17
Knowledge Engineering – Exsys MiniProject UNDER THE SUPERVISION OF: DR.MOHAMMED SAMIR GHAITH.

Restaurant Recommender Exsys. - CORVID

Embed Size (px)

Citation preview

Page 1: Restaurant Recommender Exsys. - CORVID

Knowledge Engineering – Exsys MiniProjectUNDER THE SUPERVISION OF: DR.MOHAMMED SAMIR GHAITH.

Page 2: Restaurant Recommender Exsys. - CORVID

OUTLINES Introduction. Objective of Restaurant Recommender Expert System. Reference of Knowledge . Inputs and Outputs of the Exsys. Expert System Structure.

i. Logic Blocks. (Verbal If/Then rules and variables)ii. Decision Tree of the system.iii.Command Blocks.iv.Example of using Restaurant Recommender Exsys.

Advantages and Disadvantages of Restaurant Recommender Exsys. Summary.

Page 3: Restaurant Recommender Exsys. - CORVID

INTRODUCTION

People make decisions every day. “Which should I go?”, “What should I eat?”... There are too many choices and a little time to explore them all. Expert systems help people make decisions in these complex information spaces. Recommendation expert systems are a type of information filtering that presents lists of items which are likely of user interest. Simply they compare user interest with some reference characteristics and predict the suitable rating that the user would give.This Project based on a recommendation expert system that recommend an item to a user based upon a description of the item and the user’s interests. In this project we used Knowledge Automation Expert System Technology by “CORVID Exsys Software”. It provides efficient and effective way to automate the delivery of problem-solving advice to users. Systems built with Exsys CORVID interact with users in a way that emulates a conversation with a human expert. The systems ask questions dynamically, and use the expert’s logic to produce reasoned advice tailored to the individual.

Page 4: Restaurant Recommender Exsys. - CORVID

OBJECTIVE OF “RESTAURANT RECOMMENDER” EXPERT SYSTEM

Basically, our Restaurant Recommender Exsys Project focuses on which restaurant should you go to; based on food preferences, attributes, and personal preferences.

CORVID expert system advises among a group of possible competing alternatives of restaurants (spreadsheet data), No one restaurant may be ideal in all respects, but the ability of CORVID to handle competing requirements allows it to find the "best" choices overall. The top 3 restaurants that best meet the user's requests are displayed, along with specific comments on potential disadvantages the restaurant might present for the occasion so the results recommends the most appropriate rated restaurants options to go to a specific site that meets user’s requirements.

Page 5: Restaurant Recommender Exsys. - CORVID

REFERENCE OF KNOWLEDGE USED IN YOUR EXPERT SYSTEM

Human Expert. Websites (Restaurants websites, Google Maps... etc.). Webpages (Facebook Pages, Review Blogs... etc.)

Page 6: Restaurant Recommender Exsys. - CORVID

INPUTS AND OUTPUTS OF THE SYSTEM

a) System’s InputsQuestions are asked about user preferences: Budget, Meal type, whether they need smoking area, parking area or free Wi-Fi attributes.

b) Expected Output KnowledgeList of three restaurants matching the user’s search containing restaurant's name, address, telephone and map URL. The list is arranged according to the restaurants’ score from highest to lowest.

Page 7: Restaurant Recommender Exsys. - CORVID

EXPERT SYSTEM STRUCTURE

Page 8: Restaurant Recommender Exsys. - CORVID

LOGIC BLOCKS“If/Then Rules”

Page 9: Restaurant Recommender Exsys. - CORVID

LOGIC BLOCKS“Variables”Name Type Value PromptBudget Numeric value Numeric integers only Budget (per person)

Meal type Static list BreakfastLunchDinnerSupper

What is the type of meal you would eat?

Smoking Area Static list YesNo

Do you need smoking area?

Parking area Static list YesNo

Do you need available parking area?

Free Wi-Fi Static list YesNo

Do you need free Wi-Fi?

Score Confidence Sum ScoreRestaurant Report Collection/Report Text Restaurant report

Overall Report Collection/Report text Overall report

Page 10: Restaurant Recommender Exsys. - CORVID

DECISION TREE OF THE SYSTEM

Page 11: Restaurant Recommender Exsys. - CORVID

COMMAND BLOCKS

Page 12: Restaurant Recommender Exsys. - CORVID

EXAMPLE

Page 13: Restaurant Recommender Exsys. - CORVID
Page 14: Restaurant Recommender Exsys. - CORVID

RESULT

Page 15: Restaurant Recommender Exsys. - CORVID

ADVANTAGES AND DISADVANTAGESAdvantages: Friendly user interface. Restaurant database updated easily. Fast, efficient and logically controlled. Results of three rated recommendation to help user to decide. Handle everyday decision problem which is restaurant selection.

Disadvantages: Answers may not be sufficient to select the exact preferred restaurant. Huge class of restaurants database should be supported to match the users’

preferences.

Page 16: Restaurant Recommender Exsys. - CORVID

SUMMARY

Nowadays, personal decisions cannot wait for an expert advisor. However, ES are essential for people in order to solve decision-making problems. Moreover, by means of Web any individual could access the ES. By using Exsys Corvid® Expert System Development Tool, ES could be developed quickly, even if the person in not a programmer. Accordingly, the ability to select the best overall recommendation among complex and competing factors is a very effective use of expert systems that can be applied to many types of selection problems.

Page 17: Restaurant Recommender Exsys. - CORVID

TEAM MEMBERS:

• Alaa Farag.• Esraa Fathy.• Peter Adel.• Nouran Yehia.• Nourhan Hany.