INFO201 Final Group Project Proposal
Final Project Website Link
Github Page
Copyright WRAL TechWire
Author:
- “Kaz Jiang”
- “Xiying(Nina) Zhang”
- “Yue(Ivy) Wu”
- “Bingyan Wang”
Project description:
The dataset we will be working with is the U.S. Cancer Statistics from 1999 to 2015, collecting information about all available diagnosed cancer cases and deaths for the whole U.S. population. This U.S. Cancer Statistics(USCS) is collected by Centers for Disease Control and Prevention(CDC). We will use this dataset to build some visualizations and look for the possible tendencies between cancer and other factors like geographical locations, genders, races and ages, etc.\ Our target audiences could be professional public health staff, as well as people who are interested in cancer statistics or suspicious about their own health condition. The latter one is the audience we put more focuses on, since our goal is to enable the general public to figure out their own propensity of cancer dangers.
- Possible questions answer for audience:
- How could cancer population be possibly affected by different states, years, and races?
- Which cancer types are the most prevalence ones in U.S? Is this different in different state/years?
- Does cancer statistics show any differences between male and female?
Technical Description
- In the final project we will use both API and csv in order to achieve our goal. From API, we will use
httrandjsonlitelibraries to parse the JSON data from api to a R data frame. From .csv, we will just use basic read.csv function to parse the data. - We will separate dataset to two major groups (mortality and incident) then sort the data in order to make visualize easier. Also, we will filter out the NA value in order to make dataset valid. Additionally, we will probably modify both API data and csv data in order to make a join.
- This time we will focus on libraries:
shiny,httr,jsonlite,ploty, and maybe some other libraries which will make the visualization more efficient. - The major challenges: Make a well format dataset to work with because most of cancer dataset are badly formatted. Make a beautiful visualization and also easy for client to understand. Make an attractive R based website.