Management

• filter_drama Our client's feedback of our project, sent to us on April 19, 2017.

  We have very much enjoyed working with your team on the PEACH Chatbot project.
  
  You inherited a code base that could not easily be extended as we needed. You correctly decided to go back to basics and investigate the best approach. There have undoubtedly been challenges throughout this project, mainly due to key dependencies outside your control.
  
  But you have persevered and shown initiative throughout. There has been effective communication, with appropriate questions and timely responses.
  
  It has been a pleasure working with you, and we feel that you have progressed the project to a point where it can easily be developed further to meet multiple use cases. We hope you had fun, and wish you all the best for the future.

  
  

This Medium post, published by our client, outlines the first steps we took within this project. It talks about the problem we are trying to solve, why it is important, and a hint at our solution.

This Medium post, published by a medical student helps create additional awareness for the chatbot.

These slides were created for a progress check conducted by Yun Fu. They include the name and description of the project, the progress we have made towards meeting requirements, and division of the overall contribution to the project.

This presentation was made for the pitch we had to deliver to our yeargroup, and our client. It includes the problem our project is addressing, the way we approached a solution, and a bit of information about our development process.

This repository contains our static site which contains everything you see right now.

This Gantt chart tracks our relative progress. Note that auxiliary tasks (testing, deployment) were left out. Also note that we set regular milestones. Click on the image to zoom in.

This is the form we were given by our client to base our conversation for the chatbot on. We have since based the conversational structure our chatbot uses on the questions asked in this assessment.

First biweekly report of the project. We explain the thoughts behind any decisions we made, and include minutes from all of our relevant meetings.

Following the same structure as our previous report, we outline new decisions we made, and the reasoning applied. We also outline the next steps in the process of this project.

We outline the main challenges and successes of this fortnight, including the definition of a sample dialogue for the chatbot and the building of a react component for our project.

The key successes of this fortnight was the creation of our database schema, and understanding the Rocket.chat API. We faced challenges with the API documentation and structure, as it is still in alpha stage.

Within this fortnight, we researched further Rocket.chat, and attempted to define the limits of this platform for our chatbot. We also populated our database with sample data so that we can test our API.

These 2 weeks were extremely productive, as we almost finished building the chatbot API. At this stage, we were still lacking Rocket.chat integration, which limited the testing we were able to do. We also started on the CLI development.

This fortnight was fairly slow in terms of progress, but raised a lot of valid questions about the User Experience and functionality of our chatbot. We completed developing unit tests for the summary API, and worked on migrating the CI from Python 2 to Python 3.

These 2 weeks were spent fixing small bugs in the chatbot, further developing the CLI, and starting work on the 'user commands' such as 'help' and 'exit'. The development of the CLI meant that we were able to perfect the functionality of the API.

The period of this report was very productive. We deployed the chatbot on Azure, finished the React UI demo, and further developed the 'user commands' functionality of the chatbot.