Expert System, System life cycle analysis
3
System Life Cycle Analysis (Expert System)
By: Maria Jose Sulvaran Alvarez
Research on Expert Systems
Research
Definition: Computer program that uses artificial intelligence to solve problems which normally need human expertise. Expert Systems now-a-days have commercial applications:
- medical diagnosis
- financial investing
- petroleum engineering.
Research methods:
- Observation
- Interviews
- Questionnaire
- Collecting documents
Questionnaire
Questionnaire Results
Research Method
To create the new expert system:
- we handed out questionnaires to the students
- Understand their needs and adjust the expert system to them.
- Research the aptitudes/qualities needed for each after-school activity.
Our Research
Analysis
Knowledge base: collection of facts about the systems domain
Inference Engine: and analyzes facts in knowledge base to provide an answer.
User Interface: method of interaction with the user.
Analysis
Input and Output
Input - Yes or No answer by the user
Process - Inference engine analyzes information and looks for answers.
Output - Conclusion given to user depending on knowledge base and interface engine.
Problem with CURRENT Expert System
Problem
In most schools students face problems when the time comes to sign up for the After School Activities. Most students find themselves unsure of which activity to take that will match their qualities, aptitudes and objectives. In many of this cases users don't have any expert guidance they can turn to for help.
This expert system is designed to provide the help they need and make them feel sure when they choose their activities so they can take full advantage
Hardware and Software used
NO
Hardware:
- One computer per student taking test
- Internet and international computer network
- No server needed
Software:
- Expert system used to create project is ready-made: Mcgoo
- There was no need to customize or create a software.
Hardware & Software
Design
Run the system to Test its functionality and fix errors
To create the Expert System I had to research using websites created by the government, educational institution and experts.
Design
Created User Interface using Mcgoo (Australian program) using the Tree format.
Added the information to the program in the form of simple questions for the user to answer and the system to process.
I created the user interface by connecting the options to a conclusion or consecutive question
User interface screen.
Text
Inference Engine
Knowledge base
Conclusion section
Production
Production
The After School Activities Expert System will be produced using the Australian ES - Builder program MCGOO. The research has been made in these sites:
https://www.wayup.com/guide/types-skills-best-computer-science-major/
https://www.sciencedirect.com/science/article/pii/S1888754616300223
https://www.wired.com/2011/06/sports-aptitude-nba-draft/
https://mypersonalfootballcoach.com/top-5-soccer-player-attributes-needed-to-be-a-great-footballer/
https://www.truity.com/career-profile/craft-or-fine-artist
https://www.adorama.com/alc/5-desirable-qualities-every-good-photographer-should-have
Testing
Testing is used so that we know the Expert System is working properly. In this design we know that:
- The program opens and closes correctly
- Data can be entered by selecting an option
- As the user interface is designed with multiple options answer the user cannot enter any information that the expert system cannot process.
Testing
Opens and Closes correctly
Data can be entered by selecting options
Evidence
Gives user options in case of failure
Technical Documentation
Documentation
Hardware and Software:
- Internet connection to open system
- Accessibility and log in information for use of Mcgoo
- A computer can be used by various users by changing the log in information or simply restarting test. One computer per student is recommended for grater efficiency and comfort.
Data Structure:
- String data type
- Field names are corresponding to the question number or value it is coming from.
Technical Documentation #2
Inputs and Outputs:
- The user is expected to select one of the options provided by the system, e.g. Art, Yes, No, etc.
- At the end the system will provide the conclusion in which it will inform the user of its aptitude for the activity and whether or not to take it.
Data Process:
- Depending on the option selected by the student the system will lead it to further questioning or will provide a conclusion
Implementation
For this project I have decided to use Direct Changeover as the form of implementation. This is due to the lack of a previous Expert System. The students need the help and guidance as soon as possible. Direct Change over will allow an efficient and quick introduction of the program which will give access to all the students almost immediately after the conclusion and validation of the Expert System.
Implementation
Evaluation
Evaluation & Conclusion
Average rating: 9.8
Conclusion
In conclusion I think that the Expert System Created fulfill its purpose and was able to help the user understand what their qualities can be useful for in different activities and in which activity they will enjoy the most. The research used for the creation of the expert system has already been credited in the Documentation faze.