2021, May 23 Start: 13:00 CDT Duration: 3 hours online
"Predict Loan Eligibility using Machine Learning Models"
Workshop #4
Mridul Bhandari
Developer Advocate at IBM
Khalil Faraj
Developer Advocate at IBM
AGENDA
2021, May 23-th
13:00 - 13:10
13:00 - 13:10
What is Data Science?
Explaining data science and its use cases and where can it be used and how is it used in today's world
13:10 - 13:25
13:10 - 13:25
Data Scientist vs Data Engineers vs Data Analysts
Talks about the different hats a person dealing with data wears how it goes from analysis phase to an engineering phase then to the machine learning phase
13:25 - 13:45
13:25 - 13:45
Data Science Methodology
The methodology followed when dealing with a data science based problems and what steps are required to overcome each and every step in the pipeline (5 steps)

  1. Identifying the problem and the approach to fix the problem: identify the problem a business is facing and turn it into an analytical approach to fix the problem.
  2. Deduce data requirements and collection methods: look at the approach and figure out what data we need to solve the problem the business is facing.
  3. Understand the data and data preparation: understand the data which is collected and prepare the data, this would include steps such as feature engineering as well which will generate new data from the collected data which we will use.
  4. Generate models and evaluate them: generate machine learning models and evaluate them based on accuracy and errors within the machine learning models generated and decide which model are we going to use
  5. Deploy the model and get feedback: deploy the model into production and then get feedback based on the models prediction and tune the machine learning model accordingly
13:45 - 14:05
13:45 - 14:05
Data Analysis Types
The types of data analysis which a analyst does to gain more insight about the data which they are dealing with

  • Descriptive Analysis: comprehensive and effective to visualize; metrics reports, data mining, aggregation; summary statistics.
  • Diagnostic Analysis: Data collected to drill down to the root cause and analyze it; regression analysis, sensitivity analysis, principle components analysis
  • Predictive Analysis: Using data to predict what is going to happen next, and what the future looks like for the; quantitative analysis, predictive modeling machine learning algorithms
  • Prescriptive Analysis: decides what actions are best based the desired outcomes; recommendation engines, simulation analysis, artificial intelligence, neural networks
14:05 - 14:15
14:05 - 14:15
Break
14:15 - 14:35
14:15 - 14:35
What is Machine Learning
What is machine learning and the fundamentals of machine learning the aim here is to cover the basics of machine learning and the types of machine learning models such as classification, regression, etc. and if required we will add code samples as well.
14:35 - 14:55
14:35 - 14:55
ML Process
The process to create a machine learning pipeline and how to collect, organize, analyze, infuse & modernize data

  • Collect: you extract the data from a data source may that be from various data sources and collecting them into a database or data lake
  • Organize: organize the data into a trusted source and make it business ready with built-in governance, protection, and compliance
  • Analyze: Analyze the collected data and perform cleaning of the data, get the required fields and remove redundant fields and analyze data in smarter ways and and gain new insights
  • Infuse: Apply the collected findings into a machine learning and AI approach and apply it through the organization to aid business processes drawing on predictions, automation and optimization
  • Modernize: make your data and machine learning models ready for cloud/hybrid cloud environment

It is an iterative step to build a machine learning models, we start with selecting the hyperparameters, and then we build the model and assess it based on accuracy and error rate, after which we train the model for the use case which we are trying to solve and lastly we validate the model by testing.

The data is split into various sections based on it size Train-Validate-Test Split:

  • Smaller Datasets: 60-20-20
  • Larger Datasets: 80-10-10
  • Huge Datasets (DL): 98-1-1
14:55 - 15:15
14:55 - 15:15
Data Science Tools
The various data science tools which can be used to perform various techniques such as EDA, ML, and Visualizations Covering libraries such as

  • For EDA and Visualizations: Matplotlib, Pandas, Numpy, Plotly, Seaborn, Bokeh
  • Machine Learning: NLTK, TensorFlow, Pytorch, Apache Spark, Scikit-Learn.
15:15 - 15:25
15:15 - 15:25
Break
15:25 - 16:40
15:25 - 16:40
Code Lab
A code lab predicting loan eligibility using a notebook where we explain various functions used and attendees can run the same.

