[IBM] What is Data Science? - Deep Learning & Machine Learning

What's the difference?

- BIgn Data: massive, quickly built, vary, perform with relational DB

- Data Mining: the process of automatic searching and analyzing data, discovering patterns, including preprocessing, 

- Machine Learning: a subset of AI, analyze data and make intelligent decisions, trained with large sets of data, learn from examples, allow the machines to solve problems on their own and predict accurately with provided data

- Deep Learning: a specialized subset of machine learning, layered neural networks, label and categorise the information, identity patterns

- Neural Networks: a collection of small computing units(neurons),  plateau as data increases

 

Aritificial intelligence and Data Sicence

Data Science 

- the process and method for extracting knowledge and insights from large volumes of disparate data

- interdisciplinary field involving mathematics, statistical analysis, data visualization, machine learning, and more 

- make appropriate information, see patterns, find meaning from large volumes of data, make decisions that drive business based on it 

- a broad term that encompasses the entire data processing methodology 

 

AI

- include everything that allows the computer to learn how to solve the problem and make intelligent decisions

 

 

Neural Networks and Deep Learning

- Deep Learning is layered neural networks 

- Neural Networks discovered long ago, but due to the computationally intensive

- speech recognition( words, people, images, classifying images etc.) 

- linear algebra to be studied, high-powered computational resources

 

Applications of Machine Learning

- The recommender systems are one of the major applications 

- Classifications, cluster analysis, predictive analytics

- Decision trees, Bayesian, naive 

- with R, just to understand the meaning not how to do 

 

- Recommendations using machine learning 

- fraud detection: you have to learn from all of the transactions that have happened previously and build a model, and when the charge comes through you have to compute all this stuff and say, "Yeah we think that's ok," or "hmm, that's not so good. 

 

 

Regression

- Regression is a statistical method used in finance, investing, and other disciplines that attempt to determine the strength and character of the relationship between one dependent variable (usually denoted by Y) and a series of other variables (known as independent variables).

- Sir Frances Galton in 1886 developed the statistical technique, aka regression

- The use of regression models, or their variants, is ubiquitous.

- The findings may be the one we already know from our everyday experience, but the real value added by the research rested in quantifying the magnitude of those relationships.

 

 

 Watson Visual Recognition

Visual Recognition