdict data to create heatmap showing which areas have a high proportion of non-guilty verdicts. | ||
| + | |||
| + | === Week 5 (6/26/2018 - 7/2/2018) === | ||
| + | * Continue exploring agglomerative clustering; cluster temporally to identify patterns. | ||
| + | * Create animated visualizations based on temporal data to show frequency of crime occurring at different times. | ||
| + | |||
| + | === Week 6 (7/3/2018 - 7/9/2018) === | ||
| + | * Read selected articles about machine learning, human-computer interaction, and algorithmic bias. | ||
| + | * Create presentation summarizing each article and explaining relevance to our project. | ||
| + | |||
| + | === Week 7 (7/10/2018 - 7/16/2018) === | ||
| + | * Continue reading articles and creating presentation. | ||
| + | * Continue exploring agglomerative clustering; perform trials to determine which proximity metric is best. | ||
| + | |||
| + | === Week 8 (7/17/2018 - 7/23/2018) === | ||
| + | * Create research poster. | ||
| + | |||
| + | === Week 9 (7/24/2018 - 7/30/2018) === | ||
| + | * Write first draft of final paper, send to Dr. Guha. | ||
| + | |||
| + | === Week 10 (7/31/2018 - 8/3/2018) === | ||
| + | * Present research at poster session. | ||
| + | * Create final slideshow presentation. | ||
| + | * Present research at MSCS REU presentations. | ||
| + | * Write final draft of research paper with Dr. Guha's feedback. | ||
Latest revision as of 19:22, 3 August 2018
Contents
- 1 Weekly Log
- 1.1 Week 1 (5/29/2018 - 6/4/2018)
- 1.2 Week 2 (6/5/2018 - 6/11/2018)
- 1.3 Week 3 (6/12/2018 - 6/18/2018)
- 1.4 Week 4 (6/19/2018 - 6/25/2018)
- 1.5 Week 5 (6/26/2018 - 7/2/2018)
- 1.6 Week 6 (7/3/2018 - 7/9/2018)
- 1.7 Week 7 (7/10/2018 - 7/16/2018)
- 1.8 Week 8 (7/17/2018 - 7/23/2018)
- 1.9 Week 9 (7/24/2018 - 7/30/2018)
- 1.10 Week 10 (7/31/2018 - 8/3/2018)
Weekly Log
Week 1 (5/29/2018 - 6/4/2018)
- Attend REU orientation
- Background reading on history of the project, previous research, API tools, and different clustering methods.
- Met to discuss background of project and go over previous work.
- Do preliminary exploration of crime data.
Week 2 (6/5/2018 - 6/11/2018)
- Continue researching clustering methods.
- Create presentation on the different families of clustering methods and the algorithms that fall within each. For each algorithm, discuss its steps and the points at which human interaction is required (potential biases). Next, perform the different clustering algorithms on the sample Milwaukee crime data that has geocodes available.
Week 3 (6/12/2018 - 6/18/2018)
- Add entries to Python dictionary so that court address data can be sanitized.
- Begin exploring agglomerative clustering, seperate data by type of crime, time, location.
Week 4 (6/19/2018 - 6/25/2018)
- Learn how to use the Google Maps API, get neighborhood data about each point.
- Use neighborhood data to create map of Milwaukee's neighborhoods and show which have a high proportion of non-guilty verdicts.
- Use verdict data to create heatmap showing which areas have a high proportion of non-guilty verdicts.
Week 5 (6/26/2018 - 7/2/2018)
- Continue exploring agglomerative clustering; cluster temporally to identify patterns.
- Create animated visualizations based on temporal data to show frequency of crime occurring at different times.
Week 6 (7/3/2018 - 7/9/2018)
- Read selected articles about machine learning, human-computer interaction, and algorithmic bias.
- Create presentation summarizing each article and explaining relevance to our project.
Week 7 (7/10/2018 - 7/16/2018)
- Continue reading articles and creating presentation.
- Continue exploring agglomerative clustering; perform trials to determine which proximity metric is best.
Week 8 (7/17/2018 - 7/23/2018)
- Create research poster.
Week 9 (7/24/2018 - 7/30/2018)
- Write first draft of final paper, send to Dr. Guha.
Week 10 (7/31/2018 - 8/3/2018)
- Present research at poster session.
- Create final slideshow presentation.
- Present research at MSCS REU presentations.
- Write final draft of research paper with Dr. Guha's feedback.
