In general, the CRISP$-$DM methodology should be applied and followed through the main phases of the lifecycle except the final deployment stage once the preparation is completed. Most generic tasks $($not all of them are applicable$)$ within each phase should be mapped to specific tasks meaningful to you own project.

You should consider an overall aim for your project upon which all data mining and data science activities are related to this aim.

Within the life cycle of the project, you may have regular discussions between the team members and make sensible decisions. Although you are working on your share of the work, exchanging $$good practice$$ is encouraged. Constructive criticism is always beneficial to each other.

$3.$ Project Deliverables

The deliverables of the project are in the form of a project report. Generally, you are expected to follow the CRISP$-$DM methodology and present your work according to the main phases of a data mining project lifecycle $[3].$ The report therefore should cover the following items:

a$)$ Business Understanding. This part consists of three essential elements: $($a$)$ a background study into the problem domain, project context and the purpose of the project work; $($b$)$ the project overall aim and business objectives as well as mapping from business objectives to potential data mining tasks; and $($c$)$ a comprehensive and critical literature review that consists of a broader general review of data mining used for solving the same or similar problems followed by a more detailed study of a piece of existing data mining work relevant to the aim of this project.

b$)$ Data exploration and understanding. This part of the report should describe the data set at hand in general, sum up any data characteristics identified through exploration of the data set, plus a critical evaluation of the data quality.

c$)$ Data Preparation and Pre$-$processing. This part of the report should present a set of preprocessing operations conducted in order to prepare the data for the modelling stage. The work conducted at this stage should be backed up by valid reasons. You can include multiple versions of pre$-$processing to reflect the repeated nature of the project.

d$)$ Data Modelling and model evaluation. This part of the report documents the details of the mining tasks conducted and their outcomes. The evaluations of any discovered patterns or models are also presented here. Some interpretation and further analysis of the outcomes are also described here. Any further post$-$processing tasks should also be documented and supported. Just a piece of advice here: you might conclude that the discovery operations so far are still insufficient for finding the final and useful truth. In other words, you may have a feeling of quite $$open ended$$ and you could continue and go on further. This is quite common and natural feeling to have for data science projects. So if you have already done a substantial amount with sufficient depth, you should stop here, and wrap up any possible ideas for further trials as $$Future Work$$ in the Summary part next.

e$)$ Project Evaluation and Summary. This is the final part of the report. It summarizes the work of the whole project and highlight any major findings through the project. You can recommend either the need for another round of data mining to be done or a set of possible deployment actions. Any parts of the project that could be expanded into can be presented as future work. You should also highlight potential impacts of the project results towards the business and$/$or society.