What You Will Learn!
- Principle of Data [strategy, platform, & timeline] and its relation to prediction
- philosophy of math; and its terms as an independent language in relation to prediction
- all the math which you MUST know for creating right predictive models
- practical examples on polynomial, equations, functions, trigonometry, derivatives & integrals, eigen value & vectors, combination & permutation
- the map of prediction which covers the path from analytical models to predictive models
- creation of univariate, bivariate, and multivariate PROBABILITY predictive models
- probabilistic predictive models- for Continues Values using uniform, normal, triangular, logistic Cauchy, exponential distributions
- probabilistic predictive models- for Discrete Values using uniform, binomial, geometric, Bernoulli, Poisson distributions
- creation of simple, multiple linear & nonlinear regression models
- the regularization using LASO & RIDGET for those regression models
- using the polynomial & gaussian normal basis functions transformation for nonlinear predictive models
- creation of classification models using: logistic regression, K-nearest neighbors, decision tree, SVM, & naive Biase
- comprehensive implementation from scratch using the DEEP LEARNING framework aka neural network
- creation of CNN, RNN predictive models from scratch and using ready models
- comprehensive implementation of Reinforcemtn Learning framework from scratch with real example
- creation of clustering models, that is the third elements from the map of prediction
- feature engineering- comprehensive review in terms of data preparation & dimension reduction
- BOOK APPENDIX: Python from the start to the End; beside learning the syntax of the language, exploring libraries as Pandas & NumPy, in addition Tensor Flow.
- All the outlines abstract, math, & theories, are applied using Python coding blocks
Description
This is the third element of -the Big Bang of Data Science-, that is [Prediction from the Start to The End].
I don’t want to stick to that _abstract and direct_ definition from the academic book, on the meaning of prediction, but from the industrial one. So, I believe PREDICTION is the co-concertmaster that sits in the third chair of the highest leadership position among all the other parts that are responsible for the outcome of a product that is good, successful, and intelligent.
The first two elements from -the Big Bang of Data Science-, these are _Research from the Start to the End_ and _Analysis from the start to the End_; were responsible to outcome a product that is GOOD, SUCCESSFUL, of course with main characteristic of being a quality one, the technical name of this product is known as _analytical model_. However, we knew back then that product leaks an important characteristic that to be INTELLIGENT. We have discussed the fact that to enhance that feature of intelligence into the equation of your product, then you must factor in the FUTURE.
The PREDICTION block is responsible for that kind of enhancement. Essentially, this black box comprises specific elements that shape the outcome of the _analytical model_ into a _predictive model_. In other words, the prediction step enhances the future element into the equation of the analytical model.
There are many tools, frameworks, and similar elements of that kind offer variant types of black boxes to do that kind of enhancement; using any of which are of no harm, however, the only disadvantage is the state of dependency. Let me share with you a story I call it __the text editor and you__. Let us assume you have a task to _edit_ a text which you have _authored_. What options are available to accomplish such a task? Obviously, you would employ _text editing tool_ to help you with such task. Let us assume that you have come across specific feature in that editing process and you found that your text editor does not offer such feature. Most probably, you would try to find another tool that has that feature; however, the dilemma arose when you find out that there are no such tool has such feature. Then what would you do? I believe if you have no options, then you would stick within the boundaries of that tool.
That is the same choice you would end up with, if you rely on those black boxes of AI models without comprehensive understanding of their elements. To this end, __if you are seeking a complete state of independence__ being able to __customize, create, or reengineer__ such AI models then you are at the right place.
This book material offers you a comprehensive understanding from __abstract and applied__ perspective on every window to become __AI Models Creator__. Firstly, you will have a clear understanding about mathematics inform of disciplines and as an independent language. Secondly, you will learn about selected outlines from math which you need to learn to make your own predictive model from scratch. Finally, you will enjoy the __map of prediction__ which shows you every possible kind of predictive model and above all you don't only learn the abstract but also the applied using Python.
To this end, the third book is carefully crafted to meet all the requirements to build your product on the right foundation of prediction. Here is a quick view of the content of the book.
