INFT216讲解、辅导via iLearn、讲解R语言、R程序调试
辅导Web开发|辅导Web开发
INFT216 - Data Science 191
Project: Building a Data Science Dashboard for BondTelco
Assessment Value: 20%
Due Date: Mid Week 13, Wednesday 10th April 5pm (submit via iLearn)
Management at BondTelco are impressed with the models you have built so far. They have heard of
a few other Data Science modelling techniques, and will ask you to build models using those
techniques over the coming months.
They are also starting to think about deployment. At a recent conference on Data Science and Big
Data, management attended a seminar about ShinyApps, and they think this would be the perfect
deployment approach for their models. After all, Shiny allows for building a cloud-based app which
can be written and maintained in R, and accessed over the web by their salesforce.
You have already developed models using Decision Trees and Logistic Regression. There are 2 more
models you are yet to build. For this project, you only need to implement the tree, logistic
regression, and knn models.
Management want you to build a Data Science Dashboard. Essentially this will be a dashboard
which is built using Shiny. For the tree, logistic regression and knn models, the app will build the
models and for each model, it will show visualizations of the model and its fitness for purpose. The
app would then allow a user to input the specific details of the customer they are talking to, and
receive a prediction from each of the underlying models.
You have access to the same data as you used for the assignments.
Deliverables:
Your final deliverable will be your Shiny app. Zip the app folder and submit it online through iLearn.
Remember to include the web address of your application when you submit the project.
Guidance:
You can design the app any way you like. The most obvious implementation would be a tab of
panels on the main screen – one for each model type. You may also wish to investigate Shiny
Dashboards. I am particularly interested in visualization and useability. You should also take into
account any feedback on previous assignments to update your model functions.
Note:
As is the case with all assignments I set, if you do the minimum (correctly), then you will receive half
marks. Additional marks are awarded for those assignments where you have clearly put in
additional thought, whether it be in visualization, modelling, succinctness, or coding elegance.