Open ChristinaB opened 4 years ago
Stage 1: what can we do NOW with what we have NOW?
[ ] Review Connector documentation and software and outputs with WHW demo dataset
[ ] Identify primary audience within those interested in Kaggle tasks (at UW?) who want to find the humans and data and code connections between the journal publications in the kaggle project
[ ] Run Connector for a new audience focused on COVID-19 research
[ ] Solicit inputs on a process to create WHW 2020 cohort dataset for visualization
See Existing Kaggle task submissions for
What has been published about information sharing and inter-sectoral collaboration?
COVID-19 Open Research Dataset Challenge (CORD-19)
Paul Mooney · 14 Submissions
Task Details What has been published about information sharing and inter-sectoral collaboration? What has been published about data standards and nomenclature? What has been published about governmental public health? What do we know about risk communication? What has been published about communicating with high-risk populations? What has been published to clarify community measures? What has been published about equity considerations and problems of inequity?
Specifically, we want to know what the literature reports about:
Methods for coordinating data-gathering with standardized nomenclature. Sharing response information among planners, providers, and others. Understanding and mitigating barriers to information-sharing. How to recruit, support, and coordinate local (non-Federal) expertise and capacity relevant to public health emergency response (public, private, commercial and non-profit, including academic). Integration of federal/state/local public health surveillance systems. Value of investments in baseline public health response infrastructure preparedness Modes of communicating with target high-risk populations (elderly, health care workers). Risk communication and guidelines that are easy to understand and follow (include targeting at risk populations’ families too). Communication that indicates potential risk of disease to all population groups. Misunderstanding around containment and mitigation. Action plan to mitigate gaps and problems of inequity in the Nation’s public health capability, capacity, and funding to ensure all citizens in need are supported and can access information, surveillance, and treatment. Measures to reach marginalized and disadvantaged populations. Data systems and research priorities and agendas incorporate attention to the needs and circumstances of disadvantaged populations and underrepresented minorities. Mitigating threats to incarcerated people from COVID-19, assuring access to information, prevention, diagnosis, and treatment. Understanding coverage policies (barriers and opportunities) related to testing, treatment, and care Expected Submission To be valid, a submission must be contained in a single notebook made public on or before the submission deadline. Participants are free to use additional datasets in addition to the official Kaggle dataset, but those datasets must also be publicly available on either Kaggle, Allen.ai, or Semantic Scholar in order for the submission to be valid. Participants must also accept the competition rules.
Evaluation Submissions will be scored using the following grading rubric:
Accuracy (5 points) Did the participant accomplish the task? Did the participant discuss the pros and cons of their approach? Documentation (5 points) Is the methodology well documented? Is the code easy to read and reuse? Presentation (5 points) Did the participant communicate their findings in an effective manner? Did the participant make effective use of data visualizations? Timeline Submissions will be evaluated in 2 rounds: Round 1: Submission deadline is April 16, 2020 at 11:59pm UTC. Task Submissions which the evaluation committee deems to meet the threshold criteria will be awarded prizes in this initial round. Round 2: Submission deadline is June 16, 2020 at 11:59pm UTC. The hosts may add prize-eligible tasks in round 2. We may also re-award prizes for existing tasks whose submissions have advanced from the prior award, as judged by the evaluation committee. By each of these submission deadlines, you must have accepted the rules and shared your notebook submission publicly to be considered for prizes in each respective round
Prizes Kaggle is sponsoring a $1,000 per task award to the winner who is identified as best meeting the evaluation criteria. The winner may elect to receive this award as a charitable donation to COVID-19 relief/research efforts or as a monetary payment. Some details: Prize-eligible tasks are defined and specified as such on the task by the sponsor and hosts. A team or user is eligible to receive awards for multiple tasks. If the team consists of multiple users, they may elect to divide the monetary prize equally or donate their entire award. Prizes will be distributed in 2 rounds, per details contained in the timeline. Evaluation will be judged by a committee consisting of subject matter experts. All judgments are considered final. Kaggle reserves the right to adjust the prize distribution schedule if needed.
add how to details that are generic for an organizational effort as well as specific to our project goals as they evolve. useful for any open source project management.