Compegence: Dr. Rajaram Kudli - An Introduction to Artificial Neural Network and its Applications_2012_Oct

Introduction to Artificial Neural Networks 1 www.compegence.com

an introduction to

Artificial Neural Networks

and its applications

Process, Data and Domain driven Business Decision Life Cycle

Partners in Co-Creating Success

- Dr. Rajaram Kudli




Intelligent Systems

• What is Intelligence?

– The capacity for understanding or the ability to perceive and comprehend meaning - Cognition

– System or method able to modify its action in the light of ongoing events - Adaption


• AI - Artificial Intelligence– a branch of Computer Science concerned with the problems of

reasoning, knowledge representation, planning, learning, natural language processing, communication, perception etc., thorough the approaches of statistical methods, computational intelligence and symbolic computation, aimed at engineering Intelligent Machines

• BI - Business Intelligence– a technology enabled discipline comprising of functions viz., reporting,

online analytical processing, data mining, business performance management, benchmarking, predictive & prescriptive analytics etc. that enable an enterprise to discover actionable insights from all its data

• CI - Computational Intelligence– a set of nature-inspired computational methodologies and approaches

to address complex real-world problems to which traditional approaches, viz., ab-initio modeling or explicit statistical modeling, are ineffective or infeasible

ABCs of Intelligence


Computational Intelligence - Paradigms

• Artificial Neural Networks– Human-Like Information Processing

• Fuzzy Logic– Human-Understandable Reasoning

• Genetic Algorithms– Human-Like Evolution

• Chaos Theory– Humanity-Like Complex Behavior


Human Understandable Vs Human-Like

• Human Understandable– Artificial Intelligence,

Fuzzy Logic and Genetic Algorithms

– Synthetic, rule-based logical models

– Easier to explain the knowledge & method of solution

– Easier to gain acceptance

• Human-Like– Artificial Neural

Networks; Chaos Computing

– Natural, abstract models

– Harder to extract meaning from the values

– Harder to gain acceptance



• Computational models inspired by the human brain– Massively parallel, distributed system, made up of simple

processing units., neurons– Synaptic connection strengths among neurons are used to

store the acquired knowledge.– Knowledge is acquired by the network from its

environment through a learning process


Model Complexity

Computational Complexity

HIGHARIMA

NNET

MEDAAR

AR

LOWCM

EMAWMA

SMA

HWM HLES

Model Complexity (Forecasting Application)

LOW MED HIGH

Information Complexity

Non-regression Models Regression Models


Applicability – Where & Why ?

• Why? – Ability to solve data-

intensive problems – Adaptation– Parallel Distributed

Representation & Processing

– Fault tolerance– Nonlinearity– Scalability– Universality

• Where?– Where data is noisy,

complex, imprecise, and hi-dimensional

– Where a clearly stated mathematical solution or algorithm doesn’t exist

– Where an explanation of the decision is not required


Example – CPG-Retail Sales Forecasting

0

1

2

3

4

5

6

7

8

4-Oct-

08

4-Nov-08

4-Dec-

08

4-Jan-

09

4-Feb-

09

4-Mar-

09

4-Apr-

09

4-May

-09

4-Jun-09

4-Jul-09

4-Aug

-09

4-Sep

-09

4-Oct-09

4-Nov

-09

4-Dec-

09

4-Jan

-10

4-Feb

-10

4-Mar-

10

4-Apr-

10

4-May

-10

4-Jun

-10

4-Jul-10

4-Aug

-10

4-Sep

-10

4-Oct-1

0

4-Nov

-10

4-Dec-

10

4-Jan

-11

Week End Date

Sales Tot

al Q

uant

ity

Validation Forecast

0

100

200

300

400

500

600

700

04-Oct-

08

04-Nov-0

8

04-Dec-

08

04-Jan

-09

04-Feb

-09

04-Mar-

09

04-Apr-

09

04-May-

09

04-Ju

n-09

04-Ju

l-09

04-Aug-0

9

04-Se

p-09

04-Oct-

09

04-Nov-

09

04-Dec-

09

04-Jan

-10

04-Fe

b-10

04-Mar-

10

04-Apr-

10

04-May-

10

04-Jun

-10

04-Jul

-10

04-Aug-1

0

04-Sep

-10

04-Oct-

10

04-Nov

-10

04-Dec-

10

04-Ja

n-11

Week End Date

Sales

Total Qua

ntity

ACTUAL SF_BEST

Validation Forecast

MODEL %BEST FC (1000 SKUs)

NNET 66%AAR 18%ARIMA 8%HW 2%HOLT 1%CR 0%AR 1%EMA 0%WMA 2%SMA 2%

• An Intelligent Forecasting System that evaluates 10 classical forecasting models including Neural Networks, and gives best forecast acceptable to qualitative expectations of a human expert


Example – Supervised Learning

• ALVINN, Autonomous Land Vehicle In a Neural Network, is a perception system which learns to control the NAVLAB vehicles by watching a person drive.