### Introduction
1. [✓] Course Strategy
2. [✓] Principles of data
3. [✓] Data Platform
4. [✓] Timeline representative
### The Story of Math
1. [✓] Philosophy of math
2. [✓] Area of mathematics
2.1. ➢ Geometry
2.2. ➢ Algebra
2.3. ➢ Calculus & Analysis
2.4. ➢ Discrete Mathematics
2.5. ➢ Math Logic
2.6. ➢ Decision Science
2.7. ➢ Computational Math
### You must know
1. [✓] Number Properties
2. [✓] The universe of polynomial
3. [✓] Equation & Function & System
4. [✓] Trigonometry
5. [✓] e & Natural Logarithm ln
6. [✓] Exponential Function & Logarithm
7. [✓] Derivatives & Integrals
8. [✓] Matrix, Eigenvalue, Eigenvector
9. [✓] combination and permutation
10. [✓] Python LAB- implementation on the abstract outlines
### Wourld of Prediction
1. [✓] Introduction to Prediction
2. [✓] Map of prediction
3. [✓] Elaboration on the map from left
4. [✓] Elaboration on the map from right
5. [✓] Elaboration on the map from source
### Prediction by Probability
1. [✓] Introduction to Probability
2. [✓] Univariate concept of probability- Discrete value
3. [✓] Univariate concept of probability- Continues value
4. [✓] Bivariate concept of probability
5. [✓] Multivariate concept of probability
6. [✓] Probability predictive model-discrete value using: Bernouil; Binomial; Geometric; Pascal; & Hypermetric distribution
7. [✓] Probability predictive model-continues value using: uniform; exponential; normal; gamma; and cauchy distribution
8. [✓] Python Lab- implementation on every probability model
### Prediction by RCC
1. [✓] Introduction to RCC
2. [✓] Introdution to Regression
2.1. ➢ Simple Linear Regression
2.2. ➢ Multiple Linear Regression
2.3. ➢ Nonlinear Regression
2.4. ➢ Regulization- LASO & RIDGE
3. [✓] Introduction to Classification
3.1. ➢ Logistic Regression
3.2. ➢ K-nearst Neighbor
3.3. ➢ Decision Tree
3.4. ➢ Support Vector Machine
3.5. ➢ Naive Bayse
3.6. ➢ Deep Learning- Introduction
3.7. ➢ Deep Learning- Neural Network Framework
3.8. ➢ Deep Learning- RCC Framework
3.9. ➢ Deep Learning- Reinforcement Learning Framework
3.10. ➢ Deep Learning- RNN Framework
4. [✓] Introduction to Clustering
5. [✓] Feature Engineering intro
5.1. ➢ Data Preparation
5.2. ➢ Feature Reduction
### APPENDIX- Python from the start to the End
1. [✓] Introduction
2. [✓] Setup environment
3. [✓] Informal introduction to python
4. [✓] Control flow tool kit
5. [✓] Data structure
6. [✓] Modules
7. [✓] Input & output
8. [✓] Errors & exceptions
9. [✓] Classes
10. [✓] Tour of the standard libraries
11. [✓] Development tips
12. [✓] NumPy package
13. [✓] Pandas package
14. [✓] Tensor Flow basics
## Who is this book for?
This book is for anyone with the interest in building, creating and producing a professional product that has a future enhancement feature. it's recommended to have basic knowledge about elementary math, research and analysis, with extreme enthusiasm to learn how to make the right decision. So, it is meant for an audience of: (1) students, under or postgraduate. (2) scholars, (3) researchers, (4) scientists, (5) professionals from technical or academic background in IT, computer science or similar domain.
> [!TIP]
> The trainer strongly advice on learning the materials from the first book [Research from the Start to the End]; that can absolutely help you to perform way better in this book.
> The trainer strongly advice on learning the materials from the second book [Analysis from the Start to the End]; that can absolutely help you to perform way better in this book.
# Book competitive advantage
> [!IMPORTANT]
> The main principle of this material is to fix you in the state of independence, where you can build, assemble and create your complete own AI model. For that reason, the main element to accomplish that is comprehensive level of understanding Math, so the first step it shows you how math is an independent language, and then introduces you the main field of math for you to conceptualize its importance.
> Based on the first advantage, the material selects several subjects of math that you must know and master. Those subjects are discussed not from a rigid abstract of math but from a wider level of understanding the use of it as an independent language. Moreover, every line of this subject is applied with real life example and code implementation in Python, in addition to visualization ability.
> The map of prediction is a very unique way this material illustrates the entire world of AI. It shows you the input process and output of this phenomenon. More important, in every possible type of predictive model, e.g. regression, classification or clustering, it implements them using several common types of algorithms and methods from scratch to the end. In this way you will have strong ability from abstract and applied way.
> Unlike many materials that speak about the subject of AI and machine learning, it presents some common frameworks as the basis, this material shows you that you can utilize any of those frameworks, or even re-engineer them your way. For instance, it shows you frameworks of Neural network, CNN, RNN, Reinforcement learning, which are discussed as an independent training programs, but once you master the math of AI you will see that you only need few minutes to build from scratch on your own. Moreover, you will write your own code to do so using Python.
> Finally, if you have come from background with no experience in code writing, this material introduces a whole appendix coaching you on learning Python from the start to the end. so, in this case, you can even procced from the very beginning.
Who Should Attend!
- students- post/undergraduate; scholars, scientists, professionals with IT or Natural science backgrounds