with DATA ANALYTICS, MACHINE LEARNING,

DEEP LEARNING & ARTIFICIAL INTELLIGENCE

using PYTHON, R & Weka

INTRODUCTION TO DATA SCIENCE:

 What is Data Science?
 Who is Data Scientist and who can become a Data Scientist?
 Real time process of Data Science
 Data Science Applications
 Technologies used in Data Science
 Prerequisites knowledge to learn Data Science

INTRODUCTION TO MACHINE LEARINING:

 What is Machine Learning?
 How Machine will learn like Human Learning?
 Traditional Programming vs. machine learning
 Machine Learning engineer responsibilities
 Types of learning
 Supervised learning
 Un-supervised learning
 Machine learning algorithms: KNN, Naïve-bayes, Decision trees,
Classification rules, Regression (Linear Regression, Logistic Regression),
K-means clustering, Association rules, Support Vector Machine, Random
Forest.

PYTHON PROGRAMMING:
 What is Python? History of Python
 Python Features, Applications of Python
 Downloading and Installing Python
 Python IDE: Jupyter Notebook & Spyder
 What is Anaconda Navigator?
 Downloading and Installing Anaconda, Jupyter Notebook & Spyder
 Python Programming vs. Existing Programming
 Interactive Mode Programming & Script Mode Programming
 Python Identifiers, Reserved Words
 Lines and Indentations, Quotations, Comments
 Assigning values to variables

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

  Operators - Arithmetic Operators, Comparison (Relational) Operators,
Assignment Operators, Logical Operators, Bitwise Operators, Membership
Operators, Identity Operators
 Decision Making and Loops
 Flavors in Python, Python Versions
 Data Types: int, float, complex, bool, str
 List, Tuple, Range, Bytes & Bytearray
 Set, Frozenset, Dict, None
 Inbuilt Functions in Python, Slice operator - Indexing
 Mutable vs. Immutable, Modules and Packages
 Database Connection - PyMySQL, Defining & Manipulating

NumPy with Python:

 NumPy Environment setup in Python, Features of NumPy
 Array Creation, Indexing & Slicing, Array Manipulation
 Mathematical Functions, Statistical Functions

Pandas with Python:

 Pandas Environment setup in Python
 Features of Pandas, Data Structures
 Series - Create Series, Accessing Data from Series with Position
 DataFrame - Features of DataFrame, Create DataFrame, DataFrame from
List, Dict, Row & Column Selecting, Adding & Deleting
 Panel - Create and select data from Panel
 Indexing & Selecting Data, Statistical Functions
 Merging / Joining, Categorical Data

R PROGRAMMING:
 R Programming Introduction
 R Programming vs. Existing Programming
 Downloading and Installing R, What is CRAN?
 R Programming IDE: RStudio, Downloading and Installing RStudio
 Variable Assignment - Displaying & Deleting Variables
 Comments – Single Line and Multi Line Comments
 Data Types – Logical, Integer, Double, Complex, Character
 Operators - Arithmetic Operators, Relational Operators, Logical Operators,
Assignment Operators, R as Calculator, Performing different Calculations
 Functions – Inbuilt Functions and User Defined Functions
 STRUCTURES – Vector, List, Matrix, Data frame, Array, Factors
 Inbuilt Constants & Functions

Setting Environment:
 Search Packages in R Environment
 Search Packages in Machine with inbuilt function and manual searching
 Attach Packages to R Environment
 Install Add-on Packages from CRAN

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

  Detach Packages from R Environment
 Functions and Packages Help

Vectors:
 Vector Creation, Single Element Vector, Multiple Element Vector
 Vector Manipulation, Sub setting & Accessing the Data in Vectors

Lists:
 Creating a List, Naming List Elements, Accessing List Elements
 Manipulating List Elements, Merging Lists, Converting List to Vector

Matrix:
 Creating a Matrix, Accessing Elements of a Matrix
 Matrix Manipulations, Dimensions of Matrix, Transpose of Matrix

Data Frames:
 Create Data Frame, Vector to Data Frame
 Why Characters are Converting into Factors? – stringsAsFactors
 Convert the columns of a data frame to characters
 Extract Data from Data Frame
 Expand Data Frame, Column Bind and Row Bind
 Merging / Joining Data Frames – Inner Join, Outer Join & Cross Join

Arrays:
 Create Array with Multiple Dimensions, Naming Columns and Rows
 Accessing Array Elements, Manipulating Array Elements
 Calculations across Array Elements

Factors:
 Factors in Data Frame, Changing the Order of Levels
 Generating Factor Levels, Deleting Factor Levels

Loading and Reading Data:

 DATA EXTRACTION FROM CSV
 Getting and Setting the Working Directory
 Input as CSV File, Reading a CSV File
 Analyzing the CSV File, Writing into a CSV File
 DATA EXTRACTION FROM URL
 DATA EXTRACTION FROM CLIPBOARD
 DATA EXTRACTION FROM EXCEL
 Install “xlsx” Package
 Verify and Load the "xlsx" Package, Input as “xlsx” File
 Reading the Excel File, Writing the Excel File
 DATA EXTRACTION FROM DATABASES

