Table of Contents
- Deliverables and Grade Breakdown
- Report Format and Submission
The graduate project is offered only to students enrolled in Data C200 or CS C200A. Other students are welcome to explore the questions and datasets in the project for learning, but their work will not be graded or counted towards their final grades.
The purpose of the project is to give students experience in both open-ended data science analysis and research in general. In this project, you will work with one or any combination of the following datasets provided to you to explore research questions that you define.
Note: in addition to the general guideline, each option has its own set of additional requirements for Report Format and Submission. Be sure to consult the correct section for your project option.
You will receive feedback from peer grading before the final deadline, and you are expected to incorporate the feedback into the final report and presentation. You will be graded on both the final report and presentation, as well as deliverables before the submission of the final reports, including your peer reviews.
Teamwork: You can work alone or in a group with at most two other students. If you are interested in working with others, we will have a Piazza post for teammate search. Everyone in the same group will receive the same grade. The group size will be taken into consideration for grading.
|Date (by EOD at 11:59pm)||Event / Deliverable|
|4/11||Research proposal and project groups due|
|4/30||First draft of final report due|
|4/31||Peer grading starts|
|5/4||Peer grading due|
|5/9||Revised final report due|
|5/12||Presentation video due|
|5/14||Presentation video released (at discretion)|
|Research proposal and project groups||10%|
|Submission of first draft||10%|
|Final report: analysis notebook||20%|
|Final report: project writeup||30%|
|Final presentation video||15%|
This section contains the datasets we will provide to you to explore your research questions. Please note that:
- You must incorporate at least one of the provided datasets.
- You are welcome to bring in additional datasets to complement the datasets provided here, but you must cite the sources and clearly describe the content of any additional data you use in the final report.
Please be sure to consult the references on causal inference for guidance on how to work with multiple datasets if you decide on doing that.
All the datasets (or links to access and download them) provided by us can be found inside the graduate project dataset link on DataHub. If you wish to work on the project locally, you can also download the zip files containing the datasets for each topic.
The following subsections contain the descriptions and additional requirements for each dataset.
This dataset contains US reports on COVID-19 testing and cases from the COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University and CDC (Centers for Disease Control and Prevention). You can access all the data here within
Dataset_A.zip on DataHub:
csse_covid_19_daily_reports_us.csvcontains US daily reports (documentation)
cdc_death_counts_by_sex_age_state.csvcontains US weekly reports on deaths involving COVID-19, pneumonia, and influenza reported to NCHS by sex, age, group, and state. (documentation)
cdc_death_counts_by_conditons.csvcontains US weekly reports on health conditions and contributing causes mentioned in conjunction with deaths involving COVID-19. (documentation)
You must choose to work with at least 2 of the reports above in your analysis.
This dataset contains reports from the Household Pulse Survey launched by NCHS in partnership with the U.S. Census Bureau; it focuses on how COVID-19 has affected survey correspondents’ mental health and their access to health care. In addition, it provides statistics on usage of telemedicine by healthcare providers. You can access all the data here within
Dataset_B.zip on DataHub:
nchs_covid_indicators_of_anxiety_depression.csvcontains survey estimates of responses to questions that are indicators of anxiety or depression based on reported frequency of symptoms within the past week. (documentation)
nchs_covid_mental_health_care.csvcontains survey estimates of responses to questions that ask if participants have accessed mental health care in the past 4 weeks. (documentation)
nchs_covid_health_insurance_coverage.csvcontains survey estimates of responses to questions that ask about participants’ health insurance coverage. (documentation)
nchs_covid_reduced_access_to_health_care.csvcontains survey estimates of responses to questions that ask if participants have experienced delay or been refused health care due to COVID-19. (documentation)
nchs_covid_telemedicine_usage.csvcontains survey estimates of responses to questions that ask if healthcare providers offered telemedicine (including video and telephone appointments) – both during and before the pandemic – and about the use of telemedicine during the pandemic. (documentation)
You must choose to work with at least 3 of the reports above in your analysis.
