studentHW1sol - Bryant Servello CSE 352 Homework 1 9/21/10...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Bryant Servello CSE 352 – Homework 1 9/21/10 Problem 1 (2) Expert System 1: Wildlife Recognition Description: Allow a user to identify an organism based on its traits and behavior Use: This system would be useful for anyone who is curious about a plant or animal they have seen (consider bird watching books, botanical encyclopedia, etc). By making an expert system with this functionality, it would make searching for an animal much easier. Knowledge Needed: The expert rules for this system would encode facts about a plant or animal to narrow down the search. For instance: hasFeathers(x, true) hasBeak(x, true) → type(x, bird) type(x, mammal) size(x, large) hasTusks(x, true) → species(x, elephant) The user input „facts‟ would be the attributes of the organism. For instance: Large Brown Furry 4 Legs This might be enough data for the system to identify a bear (pictures could be used for verification). Expert System 2: Vacation Scheduler Description: Help the user select a perfect vacation Use: This system would help the user pick the vacation best suited to his/her desires and budget Knowledge Needed: The expert rules for this system would process the user‟s desires. For instance: temperature(x, warm) swimming(x, true) → possibleDestination(x, Hawaii) From the user, the system would need parameters to be given for what is desirable in a vacation to form the database of facts. For instance: Warm Sunny SCUBA diving Coral Reef These „facts‟ (gained by asking the user questions) would help the system produce a list of suitable locations to travel to. (4) If I were a judge in the Loebner Contest, I think I would use mostly questions involving “why.” Since “why” questions often have very complicated, in-depth answers, they present a hurdle for a machine. Questions such as these might be tricky: “Why do people work every day?” “Why did you decide to participate in this event?”
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
I would also ask things which are more conversational than data-related. Here are a few examples: “How was your trip here?” “What‟d you think of the lecture today?” Alternatively, instead of looking for answers that do not quite hold up to my expectations, I could ask questions which an AI entity would be better qualified to answer. Questions like: “How far is it between the Earth and Mars?” “What is 9736.5 divided by 122.6?” These questions, barring an exceptionally smart human or an AI programmed carefully to avoid
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/25/2012 for the course CSE 352 taught by Professor Wasilewska,a during the Fall '08 term at SUNY Stony Brook.

Page1 / 7

studentHW1sol - Bryant Servello CSE 352 Homework 1 9/21/10...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online