** Code snippets will be added within the presentation for various topics covered in the outline we will add code snippets as well when it comes to cleaning data, preprocessing data, feature engineering and building visualizations for the exploratory data analysis phases and building machine learning models
16:40 - 17:10
16:40 - 17:10
Quiz
We will conduct a face to face quiz using mentimeter, and during the presentation we will have 2-3 self knowledge checks as well which will allow the attendees to interact.
13:00 - 13:10
What is Data Science?
Explaining data science and its use cases and where can it be used and how is it used in today's world
13:10 - 13:25
Data Scientist vs Data Engineers vs Data Analysts
Talks about the different hats a person dealing with data wears how it goes from analysis phase to an engineering phase then to the machine learning phase
13:25 - 13:45
Data Science Methodology
The methodology followed when dealing with a data science based problems and what steps are required to overcome each and every step in the pipeline (5 steps)

  1. Identifying the problem and the approach to fix the problem: identify the problem a business is facing and turn it into an analytical approach to fix the problem.
  2. Deduce data requirements and collection methods: look at the approach and figure out what data we need to solve the problem the business is facing.
  3. Understand the data and data preparation: understand the data which is collected and prepare the data, this would include steps such as feature engineering as well which will generate new data from the collected data which we will use.
  4. Generate models and evaluate them: generate machine learning models and evaluate them based on accuracy and errors within the machine learning models generated and decide which model are we going to use
  5. Deploy the model and get feedback: deploy the model into production and then get feedback based on the models prediction and tune the machine learning model accordingly
13:45 - 14:05
Data Analysis Types
The types of data analysis which a analyst does to gain more insight about the data which they are dealing with

  • Descriptive Analysis: comprehensive and effective to visualize; metrics reports, data mining, aggregation; summary statistics.
  • Diagnostic Analysis: Data collected to drill down to the root cause and analyze it; regression analysis, sensitivity analysis, principle components analysis
  • Predictive Analysis: Using data to predict what is going to happen next, and what the future looks like for the; quantitative analysis, predictive modeling machine learning algorithms
  • Prescriptive Analysis: decides what actions are best based the desired outcomes; recommendation engines, simulation analysis, artificial intelligence, neural networks
14:05 - 14:15
Break
14:15 - 14:35
What is Machine Learning
What is machine learning and the fundamentals of machine learning the aim here is to cover the basics of machine learning and the types of machine learning models such as classification, regression, etc. and if required we will add code samples as well.
14:35 - 14:55
ML Process
The process to create a machine learning pipeline and how to collect, organize, analyze, infuse & modernize data

  • Collect: you extract the data from a data source may that be from various data sources and collecting them into a database or data lake
  • Organize: organize the data into a trusted source and make it business ready with built-in governance, protection, and compliance
  • Analyze: Analyze the collected data and perform cleaning of the data, get the required fields and remove redundant fields and analyze data in smarter ways and and gain new insights
  • Infuse: Apply the collected findings into a machine learning and AI approach and apply it through the organization to aid business processes drawing on predictions, automation and optimization
  • Modernize: make your data and machine learning models ready for cloud/hybrid cloud environment

It is an iterative step to build a machine learning models, we start with selecting the hyperparameters, and then we build the model and assess it based on accuracy and error rate, after which we train the model for the use case which we are trying to solve and lastly we validate the model by testing

The data is split into various sections based on it size Train-Validate-Test Split:

  • Smaller Datasets: 60-20-20
  • Larger Datasets: 80-10-10
  • Huge Datasets (DL): 98-1-1


14:55 - 15:15
Data Science Tools
The various data science tools which can be used to perform various techniques such as EDA, ML, and Visualizations Covering libraries such as

  • For EDA and Visualizations: Matplotlib, Pandas, Numpy, Plotly, Seaborn, Bokeh
  • Machine Learning: NLTK, TensorFlow, Pytorch, Apache Spark, Scikit-Learn
15:15 - 15:25
Break
15:25 - 16:40
Code Lab
A code lab predicting loan eligibility using a notebook where we explain various functions used and attendees can run the same.