This dataset contains (in full-text and metadata form) scholarly articles related to COVID-19. The data are optimized for machine readability and made available for use by the global research community. The dataset is intended to mobilize researchers to generate new insights from the articles in support of the fight against this infectious disease. You can access the link to obtain the data here within
Dataset_C.zip on DataHub:
covid_open_research_dataset.txtcontains the link that will guide you to obtain the full-text and metadata dataset of COVID-related research articles. (documentation)
This dataset contains some general statistics and measurements of various aspects of the climate and the environment. You can access all the data here within
Dataset_A.zip on DataHub. It includes the following reports:
daily_global_weather_2020.csvcontains data on daily temperature and precipitation measurements. To learn how to use the data from this file, please read the following section on the first report.
us_greenhouse_gas_emission_direct_emitter_gas_type.csvcontain data reported by EPA (Environment Protection Agency) on greenhouse gas emissions, detailing the specific types of gas reported by facilities and general information about the facilities themselves. The dataset is made available through EPA’s GHGRP (Greenhouse Gas Reporting Program).
us_air_quality_measures.csvcontains data from the EPA’s AQS (Air Quality System) that measures air quality on a county level from approximately 4000 monitoring stations around the country. (source)
The following subsection contains more details on how to work with the first report on global daily temperature and precipitation:
The first report on daily temperature and precipitation is measured by weather stations in the Global Historical Climatology Network for January to October 2020.
The data in
daily_global_weather_2020.csv is derived from the source file at https://www1.ncdc.noaa.gov/pub/data/ghcn/daily/by_year/2020.csv.gz.
To help you get started with a dataset of manageable size, we have preprocessed the GHCN dataset to include only the average temperature and precipitation measurements from stations that have both measurements. Each row in the preprocessed dataset contains both the average temperature and precipitation measurements for a given station on a given date.
If you wish to explore the climate data for a different year, you can use the
GHCN_data_preprocessing.ipynb notebook to download and perform the preprocessing described above. Please be advised that depending on the dataset size for a given year,
GHCN_data_preprocessing.ipynb may not run on DataHub. We will not be providing infrastructural support for running the notebook, but you are welcome to run it on a different machine you have access to or ask a GSI to dump the data for you.
The data contains only the (latitude, longitude) coordinates for the weather stations. To map the coordinates to geographical locations, the reverse-geocoder package mentioned in the References section might be helpful.
This dataset contains studies focused specifically on the impact of environmental and climate changes on biodiversity and the local ecosystems. You can access all the data here within
Dataset_B.zip on DataHub. It includes the following reports:
bioCON_plant_diversity.csvcontains data collected as part of an ecological experiment, BioCON (Biodiversity, CO2, and Nitrogen), that started in 1997 and focused on studying biodiversity within the plant species at Cedar Creek Ecosystem Science Preserve. (documentation)
plant_pollinator_diversity_set2.csvcontain ecological data collected from a long-term observation study from 2011 to 2018 that focuses on plant-pollinator interaction and its impact on local biodiversity. (documentation)
national_parks_biodiversity_species.csvcontain data published by the National Park Service on animal and plant species identified in individual national parks.
This dataset contains a set of reports from pioneering researches that explore the outer space. Much of the data from these studies have provided a rich foundation for a variety of large-scale research projects that explore widely discussed topics such as habitable exoplanets or search for extraterrestrial life.
You can access all the data here within
Dataset_A.zip on DataHub. It includes the following reports:
kepler_exoplanet_search.csvcontains data collected by NASA from the Kepler Space Observatory as part of a long-term study on finding habitable exoplanets from over 10,000 candidates. (source)
kelper_planetary_system_composite.csvcontains data collected by NASA from the Kelper Space Observatory as part of an ongoing study that tabulates all confirmed planetary systems outside the solar system. You are encouraged to use the composite data in conjunction with the exoplanet search results above. (source)
nasa_neows.csvcontains data collected from NASA’s NeoWs (Near Earth Object Web Service) that collects information on near earth asteroids.