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

  RMySQL Package, Connecting to MySql
 Querying the Tables, Query with Filter Clause
 Updating Rows in the Tables, Inserting Data into the Tables
 Creating Tables in MySql, Dropping Tables in MySql
 Using dplyr and tidyr package

STATISTICS:
 Mean, Median and Mode
 Data Variability: Range, Quartiles, IQR, Calculating Percentiles
 Variance, Standard Deviation, Statistical Summaries
 Types of Distributions – Normal, Binomial, Poisson
 Probability Distributions, Skewness, Outliers
 Data Distribution, 68–95–99.7 rule (Empirical rule)
 Descriptive Statistics and Inferential Statistics
 Statistics Terms and Definitions, Types of Data
 Data Measurement Scales, Normalization
 Measure of Distance, Euclidean Distance
 Probability Calculation – Independent & Dependent
 Hypothesis Testing, Analysis of Variance

DATA VISUALIZATION:
 Data Visualization with MatPlotLib and Seaborn
 Data Visualization with Graphics and GrDevices
 High Level Plotting and Low Level Plotting
 Pie Charts - Title, Colors, Slice Percentages, Chart Legend
 3D Pie Charts
 Box Plots - Outliers, Ranges, IQR, Quantiles, Median, Data Distribution
Analysis, 68–95–99.7 rule (Empirical rule)
 Bar Charts - Label, Title, Colors, Group Bar, Stacked Bar Charts
 Histograms - Range of X and Y Values
 Line Graphs - Types: Points, Lines, Both, Overplotted, Steps
 Scatterplots
 Combining Plots - Par and Layout

LAZY LEARNING – CLASSIFICATION USING NEAREST NEIGHBORS:

 Understanding Classification Using Nearest Neighbors
 The KNN algorithm
 Calculating distance
 Choosing an appropriate k
 Preparing data for use with KNN
 Why is the KNN algorithm lazy?
 Diagnosing breast cancer with the KNN algorithm
 Collecting data
 Exploring and preparing the data
o Transformation-normalizing numeric the data

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

 o Data preparing –creating training and test datasets
 Training a model on the data
 Evaluating model performance
 Improving model performance
o Transformation –z-score standardization
o Testing alternative values of k

PROBABILISTIC LEARNING – CLASSIFICATION USING NAÏVE

BAYES:
 Understanding Naïve-Bayes
 Basic concepts of Bayesian methods
 Probability
 Joint probability
 Conditional probability with Bayes’ theorem
 The Naïve Bayes Algorithm
 The Naïve Bayes classification
 The Laplace estimator
 Using numeric features with Naïve Bayes
 Filtering Mobile Phone Spam with the Naïve-Bayes Algorithm
 Collecting data
 Exploring and preparing the data
 Data preparation –processing text data for analysis
o Data preparation –creating training and test datasets
o Visualizing text data-word clouds
o Data preparation-creating indicator features for frequent words
 Training a model on the data
 Evaluating model performance
 Improving model performance

DIVIDE AND CONQUER – CLASSIFICATION USING DECISION TREES

AND RULES:
 Understanding decision trees
 Divide conquer
 The C5.0 decision tree algorithm
o Choosing the best split
o Pruning the decision tree
 Identifying risky bank loans using C5.0 decision trees
 Collect data
 Exploring and preparing the data
o Data preparation-creating random training and test datasets
 Training a model on the data
 Evaluating model performance
 Improving model performance
o Boosting the accuracy of decision trees
o Making some mistakes more costly than others

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

  Understanding classification rules
 Separate and conquer
 The one rule algorithm
 The RIPPER algorithm
 Rules from decision trees
 Identifying poisonous mushrooms with rule learners
 Collecting data
 Exploring and preparing data
 Training a model on the data
 Evaluating model performance
 Improving model performance

FORECASTING NUMARIC DATA – REGRESSION METHODS:

 Understanding regression
 Simple linear regression
 Ordinary least squares estimation
 Correlations
 Multiple linear regressions
 Predicting medical expenses using linear regression
 Collecting data
 Exploring and preparing data
o Exploring relationships among features- the correlation matrix
o Visualizing relationships among features –the scatter plot
matrix
 Training a model on the data
 Evaluating model performance
 Improving model performance
o Model specification –adding non-linear relationships
o Transformation –converting a numeric variable to a binary
indicator
o Model specification –adding interaction effects
o Putting it all together-an improved regression model
 Understanding regression trees and model trees
 Adding regression to trees
 Estimating the quality of wines with regression trees and model trees
 Collecting data
 Exploring and preparing the data
 Training a model on the data
o Visualizing decision trees
 Evaluating model performance
o Measuring performance with mean absolute error
 Improving model performance

