Artificial Intelligence (Ai) and Deep Learning with pictorial illustrations of Ai classifications and Machine Learning. This is a Research Paper Presentation on topic (Deep Learning Previous and Present Applications)
4. ARTIFICIAL INTELLIGENCE (AI)
Any TASK performed by a MACHINE that, if a
human carried out the same activity, we would say
the human had to apply INTELLIGENCE to
accomplish the task.
6. MACHINE LEARNING & SYMBOLIC AI
Symbolic Artificial Intelligence, also known as Good Old Fashioned AI makes
use of strings that represent real-world entities using simple (if-else) conditions.
MACHINE LEARNING
• SYMBOLIC AI:
• MACHINE LEARNING:
7. DEEP LEARNING
• Deep learning is the fastest growing segment of
artificial intelligence (AI).
• It uses many layered Deep Neural Networks
(DNNS) to learn levels of representation and
abstraction that makes sense out of data.
11. RESEARCH OBJECTIVES:
RESEARCH QUESTIONS:
• How and in what major applications deep learning algorithms have been utilized
• The purpose of this review is to give a general concept of deep learning and how its advantages
have improved during the last years.
12. LITERATURE REVIEW
Name of the Researcher
and Title
Application Technique Conclusion
• Marra, Poggi, Sansone,
& Verdoliva (2017)
• Raja,
Raghavendra,Vemuri &
Busch (2015)
• Ossama, Abdelrahman,
Hui & Gerald (2012)
• Iris sensor model identification
• Iris Recognition by using
smartphones’ cameras
• Multi-speaker speech
recognition
• Convolutional neural
networks
• Deep sparse filtering
• Local filtering and
max pooling
infrequency domain
• Iris sensor model is
important for
forensics
applications.
• The characteristics of
iris are significantly
unique for each and
can be recognized
from distance.
13. LITERATURE REVIEW
Name of the Researcher
and Title
Application Technique Conclusion
• Hinton, E, Osindero &
WhyeTeh (2006).
• Mohamed, Abdelrahman,
Dahl & Hinton (2009).
• Digit Classification
• Deep Belief Networks for phone
recognition
• Complementary
Priors on Belief
networks
• Back propagation
and associative
memory architecture
• The difficulty of
visual pattern
recognition becomes
apparent if a program
is designed to
recognize digits.
14. SIGNIFICANCE OF THE STUDY
The era of digitalization has
allowed the technology to flourish
and computers that are able to
analyze massive amounts of
complex data known as ‘big data’
can now provide more accurate
results. This helps companies
increase their chance to identify
profitable opportunities or avoid
unknown risks.
15. LIMITATION OF THE STUDY
• Number of deep learning publications were observed from ScienceDirect
database per year from 2006 till June 2017.
• The IEEE Digital Library’s number of conference publications and journals &
magazines from Jan 2006 till the Jun 2017 were analyzed.