Fundamentals of Artificial Intelligence - Spring 2022 Lab 1: Expert Systems

This document provides instructions for Lab 1 of the Fundamentals of Artificial Intelligence course in spring 2022. Students are tasked with building an expert system to detect different types of tourists visiting Luna-City by distinguishing their characteristics from residents. The system must include a database of 5 tourist types, rules to match user answers to types, forward and backward chaining algorithms, and an interactive question-asking interface. Students should submit a zip file of their Python code and an explanatory report in ELSE by February 2nd for grading based on implementing the required tasks and report quality.

Uploaded by

Mihai Filipescu

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

152 views2 pages

Fundamentals of Artificial Intelligence - Spring 2022 Lab 1: Expert Systems

Uploaded by

Mihai Filipescu

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 2

Fundamentals of Artificial Intelligence - spring 2022

Lab 1: Expert Systems

Handed out: Monday, January 24, 2022

Submissions until: Wednesday, February 2nd, 2022, 7:59 AM

Searching for tourists, expertly

Finally, your dream came true and you’ve landed a job at ”HeinleinAI” – the biggest AI
company in the whole Luna-City! You’re on a testing period so you will want to make sure
that you do your best.
While you are daydreaming about free lunch, the mentor comes and hands you your first
task – you’ll need to build an expert system. They say that while going through the central
library database they stumbled upon this ancient approach that looked like it could solve a
particular lunar problem – detecting tourists.
The tourists on Luna-City are a big source of income and while almost every salesman and
hotel administrator can easily tell one from a Loonie, our mentor researches a more systematic
way of detecting them. While this would prove a trivial task to the city’s main computer,
resources are scarce these days and so you will be researching alternative, more special ap-
proaches.
For starters, research the types of tourists that visit Luna-City and collect a database of
at least 5 tourist types and the criteria by which they can be distinguished from Loonies
and between themselves (i.e. clothes, accent, gait, height and opinion on politics). After your
database is done, create a system that would allow the user to answer some questions about a
possible tourist. If the set of given answers matches a type of tourist from the database, this
should be the system’s answer. If on the other hand, the system determines that the person
in question is a Loonie, the answer should be returned accordingly. Make sure to consider the
case when the set of answers does not find a match in the database (highly improbable if you’ve
done your research well).
Another thing to consider is that the machines that are at your disposal are only capable
of running Python code - either .py files or Jupyter notebooks (.ipynb).
The system that the Luna-City is using is called ELSE. You should upload both the code
and a report describing your work in detail on this system.
The Company has a grading policy which is described in the next page. If you want to
be sure that the company appreciates your work, be sure to follow the guidelines and try to
obtain many points.
General guidelines
• The coding language is Python.

• Submit your solution as a .zip archive (containing .ipynb and .py code files), and a
PDF report uploaded on ELSE.

• For this laboratory you can use the provided code from the archive. Alternatively, you
can implement your own code for the Expert System, but you will have to implement the
entire algorithm yourself, including Forward chaining and Backward chaining.

• For the Report you are free to choose between LaTeX, Jupyter notebook or Word ex-
ported to PDF. Highly recommended Jupyter notebook/Colab.

• Please don’t host your solution in public repositories (e.g Github etc). You can use private
repositories if you need.

• Please refrain from plagiarism.

Grading policy
• Task 1: Define 5 types of tourists. Draw the Goal Tree representing these types of
tourists. (1p)

• Task 2: Implement the rules from the defined tree in task 1 in your code (use the IF,
AND, OR, THEN etc rules which are already implemented in the code). (1p)

• Task 3: If you are using the provided code, check how the Forward chaining algorithm
works and show an example. If you are implementing your own code, implement the
Forward chaining algorithm yourself. (0.5p)

• Task 4: Implement the Backward chaining algorithm for the Goal tree. (2p)

• Task 5: Implement a system for generating questions from the goal tree. Have at least
2-3 types of questions (e.g yes/no, multiple choice, etc). (2p)

• Task 6: Wrap up everything in an Interactive Expert System that will dynamically ask
questions based on the input from the user. Both forward chaining and backward chaining
should be working. (2p)

• Task 7: Format the output and questions to human readable format. (0.5p)

• Report & Presentation of the solution:

Clear explanations, report formatting, code quality, visualisations if relevant etc. (1p)

Good Luck!

The code source and the material was inspired and adapted from 6.034 Artificial Intelligence Lab 1 MIT OpenCourseWare http://ocw.mit.edu.