A recommender system is an information filtering system that focuses on predicting the preference a user would give to an item by predicting its rank; it is used in a variety of areas, such as search engines, online shopping platforms, etc. This dataset contains a set of reports on various tools using a recommender system.
You can access the links to obtain all the data here within
Dataset_B.zip on DataHub. It includes the following reports:
fitness_recommendation.txtcontains a link to access the fitness data from sequential sensors for various workouts. (documentation)
amazon_reviews.txtcontains a link to access the data on a subset of Amazon product reviews. The report includes metadata such as ratings and text on the reviews and general information about the product. (documentation)
The project submission should include the following two components.
[Component 1]. Analysis Notebooks. The Jupyter Notebook(s) containing all the analyses that you performed on the datasets to support your claims in the narrative notebook. Make sure that all references to datasets are done as
data/[path to data files]. You can copy the datasets from
data/ at the top-level directory for your project on DataHub to do this.
Your analysis notebook(s) should address all of the following components in the data science lifecycle. Please note that a thorough explanation of your thought process and approach is as important as your work. We have provided a few preliminary questions/tips you can think about for each part:
- Data Sampling and Collection
- How were the data collected?
- Was there any potential bias introduced in the sampling process?
- Data Cleaning
- What type of data are you currently exploring?
- What is the granularity of the data?
- What does the distribution of the data look like? Are there any outliers? Are there any missing or invalid entries?
- Exploratory Data Analysis
- Is there any correlation between the variables you are interested in exploring?
- How would you cleanly and accurately visualize the relationship among variables?
- Data Modeling and Inferences
- Please note that the following datasets have a data modeling requirement, i.e. you need to utilize at least 1 machine learning model we teach in this class in your project: Topic 1 - Dataset A, Topic 1 - Dataset C, Topic 2 - Dataset A, Topic 3 - Dataset A, Topic 3 - Dataset B. For datasets not mentioned above, you are welcome to continue building machine learning model(s). Otherwise, we will be placing more emphasis on the inference part instead.
- Here are a few components your notebook must address:
- What type of machine learning problem are you investigating?
- What model do you plan on using and why?
- Does your model require hyperparameter tuning? If so, how do you approach it?
- How do you engineer the features for your model? What are the rationales behind selecting these features?
- How do you perform cross validation on your model?
- What loss metrics are you using to evaluate your model?
- From a bias-variance tradeoff standpoint, how do you assess the performance of your model? How do you check if it is overfitting?
- How would you improve your model based on the outcome?
- If you are choosing to pursue your research question from an inference angle, your notebook must demonstrate sufficient analysis and visualization to support your conclusion. We will not restrict you to the type of analysis as there are many different statistical techniques that may apply to your case. However, we also ask that you provide detailed justification for the techniques you choose and how it allows you make those inferences.
[Component 2]. Project Writeup (previously narrative notebook). This is a single PDF that summarizes your workflow and what you have learned. It should be structured as a research paper and include a title, list of authors, abstract, introduction, description of data, description of methods, summary of results, and discussion. Make sure to number figures and tables and include informative captions.
If you wish, you can render the PDF using LaTeX, provided that the provenance of the figures is clearly labeled in the main narrative, and the figures can be reproduced by running the analysis notebooks
Specifically, you should address the following in the narrative:
- Clearly stated research questions and why they are interesting and important. You must include at least one research question involving at least one or more datasets from one of the topics we provided, but you may include additional research questions about each individual dataset. At least one of your research questions has to include a modeling component, e.g., can we build a model using climate data to predict growth in COVID-19 cases accurately?
- A brief survey of related work on the topic(s) of your analysis and how your project differs from or complements existing research.
- If applicable, descriptions of additional datasets that you gathered to support your analysis.
