Artificial Intelligence and Robotics
Feasibility studies
- Technical Feasibility study - Environmental - Economic - Organizational - Health
Robotics uses
- Work - Home - Warfare - Online - Underwater exploration
Internet Search engine
A tool used to look for information on the internet
Artificial Intelligence Techniques
Pattern recognition Searching Machine Learning Heuristics
Implementation stage
Developers create the system, following the design documents previously created.
Direct Changeover
Immediate removal of an old system and complete replacement with a new IT system.
Analysis stage
Investigating the current system, determining the organizations requirements for a new system, investigating possible solutions, determining which are feasible, and choosing the most appropriate.
Identification of possible IT solutions
Many projects have possible solutions, ranging from keeping existing systems. Some cases it may be viable to implement a new solution which extends functionality of existing solution
Examples of expert systems
Medical diagnosis Medical image analysis Identifying agricultural pests and diseases Spelling and grammar checking Finance - Decisions on loans Fault diagnosis in various fields
SSADM (structured systems analysis and design method)
Methodology which focuses primarily on the Analysis and Design stages of the SDLC. SSADM has 7 stages
CAPTCHA
Modern form of Turing Test, Completely Automated Public Turing test to tell Humans Apart are designed to prevent spam bots from posting comments or creating false accounts.
Project initiation documents
This a document that bundles up all the information gathered when the project was first initiated.
Computational intelligence
Computational Intelligence usually refers to the ability of a computer to learn specific tasks from data or experimental observation. Computational intelligence focuses on creating systems that 'think' in the same way humans think.
Installation stage
Concerns prepping the organization for the installation of the new system, the hardware and software installation, and the removal of the old system.
Design stage
Covers inputs, process, data structures and outputs required by a system, and relationship between these items.
Purpose of algorithms in expert systems
Finding faults Analysis Product development
Fuzzy logic
Fuzzy logic is the different degrees of truth. Fuzzy logic is a form of many-valued logic in which the truth values of variables may be any real number between 0 and 1 both inclusive.
Inference rules
IF...THEN.. statements which describes rules for a knowledge domain.
Maintenance stage
Maintenance accounts for a large amount of the time and cost associated with IT projects.
Phased changeover
One part of an organisation switches to a new IT system to test it, while others remain using the old system.
Pattern Recognition
Pattern recognition is the automated recognition of patterns and regularities in data. Pattern recognition is closely related to AI and machine learning. It is the process of recognizing patterns by using machine learning algorithms. How it works - The raw data in a pattern recognition application is processed and converted into a form that is amenable for a machine to use. Pattern recognition involves classification and cluster of patterns.
Alpha testing
Performed to verify the software works according to the requirements specification and design documentation.
Project plan
Project manager will be responsible for seeing the project is completed. Project management softwares will be used to monitor the project completion
Prototype
Prototypes of the product may be created at this stage to demonstrate features to the client and check the project is meeting expectation
Parallel running
Running the old system and the new system side by side.
Agile development model
Seeks to address some of the waterfalls models weaknesses and by allowing for greater adaptability.
Expert Systems
an expert system is a computer system that emulates the decision-making ability of a human expert. Expert systems are designed to solve complex problems by reasoning through bodies of knowledge, represented mainly as if-then rules rather than through conventional procedural code
Technical documentation
targeted at system administrators and other developers who may expand or change the system in the future.
Artificial Neural Networks
the attempt to make computers learn in a similar way to humans, by representing neurons in human brains and the electrical impulses which flow between them.
Machine Learning
the extraction of knowledge from data based on algorithms created from training data
Phase out
to gradually stop using something. Gradually stop using a system
Designing softwares
- Data flow diagrams - Entity relationship Diagrams
Problems with Expert systems
- Data may be incorrectly entered, not complete, or out of date. This will have an affect on the reliability and accuracy of the answers by the system - Fault diagnosis system suggesting incorrect solutions may cause inconvenience. - Cannot address problems outside the set domain - Unsuitable for some problems
Project management softwares
- Gantt chart - PERT Chart (Program, Evaluation and review technique)
Methods of Data collection
- Interview - Questionnaires - Observations - literature searches
Disadvantages of Fuzzy logic
- Not accurate as there are different results - Fuzzy systems don't have the capability of machine learning as-well-as neural network type pattern recognition. -
(PMBOK) Project Management Body of Knowledge
A collection of the knowledge and best practices of the project management profession
Feasibility study
A feasibility study is an assessment of the practicality of a proposed project or system.
System Development Life Cycle (SDLC)
The overall process for developing information systems from planning and analysis through implementation and maintenance
Machine Translation
Translating from one human language to another
Quality Assurance
Used to ensure the development team are following standardized practices which is suitable foe the project under development
Expert system components
User interface - Presents input from the user and accepts input from them. Can also present answers and reasoning for answers Knowledge base - Contains data and facts which form the knowledge in specific knowledge domain. Inference engine - Has the job of matching the users input from user interface with the data contained in knowledge base to find appropriate answers
Testing stage
Versions of the software is given to beta testers with the aim of detecting any remaining bugs and testing the software's usability under real world conditions
Waterfall model
Waterfall model flows from analysis to design till maintenance and back to analysis. + Highly structured and allows long term planning + Ensures problems in the analysis stage are found quickly - Lacks adaptability: large IT projects may take such a long time to develop that by the time development stage is reached, requirements identified in analysis stage may have changed.
User Training
Will be carried out before new system can be used under real world conditions.
user documentation
Written or other visual information about an application system, how it works, and how to use it.
Beta testing
a trial of machinery, software, or other products, in the final stages of its development, carried out by a party unconnected with its development
Limitations of AI
Cannot detect human emotions
Define Intelligence
- Ability to respond to the environment - Ability to learn new knowledge - Ability to use logic or reasoning to come to conclusion - Ability to learn from experience - Ability to make evaluations
Advantages of Fuzzy logic
- Allows expert systems to provide multiple answers, with different degrees of certainty. - Similar to human reasoning - High precision - Mostly robust as not many inputs required - Validation and Verification of a fuzzy knowledge-based system needs extensive testing with hardware -
AI advantages and disadvantages
Advantages + Reduction In human error + Takes more risks that Humans cannot take or are afraid to take + Works 24/7 - good for businesses as it can speed up businesses production + Faster decision making compared to humans Disadvantages - Costly to use as AI is updating every day. The hardware and the software need to get updated with time to meet the latest requirements. Makes humans lazy as we depend on AI to come up with inventions and different ideas. We get too addicted to it Artificial Intelligence basically does humans jobs for us which makes us humans useless which leads to unemployment
Project management methodologies
Aim to describe the best approaches for managing those steps, moving between them, or recording successes and failures to enable future improvement
Legacy system
An computer system that is no longer available for purchase or is no longer supported by the manufacturer.
embeded system
An operating system that combines processors and software in a device. - Cell Phones - Computers - GPS - Washing machine
AI capabilities
Capable of making decisions that humans cannot Can do things humans cannot such as calculations Less time consuming Risk-free
Expert system shell
Set of programs used allowing the building of an expert system through the creation if knowledge and rules.
Set Theory
Set theory is when an object either belongs to a set or not. For example, the food is either hot or cold, not both.
7 stages of SSADM
Stage 0 - Feasibility study Stage 1 - Investigation of the current environment Stage 2 - Business system options Stage 3 - Requirement specification Stage 4 - Technical system options Stage 5 - Logical design Stage 6 - Physical design
PRINCE2 (PRojects IN Controlled Environments)
Starts by considering a business case for a project. Planning, organization and risk management are key features of PRINCE2 approach
Turing test
Test by Alan Turing to test whether or not a machine is intelligent.
Natural language processing
The ability of a computer to understand human languages such as English. Can be input such as voice commands or output such as translation of texts
Artificial Intelligence
The development of computer systems that should be able to perform tasks that normally require human intelligence and human knowledge. For example, speech recognition, decision making, or translation between languages.