Definitions Flashcards
Define AÍ
Human intelligence exhibit by machines
Machine learning
An approach to achieve AI
Involves teaching a machine to recognize patters by example
Creating algorithm that learn complex functions from data and make predictions on it
Takes data
Learns patters
Classifies new data based on what was learn
Deep learning
Técnica para implementar machine learning atrás do uso de DNA- deep neural networks
Data science
Conjunto de princípios que suportam e guiam a extração de informação e conhecimento dos dados
Data mining
E a aplicação de algoritmos aos dados de forma a obter conhecimento; incorpora os princípios de data science.
Secondary analysis of large databases in order to find unsuspected relationships which are of the interest of data owner
Process of discovering interesting patters and knowledge from large amounts of data
Data eng vs data scientist vs business analyst
Data eng - creates the data foundation. creates the database from scratch; they design the way data will be retrieved, processed, and consumed.
Data scientist- focus em optimizar data; data modeling e algoritmos;
Business analyst- business acumen and can communicate with c suit and data scientist to help data driven decisions
Primary data vs secondary data
Primary- data collected to answer a specific goal
Secondary - data collected for any purpose from which we try to get information from. Data mining takes this one
Feature vs label
Feature são características - what we will use to identify the label
Label- what we want to predict
Feature is used to identify label
A real estate agency wants to estimate the price range for each customer based on their income;what is the feature and label
Feature is the income
Label is is which price of the house can we afford based on income
Inductive vs deductive models
Deductive: All men are mortal. Joe is a man. Therefore Joe is mortal.
(In deductive reasoning, a conclusion is reached by applying general rules to specific instances)
Inductive: This cat is black. That cat is black. A third cat is black. Therefore all cats are black.
(In inductive reasoning, the conclusion is reached by extrapolating from specific cases to general
Over fitting
Overfitting is the production of an analysis that corresponds too closely or exactly to a particular set of data, and fails to fit additional data or predict future observations reliably;
3 types of data set when testing a module
Training to develop module
validation test module
test apply module to real data
Training set ..The bigger, the better is what?
Classifier
Validation set ..The bigger, the better is what?
Better estimatation of optimal training
Test set ..The bigger, the better is what?
Performance of classifier