- Methodology: carefully describe the methods you use and why they are appropriate for answering your search questions. It must include
- a brief overview of causal inference, which should be written in a way such that another student in Data 100 who has never been exposed to the concept can carry out the analyses involving the datasets in your project.
- a detailed description of how modeling is done in your project, including inference or prediction methods used, feature engineering and regularization if applicable, and cross-validation or test data as appropriate for model selection and evaluation.
- Interesting findings* about each dataset when analyzed individually. Include visualizations and descriptions of data cleaning and data transformation necessary to perform the analysis that led to your findings.
- Interesting findings* involving your datasets. Include visualizations and descriptions of data cleaning and data transformation necessary to perform the analysis that led to your findings.
- Analysis of your findings to answer your research question(s). Include visualizations and specific results. If your research questions contain a modeling component, you must compare the results using different inference or prediction methods (e.g., linear regression, logistic regression, or classification and regression trees). Can you explain why some methods performed better than others?
- An evaluation of your approach and discuss any limitations of the methods you used.
- Describe any surprising discoveries that you made and future work.
* Examples of interesting findings: interesting data distributions and trends, correlations between different features, the relationship between the data distribution for the general population and specific datasets (e.g., the gender distribution in the census dataset vs. in the mental health dataset), specific features that are notably effective/ineffective for prediction.
The narrative notebook should include figures sparingly to support specific claims. It can include runnable components, but it should not have large amounts of code. The length of the report should be 8+/-2 pages when it is printed as a PDF, excluding figures and code.
Tip: if you need to write a large amount of $\LaTeX$, you may want to use the
%%latex cell magic.
Please submit everything as a zip file to the final report submission link. Please make sure the folder in the zip file has the following structure:
studentIDs/ data/[all datasets used] analysis/[analysis notebooks] narrative/[narrative notebook] figures/[figures included in the narrative notebook]
For groups with multiple members, please use student IDs joined by
_ as the name for the top-level directory. The analysis notebooks must be runnable within this directory structure. If the narrative notebook includes any figures that are created in the analysis notebooks, the figures should be saved to
figures/ by the analysis notebooks and imported from
figures/ by the narrative notebook.
Each group will peer grade the projects from another group. The review will be graded out of a total of 15 points.
Each review should include the following components:
- A summary of the report (5 points). The summary should address at least the following:
- What research question does the group propose? Why is it important?
- How does the dataset relate to the research question?
- What data modeling/inference techniques do the group primarily use to gain insights into their research question? Why are these techniques suitable for the task?
- What are the next steps a researcher can take if they want to investigate the question further based off the work in the project?
- An evaluation of the report based on the Data Science Lifecycle (10 points, 2 points per component). The review should include at least one strong point and one suggestion for improvement for each of the following components in the project:
- Data collection and sampling
- Data cleaning
- Exploratory data analysis (including data wrangling and visualization)
- Data modeling (including feature engineering, selection of the model, and evaluation of the model’s performance)
- Inference (do the results from the model sufficiently support the conclusion within the report?)
The analysis notebook will be graded out of a total of 20 points based on the following set of criteria:
|Code readability and documentation||5|
|Proper and sufficient utilization of Python libraries||5|
|Overall code quality||3|
|Replicability of the results||7|
The project writeup will be graded out of a total of 30 points based on the following set of criteria:
|Introduction, motivation, and presentation of the research question(s)||3|
|Exploratory data analysis||5|
|Modeling and inference techniques||7|
|Analysis of results||7|
|Implementation of peer review feedback||3|
|Discussion of potential societal impacts and/or ethical concerns||2|
|Overall clarity and structure of the report||3|
When studying the relationship between datasets, you might want to consult the following references on causality vs. correlation. Oftentimes, it is tempting to make claims about causal relationships when there is not enough evidence from the data to support such claims. Please review the following references, or other reputable references that you find on the topic to familiarize yourself with relevant concepts and methods.