Week14 final

Lecture 14

Future ICT trends

1. Artificial Intelligence2. Ubiquitous (Pervasive) Computingo Personal area networkso Wireless sensor networkso RFIDs

3. Next-generation Networking (NGNs)o IP Convergenceo Grid/ Cloud computing

4. Conclusions

Today’s lecture outline

1

Image courtesy: http://pages.cs.wisc.edu/~dyer/images/AI.jpg

ArtificialIntelligenceAI is the study of how to make computers do things at which, at the moment, people are better.

• Natural Language Processing–make a computer understand English

• Machine Learning– change behavior based on experience– recognize patterns

• Perception of a visual scene– be able to pick out people, objects, etc.

• Robotics– navigation, accomplishing tasks, etc.

Applications of AI

IBM Deep Blue defeats Kasparov

What qualifies as AI?---Turing Test

Reverse Turing test---CAPTCHAs

Loebner Prize

Virtual Chat bots

Expert Systems

Natural Languages Processing

Computers find it hard to process natural language because of its ambiguity.

Resolved through knowing the context.

A word can have different meanings depending on the context; this linguistic ambiguity confuses computers.

Whose ??Computer’s? or

natural language’s?

Natural Languages Processing

Robotics

Research area in which AI agents are equipped with sensors to perceive the world and effectors to change it

An active, artificial agent whose environment is the physical world.

Three basic parts of a robot:Sensors Computer Actuators

1. Industrial Robots

– Operates in a stable and known environment

– Fixed or limited mobility

– Relatively simple control program

Two major types of Robots

2. Mobile Robots– Operates in the

“real” world

–Mobile!

– Requires a high degree of autonomy

Two major types of Robots

Computer or Machine VisionComputer emulation of human vision

Inverse of Computer Graphics

Applications: 1. Face recognition, 2. Image processing, 3. Motion detection, etc.

ImagesComputer

visionWorld model

Computergraphics

World model Images

Ubiquitous Computing

2

Image courtesy: http://www.timboucher.com

Ubiquitous (Pervasive) ComputingPersonal Area

Network

Ubiquitous (Pervasive) ComputingWireless Sensor

Networks

Ubiquitous (Pervasive) ComputingWireless Mesh

Networks

Radio Frequency Identification (RFID)

Ubiquitous (Pervasive) Computing

Image courtesy: http://www.ida.gov.sg/

Ubiquitous (Pervasive) Computing

Image courtesy: http://www.rfid-library.com/

Radio Frequency Identification (RFID)

Next-GenerationNetworks

3

Image courtesy: http://fusion.demon.net

Single converged IP

backbone

Mobile telephony

Television

Fixed telephony

Data

Next-generation Networks

Cloud ComputingSoftware as a service

(SAAS)

Grid Computing

Conclusions1. Artificial Intelligence2. Ubiquitous (Pervasive) Computingo Personal area networkso Wireless sensor networkso RFIDs

3. Next-generation Networking (NGNs)o IP Convergenceo Grid/ Cloud computing

4. Conclusions

???Questions/

Confusions?

Credits/ Acknowledgement can be found

at the course website:

http://tinyurl.com/5hb8pp