Thesis subject

Collaborative Agricultural Platform with Chatbot

A chatbot is a software that conducts a conversation via auditory or textual methods. In recent years, chatbot has more functions based on artificial intelligence technology. Some chatbots, such as Siri, Alexa, Google Assistant etc, are able to gauge consumer’s needs and then help them. Chatbots are widely used in a commercial transaction, hotel booking, form submission, etc.

At the same time, Community-based Question and Answering (CQA) systems have become a popular platform where users share knowledge and obtain answers from community members and experts. CQA has received much attention in industry and in academia.

In this project, the student is required to build a collaborative CQA system. The system is based on an instant message system, such as whatsup, wechat, and telegram. Meanwhile, the system includes chatbots in instant messaging discussions.

background
A typical workflow of the system is as follows.

  • Firstly, several people have a conversation;
  • Secondly, a people in the conversation raise a question;
  • Thirdly, if all the other people do not know the answer, the chatbot is raised.
  • Fourthly, the chatbot analyzes the question and provide an answer. In this project, the student is required to add an agriculture-related detection component in the chatbot.
    For example, the questions could be a photo of crop pests and diseases. The server of chatbot is able to make image recognition/categorization based on the uploaded photo, and provide the answer to the CQA system.

To finish the project, the student must be able to achieve the following work steps.

  1. Set up an instant message system based on open-source software.
  2. Make image recognition for a photo of crop pests and diseases. In this step, the student could decide the type of image recognition according to the available agricultural data he or she can get.
  3. Integrate the image recognition system to the messaging platform. The challenge of this step is to guarantee that the chatbot is able to capture the question correctly in a group of instant messages.

Literature

  • Andrea Burattin. Integrated, Ubiquitous and Collaborative Process Mining with Chat Bots. Proceedings of the Dissertation Award, Doctoral Consortium, and Demonstration Track at {BPM} 2019 co-located with 17th International Conference on Business Process Management, {BPM} 2019, Vienna, Austria, September 1-6, 2019.
  • Xiaoxue Shen, Liyang Gu, Adele Lu Jia. Cultivating Online: Question Routing in a Question and Answering Community for Agriculture. arXiv 2019.

Requirements (optional)

  • Python programming
  • Basic knowledge on image recognition using neural network.

Contact person

  • qingzhi.liu@wur.nl