Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

IISPL Noida Data Analytics Machine Earning Module

180 views

Published on

IISPL Noida Data Analytics Machine Earning Module
ISPL is going to start a batch of Data Analytics and Machine Learning  from coming Saturday, (26th August 2017)

To be familiar with the conceptual understanding of Data Analytics and its need in current business scenario

We will cover following topics

● Data Challenges

● Process challanges

● Management challanges

● Big data analytics

● Statistical & mathematical modeling techniques

For any further information

Feel free to contact us : 8447460060
8860352949

" Regards "
IISPL Academy

Published in: Education
  • Be the first to comment

  • Be the first to like this

IISPL Noida Data Analytics Machine Earning Module

  1. 1. Data Analytics & Machine Learning PROF. (Dr.) S. PATHAK , PH.D, M.TECH, Senior Data Scientist & Er. J.K. JHA ( Corporate Trainer, BIG Data & Machine Learning ) Reach: info@iispl.co.in www.iispl.co.in IISPL ACADEMY
  2. 2. Introduction to Analytics & Data Analysis tools  What is data analytics?  Importance of analytics.  Introduction to various analysis techniques  Applications of data analysis in various industries  Introduction to SAS/R/Python/SPSS  Basics of programing in SAS/R/Python/SPSS  Data handling in SAS/R/Python/SPSS  BI reporting in SAS/R/Python/SPSS  Performing statistical analysis on SAS/R/Python/SPSS Analyzing the data with simple descriptive statistics  Variance and standard deviation IISPL ACADEMY
  3. 3. Data Validation & Cleaning  Introduction to validating and cleaning data  Examining data errors when reading raw data files  Validating data with the CONTENTS, PRINT, FREQ, MEANS and UNIVARIATE procedures.  Cleaning invalid data: Missing value identification and treatment.  Outlier identification and treatment  Project Work IISPL ACADEMY
  4. 4. Introduction to machine learning:  What is machine learning?  Learning system model  Training and testing  Performance  Algorithms  Machine learning structure  What are we seeking?  Learning techniques IISPL ACADEMY
  5. 5. Nearest neighbor classification:  Instance based classifiers  Nearest-Neighbor classifiers  Lazy vs. Eager learning  k-NN variations  How to determine the good value for k  When to consider nearest neighbors  Condensing  Nearest neighbor issues  Project Work IISPL ACADEMY
  6. 6. IISPL ACADEMY Naive Bayes classification  Naive Bayes learning  Conditional probability  Bayesian theorem: basics  The Bayes classifier  Model parameters  Naive Bayes training  Types of errors  Sensitivity and specificity  ROC curve  Holdout estimation  Cross-validation
  7. 7. Decision Trees - Part I  Key requirements  Decision tree as a rule set  How to create a decision tree  Choosing attributes  ID3 heuristic  Entropy  Pruning trees - Pre and post  Subtree Replacement  Raising Decision Trees - Part II  Tree induction  Splitting based on ordinal attributes  How to determine the best split  Measure of impurity: GINI  Splitting based on GINI  Attributes binary  Categorical -GINI  Strengths and weakness of decision trees IISPL ACADEMY
  8. 8. Ensemble Approaches  Ensemble approaches  Bagging model  Boosting  The Ada Boost algorithm  Gradient boosting  Random forests  RIF  RIC  Advantages  Disadvantages IISPL ACADEMY
  9. 9. Artificial Neural Network  Background of brain and neuron  Neural networks  Neurons diagram  Neuron models- step function  Ramp func etc  Perceptrons  Network architectures  Single-layer feed-forward Artificial Neural Network continued  Multi layer feed-forward NN (FFNN)  Back propagation  NN design issues  Recurrent network architecture  Supervised learning NN  Self organizing map  Network structure  SOM algorithm IISPL ACADEMY
  10. 10. Project I  Mentee can select project from predefined set of AcadGild projects or they can come up with their own ideas for their projects  Mentee can select project from predefined set of AcadGild projects or they can come up with their own ideas for their projects IISPL ACADEMY
  11. 11. Support Vector Machine Classifiers  Support vector machines for classification  Linear discrimination  Nonlinear discrimination  SVM mathematically  Extensions  Application in drug design  Data classification  Kernel functions  Project IISPL ACADEMY
  12. 12. Linear Models in R  Introduction to regression  Why do regression analysis  Types of regression analysis  OLS regression  Dependent and independent variable(s)  Steps to implement a regression model  Simple linear regression  Understanding terminology of each of the output of linear regression  Project IISPL ACADEMY
  13. 13. Correlation and Regression  Correlation  Strength of linear association  Least-squares or regression line  Linear regression model  Correlation coefficient R  Multiple regression  Regression diagnostics Assumptions in Regression Analysis  The assumptions  Assumption 1 and explanation- residuals and non normality  Assumption 2 and explanation- heteroscedasticity  Assumption 3 and explanation- additivity  Assumption 4 and explanation- linearity ; Independence assumption; Residual plots  Project IISPL ACADEMY
  14. 14. Model Selection in R  Fitting the model  Diagnostic plots  Comparing models  Cross validation  Variable selection  Relative importance  AIC  Dummy variable  Box cox transformations Creating the model  Residuals vs fitted  Residuals vs regression  Diagnostic plots IISPL ACADEMY
  15. 15. Logistic Regression  Binary response regression model  Linear regression output of proposed model  Problems with linear probability model  Logistic function  Logistic regression & its interpretation  Odds ratio  Goodness of fit measures  Confusion matrix  What is cluster analysis?  Project IISPL ACADEMY
  16. 16. Introduction to Cluster Analysis  Types of data in cluster analysis  A categorization of major clustering methods  Partitioning methods  Hierarchical methods  Density-based methods  Grid-based methods  Model-based clustering methods  Supervised classification  Project IISPL ACADEMY
  17. 17. Principal Component Analysis (PCA)  Curse of dimensionality  Dimension reduction  Why factor or component analysis?  Principal component analysis  PCs variance and least-squares  Eigenvectors of a correlation matrix  Factor analysis  PCA process steps  Project IISPL ACADEMY
  18. 18. Forecasting Principles  Basic time series and it's components  Moving averages (simple & exponential)  R'Â’s inbuilt function ts()  Plotting of time series  Business forecasting using moving average methods  The ARIMA model  Application of ARIMA model in business  Project IISPL ACADEMY
  19. 19. IISPL ACADEMY Final Project Project Work With Sophisticated Statistical and Mathematical Tools & Techniques
  20. 20. Thanking You !! With You Until Success & Beyond… IISPL ACADEMY

×