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

FINDING PATTERNS - MARKET BASKET ANALYSIS USING
ASSOCIATION RULES:
 Understanding Association Rules
 The Apriori algorithm for association rule learning
o Measuring rule interest –support and confidence
o Building a set of rules with the Apriori
 Identifying frequently purchased groceries with association rules
 Collecting data
 Exploring and preparing the data
o Data preparation – creating a sparse matrix for transaction
data
o Visualizing item support –item frequency plots
o Visualizing transaction data-plotting the sparse matrix
 Training a model on the data
 Evaluating model performance
 Improving model performance
o Sorting the set of association rules
o Taking subsets of association rules
o Saving association rules to a file or data frame

FINDING GROUPS OF DATA - CLUSTERING WITH K-MEANS:

 Understanding Clustering
 Clustering as a machine learning task
 The K-means algorithm for clustering
o Using distance to assign and update cluster
o Choosing the appropriate number of cluster
 Finding teen market segments using K-means clustering
 Collecting data
 Exploring and preparing the data
o Data preparation –dummy coding missing values
o Data preparing –imputing missing values
 Training a model on the data
 Evaluating model performance
 Improving model performance

EVALUATING MODEL PERFORMANCE:

 Measuring Performance for Classification
 Working with classification prediction data in R
 A closer look at confusion matrices
 Using confusion matrices to measure performance
 Beyond accuracy – other measure of performance
o The kappa statistic
o Sensitivity and specificity
o Precision and recall
o The F- measure

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

  Visualizing performance TRADEOFFS
o ROC curves
 Estimating future performance
 The holdout method
 Cross-validation
 Bootstrap sampling

IMPROVING MODEL PERFORMANCE:

 Tuning Stock Models for Better Performance
 Using caret for automated parameter tuning
o Creating a simple tuned model
o Customizing the tuning process
 Improving Model Performance with Meta – Learning
 Understanding ensembles
 Bagging
 Boosting
 Random forests
o Training random forests
o Evaluating random forest performance

DEEP LEARNING:
 Installation of Theano, TensorFlow, Keras, OpenCV
 Relating Deep Learning and Traditional Machine Learning
 Basics of Neural Networks
 Artificial Neural Networks
 Deep Neural Networks
 Convolutional Neural Networks
 Recurrent Neural Networks
 Deep learning with Theano
 Deep Learning with TensorFlow
 Deep Learning with Keras
 Deep Learning with OpenCV
 Implementation of Deep learning

ARTIFICIAL INTELLIGENCE:
 AI Introduction
 AI Intelligent Systems
 AI Popular Search Algorithms
 AI Fuzzy Logic Systems
 AI Natural Language Processing
 AI Robotics
 AI Neural Networks

INTRODUCTION TO WEKA
 EXPLORE WEKA MACHINE LEARNING TOOLKIT

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

  Installation of WEKA
 Features of WEKA Toolkit
 Explore & Load data sets in Weka
 PERFORM DATA PREPROCESSING TASKS
 Apply Filters on data sets
 PERFORMING CLASSIFICATION ON DATA SETS
 J48 Classification Algorithm
 Decision Trees Algorithm
 K-NN Classification Algorithm
 Naive-bayes Classification Algorithm
 Comparing Classification Results
 PERFORMING REGRESSION ON DATA SETS
 Simple Linear Regression Model, Multi Linear Regression Model
 Logistic Regression Model, Cross-Validation and Percentage Split
 PERFORMING CLUSTERING ON DATA SETS
 Clustering Techniques in Weka
 Simple K-means Clustering Algorithm
 Association Rule Mining on Data Sets
 Apriori Association Rule Algorithm
 Discretization in the Rule Generation Process
 GRAPHICAL VISUALIZATION IN WEKA
 Visualization Features in Weka
 Visualize the data in various dimensions
 Plot Histogram, Derive Interesting Insights

 Trainer received Masters of Technology in Computer Science &

Engineering from JNTU, MICROSOFT Certified Professional, Certified
from IIT Kanpur & IIT Ropar.
 Having 10+ Years of Experience in Software & Training.
 His experience Includes Managing, Data Processing, Data Cleaning,
Predicting and Analyzing of Large volume of Business Data.
 Expertise in Data Science, Data Analytics, Machine Learning, Deep
Learning, Artificial Intelligence, Python, R, Weka, Data Management &
BI Technologies.
 Having publications and patents in various fields such as machine
learning, data security, and data science technologies.
 Professionally, he is Data Science management consultant with over 7+
years of experience in finance, retail, transport and other industries.

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440

 Best training materials are provided with Lab Exercises, Data sets,
Codes, Quizzes, Case studies on real data.
 For every online session Recorded video & live running notes will
provide.
 Real time Training with live Scenarios and Applications.
 Support in Resume preparation and Interview preparation.
 Conduct Mock interviews through Skype and Telephonic after course
completion.
 You can shift the batch to weekday batches (morning or evening) and
weekend batches.
 Any number of batches can be attend in a year without any extra fees
 Job support for 1 month after successfully placing the candidates.
 Online help on Doubt Clearance, Career Guidance, Resume
Preparation and Interview Preparation.

DATAhill Solutions, Near Malabar Gold, KPHB, Hyderabad. Ph: 9292005440