Applicable Artificial Intelligence

Building a ChatBot integrated with an LLM

Objective
The objective is to develop a functional chatbot integrated with a Large Language Model (LLM), perform vectorisation on an organisation's dataset, and write a report on the work done. The chatbot should be capable of interacting with users and providing meaningful responses, while the vectorisation process aims to convert the textual data in the organisation's dataset into numerical representations such that they can be used by an LLM.

Task 1: Building a Chatbot

1.1 Chatbot Requirements:
Choose a suitable Language Model (e.g., GPT-3, BERT, Starling, Llama, Mistral, etc.) for the chatbot implementation.
Implement a user-friendly interface for interacting with the chatbot.
The chatbot should understand and respond to natural language queries.
In the chatbot instructions, include a set of predefined responses for common queries related to the organisation.
Ensure the chatbot can handle context and maintain a coherent conversation.

1.2 Integration with LLM:
Implement the integration of the chatbot with the selected Language Model (LLM).
Utilize the LLM to enhance the chatbot's understanding and generation of responses.
Demonstrate how the chatbot benefits from the capabilities of the chosen LLM.

1.3 User Interaction:
Design and implement a user interface for interacting with the chatbot.
Include examples of conversations that showcase the chatbot's capabilities.
Ensure a smooth and intuitive user experience.

Task 2: Vectorising the Organisation's Dataset

2.1 Dataset Description:
Provide details about the given organisation's dataset, such as the type of data it contains and its structure.
Include information on any specific challenges that you have experienced or that may arise during vectorisation.

2.2 Vectorisation Process:
Choose an appropriate vectorisation technique (e.g., TF-IDF, Word Embeddings) for the organisation's dataset.
Implement the vectorisation process to convert textual data into numerical representations.
Explain the rationale behind the selected vectorisation technique and how it benefits the organisation.

2.3 Data Exploration:
Conduct exploratory data analysis (EDA) on the vectorised dataset.
Try to create data visualization plots and images through the LLM on some aspects of the data representation.
Provide clear documentation on the vectorisation process