Example– Unsupervised Learning

• WEBSOM is a method for automatically organizing collections of text documents and for preparing visual maps of them to facilitate the mining and retrieval of information.


• Image Storage & Reconstruction by Hopfield network trained on the sample images and then presented with either a noisycue or a partial cue.

Example – Associative Memory

Original, Stored Degraded Cue Reconstructed


Example - Controls

• HOAP, or Humanoid for Open Architecture Platform, represents a fundamentally different approach to creating humanoid robots, in harnessing the power of a neural network to tackle movements and other tasks.


Application Areas in Engineering

• Aerospace: High performance aircraft autopilots, flight path simulations, aircraft control systems, autopilot enhancements, aircraft component simulations, aircraft component fault detectors

• Automotive: Automobile automatic guidance systems, warranty activity analyzers• Electronics: Code sequence prediction, integrated circuit chip layout, process control,

chip failure analysis, machine vision, voice synthesis, nonlinear modeling • Mechanical: Condition monitoring, Systems modeling and control• Manufacturing: Manufacturing process control, product design and analysis, process

and machine diagnosis, visual quality inspection systems, beer testing, welding quality analysis, paper quality prediction, computer chip quality analysis, analysis of grinding operations, chemical product design analysis, machine maintenance analysis, project bidding, planning and management, dynamic modeling of chemical process systems

• Robotics: Trajectory control, forklift robot, manipulator controllers, vision systems• Telecommunications: Data compression, signal processing, pattern recognition: Face,

Objects, Fingerprints, Speech recognition; automated information services, real-time translation of spoken language, customer payment processing systems, Equalisers, Network Design, Management, Routing and Control, ATM Network Control, Fault Management, Network Monitoring


Application Areas in Business

• Business Analytics: Market Research, Market Structure, Market Mix, Customer behavior modeling, Propensity modeling for Purchase, Renewals, Default, Attrition, Fraud, Market & Customer Segmentation

• Banking: Credit/Loan application evaluators, Fraud and Risk evaluation, Credit card attrition, Delinquency

• Financial: Real estate appraisal, loan advisor, mortgage screening, corporate bond rating, credit line use analysis, portfolio trading program, corporate financial analysis, currency price prediction

• Education: Modeling Students’ performance, Personality Profiling, Diagnostics of a modern state, analysis, and forecasting of dynamics of a system of education

• Defense: Weapon steering, target tracking, object discrimination, facial recognition, new kinds of sensors, sonar, radar and image signal processing including data compression, feature extraction and noise suppression, signal/image identification; Counter-terrorism

• Medical: Breast cancer cell analysis, EEG and ECG analysis, prosthesis design, optimization of transplant times, hospital expense reduction, hospital quality improvement, emergency room test advisement

• Securities: Market analysis, automatic bond rating, stock trading advisory systems• Transportation: Truck brake diagnosis systems, vehicle scheduling, routing systems


#1 - Predicting Stock Prices

• Walkrich Investment Advisors used Neural Networks to produce an investment tool WRRAT based loosely on Warren Buffett's ideas to predict stock prices, and determine which stocks are trading below their market value. The results from January 1995 to January 1996 showed that a Portfolio of WRRAT's most under-priced shares saw an average advance of 33%.


#2 - Predicting S&P 500 Index

• LBS Capital Management used a neural network software to predict the S&P 500 index. The company uses an expert system to provide instructions to the neural network, which then processes the data accordingly. When tested with hundreds of previous days data the neural network LBS trained predicts the S&P 500 with an accuracy of about 95%.


#3 – Predicting Currencies

• O'Sullivan Investments successfully used many neural networks in order to advise them of market trends. Mr James O'Sullivan produced an article Neural Nets: A Practical Primer, AI In Finance, 1994 outlined some of the networks used. One of the most important factors in producing a successful net is to ask the right kind of question. Rather than simply ask what the projected price of a currency might be, he asks at what price the market is likely to take off in one direction or the other etc.


#4 – Predicting Natural Gas Price

• Northern Natural Gas is a regulated wholesaler of natural gas. They must develop and file a rate for the gas they sell based on the average cost of the gas. By developing a neural network that use factors such as the quarter of the year, season, temperature last month etc. to predict the following months oil price, the company was better able to plan rates.


#5 – Predicting Bonds

• G. R. Pugh & Company does consulting to predict the prices of bonds of public utilities. The company used neural networks to help forecast the following years corporate bond prices and ratings of over 100 public utility companies. The network they used compared very favourably to conventional mathematical analysis. Whereas the network was able to predict a utilities rating (A, B, C) with 95% accuracy, conventional mathematical analysis was only effective 85% of the time. The only difficulties encountered by the network were associated with companies experiencing particularly unusual problems that were not incorporated into the networks inputs.


#6 – Direct Mail Marketing

• Microsoft used neural networks to maximise the effectiveness of their marketing campaign. Each year the company sent out mail to its registered customers. Most of this mail offered upgrades or new software but the response rate was rather low. The company used a neural network that was fed various variables such as how recently they registered, how many products they have bought etc. to target users more effectively. The results showed an average mailing lead to a 35% cost savings.


#7 – Credit Scoring

• Research conducted by Dr Herbert Jensen PhD demonstrated that "building a neural network capable of analysing the credit worthiness of loan applicants is quite practical and can be done quite easily". The neural network was trained on no more than 100 loan applications to process application data such as occupation, years with employer etc. Despite the relatively small training set the network achieved a 75-80% success rate. This compared well with more traditional scoring methods that resulted in about a 75% success rate.


#8 – Real Estate Appraisal

• Several neural networks have been used to predict the sale prices of homes in order to help appraisers assess, sellers estimate asking prices, and home owners decide on improvements. Richard Borst successfully trained a neural network to appraise real estate in the New York area. His network incorporated almost 20 variables including the square feet of living area, age, etc. He used over 200 sales records from 1988 and 1989 to train the network with about 90% accuracy.


Summary

• ANN can solve the direct (prediction) and inverse (control) problem easy and fast in spite of incompleteness of data

• ANN can solve problems of higher complexity of modeling, recognition, predictions, and control in engineering & business, better than traditional solutions

• ANN paradigms provide powerful approaches to the problem domains with high contact of theory, simulation, experiment, data and human expertise


ANN Paradigms – Theory & Practice

• Feed Forward Networks (FFNN)– Multi-Layer Perceptrons (MLP)

• Competitive Learning Networks (CLNN)– Self-Organizing Maps (SOM)

• Recurrent Neural Networks (RNN)– Hopfield Networks (HNN)


Process, Data and Domain driven Information Excellence

ABOUT COMPEGENCE


Market Actions

Systemic Changes

Business Landscape

Business Intent

Business Usage

Cost

EffortSkills &

Competency

Sustainable

Scalable

Flexible

Timely

Usable

Actionable

Process

Systems

DataDomain

Information

Process, Data and Domain Integrated Approach

Decision ExcellenceCompetitive Advantage lies in the exploitation of:

–More detailed and specific information–More comprehensive external data & dependencies–Fuller integration

–More in depth analysis–More insightful plans and strategies

–More rapid response to business events–More precise and apt response to customer events

We complement your “COMPETING WITH ANALYTICS JOURNEY”


Value Proposition

Data Qua lit y and Pro cess Aud it

Custo mer Data

Tra nsla teSegme ntDe riv e

Sum ma rize

Profiling

Sour ce DataSour ce D ata Extrac tExtra ct Sta gingSta gin g Tra nsfo rmTra nsfo rm Lo adLo ad Ap pl ica tion sAp pl icat ion s

Meta data Laye r f or C on sistent Bu sin e ss Unde rstandi ng

Assets

L i ab il i ti es

I n ve stmen t

Cards

Reference Data(Bra nch, P rodu cts)

P artn er Data

CRM / Marketin g P rograms

Inte gra te

Analysis

Reports

Dashboar ds

Excel Interface

Busine ss Rule s

Trusted Da ta

Fou ndat ion

with DW

Pla tf orm

Trusted Da ta

Fou ndat ion

with DW

Pla tf orm

ConstraintsAlternativesAssumptionsDependencies

Concerns / RisksCost of Ownership

Technology Evolution

Repeatable ReusableLeverageTrade Offs

Ease of Use: Drill Down, Up, Across

Tools

Technologies

Trends

Platforms

People

Processes

Partners

Cost

Time

TeraBytes

Reports

Dashboards

Decisions?

Actions?

Results?

Returns?

Jump Start the “Process and Information Excellence” journey

Focus on your business goals and “Competing with Analytics Journey”

Overcome multiple and diverse expertise / skill-set paucity

Preserve current investments in people and technology

Manage Data complexities and the resultant challenges

Manage Scalability to address data explosion with Terabytes of Data

Helps you focus on the business and business processes

Helps you harvest the benefits of your data investments faster

Consultative Work-thru Workshops that help and mature your team

Data

Processes

Decisions

Actions

Results

Returns

COMPEGENCE

People

Current State


Our Expertise and Focus Areas

Process + Data + Domain => Decision

Analytics; Data Mining; Big Data; DWH & BI

Architecture and Methodology

Partnered Product Development

Consulting, Competency Building, Advisory, Mentoring

Executive Briefing Sessions and Deep Dive Workshops


Process, Data and Domain driven Information Excellence

Process, Data and Domain driven Business Decision Life Cycle

Partners in Co-Creating Success

[email protected]

Compegence: Dr. Rajaram Kudli - An Introduction to Artificial Neural Network and its Applications_2012_Oct

Documents