** Code snippets will be added within the presentation for various topics covered in the outline we will add code snippets as well when it comes to cleaning data, preprocessing data, feature engineering and building visualizations for the exploratory data analysis phases and building machine learning models
16:40 - 17:10
Quiz
We will conduct a face to face quiz using mentimeter, and during the presentation we will have 2-3 self knowledge checks as well which will allow the attendees to interact.
WHAT YOU WILL LEARN
In this tutorial, we'll build a predictive model to predict if an applicant is able to repay the lending company or not.
We will prepare the data using Jupyter Notebook and use various models to predict the target variable.
SPEAKER'S BIO
Mridul Bhandari
Developer Advocate at IBM
Mridul Bhandari is a Developer Advocate at IBM with experience in Specialized Delivery with a demonstrated history of working in Information Technology, Project Management and Non-profit Organizations. Strong Program and Project Management Professional graduate with B.E (Hons) Computer Science from Birla Institute Of Technology and Science, Pilani Dubai.
Khalil Faraj
Developer Advocate at IBM
Khalil Faraj is a Developer Advocate at IBM. He studied Computer Science and has a Master's degree in Software Engineering from the University of Southampton. He is a tech enthusiast, an app developer especially in Android, and has experience in multiple programming languages mainly Java and Python.
WORKSHOP FORMAT
The best way to improve your professional skills
Introduction & overview - 10 Minutes
Code Lab — 75 Minutes
Basics of Data Science, Machine Learning — 80 min
Quiz & Q&A — 15 minutes
FOR WHOM?
Loans are the core business of banks. The main profit comes directly from the loan's interest. The loan companies grant a loan after an intensive process of verification and validation. However, they still don't have assurance if the applicant is able to repay the loan with no difficulties.
In this tutorial, we'll build a predictive model to predict if an applicant is able to repay the lending company or not. We will prepare the data using Jupyter Notebook and use various models to predict the target variable.

- Students who are interested in AI, Data Science but don't know where to start
- Professional Developers who want to know more about the world of Data & AI
- Anyone who wants to do Data Cleaning without code
- Basic to Intermediate audience
Loans are the core business of banks. The main profit comes directly from the loan's interest. The loan companies grant a loan after an intensive process of verification and validation. However, they still don't have assurance if the applicant is able to repay the loan with no difficulties.
In this tutorial, we'll build a predictive model to predict if an applicant is able to repay the lending company or not. We will prepare the data using Jupyter Notebook and use various models to predict the target variable.

- Students who are interested in AI, Data Science but don't know where to start
- Professional Developers who want to know more about the world of Data & AI
- Anyone who wants to do Data Cleaning without code
- Basic to Intermediate audience
SHOW MORE
WHAT YOU NEED BEFORE WE START:
1
Basic knowledge of Python
2
IBM cloud
COURSE BENEFITS:
Our aim with this tutorial is for the attendees with basic to intermediate experience in data science to understand the concepts and get hands-on experience with a simple example that will give them an overview and spark their interest to learn more.
PURCHASE INFO
$199
"Predict Loan Eligibility using Machine Learning Models"
Workshop #4
by Mridul Bhandari&&Khalil Faraj
24:45:54
workshop price
OR
Workshop #4
"Predict Loan Eligibility using Machine Learning Models"
Responsible for payment:
Additional information
Workshop #4 - "Predict Loan Eligibility using Machine Learning Models" - Waiting List

Geekle Corp.
910 Foulk Road, Suite 201
Wilmington, DE
19803, USA
Designed by: