CS 583: Project

Due: April 30th, 11:55pm

I. Assignment

In this project, we will work on the Covid-19 Q&A challenge:

https://www.kaggle.com/allen-institute-for-ai/CORD-19-research-
challenge#0072159e1ebecc889e9bcabb58bb45c47e18a403.json

There are 10 tasks:

https://www.kaggle.com/allen-institute-for-ai/CORD-19-research-challenge/tasks

You are free to choose one of them in Doodle

https://doodle.com/poll/8as8cpgxdz4d74v2

You can build a team of 1, 2, or 3 students to tackle the task you chose.

If you work in a group, please indicate all student names in the Doodle choice.

Build a system that automatically read literature in

https://pages.semanticscholar.org/coronavirus-research

and answer all the questions in your chosen task.

The output of your system should follow the evaluation guideline in the challenge.

Make use of open-source toolkits of Question/Answering and other DL resources.

II. Submission guideline

The code should be self-explainable.

Write very clear comments to your codes.

All experiments, including a readme file, source code, data, systems, and results, should be
zipped and submitted to the course instructor.

III. Assignment Policies

Collaboration and IP Policy. Do ask questions if anything is not clear. You can collaborate to
understand the material. Do not exchange code with other teams. Any solution that is similar or
the same with another submission will be penalized.

Late Policy. Late submission will be penalized on the final score.

Submission Deadline. April 30th
这是一门 course project. 要用 python的 jupyter notebook上面完成 code和解释的部分。要
结合 deep learning和 text mining的知识。

我选的是 task 10: What has been published about information sharing and inter-sectoral
collaboration?
https://www.kaggle.com/allen-institute-for-ai/CORD-19-research-challenge/tasks?taskId=583


这是导师给的建议。需要从这个方向来入手。

做 project的时候要能及时更新反馈给我这样我能及时去做 ppt。不会的要能给我解释，我再
询问导师。然后截至日期是这个月底。不要求 kaggle上面比赛的排名，但是提交内容不能和
kaggle上面的提交内容一样。要从上面截图老师给的建议方向入手。