Explore
For Enterprise
Join for Free
Log In

Imperial College London
Design and Critique Epidemiological Research
Develop skills that will allow you to conduct and critically assess research in epidemiology.
deeplearning.ai
New! TensorFlow: Data and Deployment Specialization
Join Andrew Ng and Laurence Moroney in the latest Specialization from deeplearning.ai.

Explore todo o Coursera

Navegar
Os melhores cursos
For Enterprise
Entrar
Inscreva-se gratuitamente

Segment 8 - Bias and Variance in your Data

Loading...

Project Planning and Machine Learning

Universidade do Colorado em Boulder

4.5 (17 classificações) | 1.1K alunos inscritos

Curso 2 de 3 no Developing Industrial Internet of Things Programa de cursos integrados

Cadastre-se gratuitamente

This course can also be taken for academic credit as ECEA 5386, part of CU Boulder’s Master of Science in Electrical Engineering degree. This is part 2 of the specialization. In this course students will learn : * How to staff, plan and execute a project * How to build a bill of materials for a product * How to calibrate sensors and validate sensor measurements * How hard drives and solid state drives operate * How basic file systems operate, and types of file systems used to store big data * How machine learning algorithms work - a basic introduction * Why we want to study big data and how to prepare data for machine learning algorithms

Visualizar o programa do curso

Avaliações

4.5 (17 classificações)

5 stars
11 ratings
4 stars
3 ratings
3 stars
3 ratings

Na lição

Big Data Analytics

In this module you will learn about big data and why we want to study it. You will learn about issues that can arise with a data set and the importance of properly preparing data prior to a ML exercise.

Introduction0:33

Segment 1 - Learning Outcomes, Definition of Big Data3:22

Segment 2 - Importance of Big Data, Characteristics of Big Data4:57

Segment 3 - Size of Big Data4:56

Segment 4 - Introduction to Predictive Analytics2:41

Segment 5 - Role of Statistics and Data Mining3:01

Segment 6 - Machine Learning, Generalization and Discrimination7:55

Segment 7 - Frameworks, Testing and Validating5:32

Segment 8 - Bias and Variance in your Data3:24

Segment 9 - Out-of-sample Data and Learning Curves5:07

Segment 10 - Cross Validation5:27

Segment 11 - Model Complexity, Over- and Under-fitting3:35

Segment 12 - Processing Your Data Prior to Machine Learning8:35

Segment 13 - Good Data, Smart Data6:07

Segment 14 - Visualizing Your Data1:23

Segment 15 - Principal Component Analysis (PCA)2:37

Segment 16 - Prognostic Health Management, Hadoop Machine Learning Library11:01

Segment 17 - My Example: Predicting NFL Football Winners18:17

Segment 18 - Tom Bradicich, Hewlett Packard's Viewpoint on Big Data20:52

Ministrado por

David Sluiter
Professor Adjunct

Experimente o curso Gratuito

Transcrição

So, the segment here, I'm talking about what can go wrong with your data. Sample bias, this is something to be on the lookout for. Biases and offset in the data, it's also known as selection bias. A famous example was on the 1936 US presidential election, it was Alfred Landon versus Franklin Deleanor Roosevelt in that election. A respectable magazine at the time polled 10 million names from every telephone directory, plus magazine subscription lists and clubs, and polled them all and got 2.4 million responses. From that data, the poll indicated that 57% would vote for Landon and yet Roosevelt won with 62% of the vote. So, what happened? It was the largest error ever for a public opinion poll. Any idea what might have gone wrong there? In 1936, the middle to the end of the depression, having a phone and magazine subscriptions, and belonging to a club meant that you were affluent, you were wealthy, you were rich, and so the samples were pulled from only effluent voters. So, that entire dataset was skewed way in favor of Landon because that's what they predicted. They polled all these people, and the poll said that Landon was going to win. Roosevelt actually won with 62%. That's an example of how bias in your data can affect your outcome. Point is if all your training data is bias, your results are going to be biased, and you're not going to be satisfied at all with your results. Sample variance, this is another challenge. So, it's the average of the squared differences from the mean. So, all of you have had statistics and seen this before. U is mean, and these are all your values of your data, and so you take each one and subtract it from the mean and square it, sum that all up and take one over n and that's how you calculate the variance. So, use of mean and variance measures how far a dataset is spread out. Dataset with low variance is very tightly packed together, and a dataset that has wide variance is spread out. It's important to be aware of this to understand which features may be the important one in a multi-dimensional dataset, and even features along one axis in a multi-dimensional dataset, you may have some called outliers. That's coming up here in a second. It may indicate if your model has sensitivities to variance if the model starts to behave or act erratically. It may also indicate that some samples may be just completely invalid.

Explore nosso catálogo

Registre-se gratuitamente e obtenha recomendações, atualizações e ofertas personalizadas.

O Coursera proporciona acesso universal à melhor educação do mundo, fazendo parcerias com as melhores universidades e organizações para oferecer cursos on-line.

© 2020 Coursera Inc. Todos os direitos reservados.

Baixar no Google Play

Coursera
Sobre
Liderança
Carreiras
Lista de cursos
Certificados
Certificados MasterTrack™
Notas
For Enterprise
For Government
For Campus

Comunidade
Learners
Partners
Developers
Beta Testers
Translators

Conectar
Blog
Facebook
LinkedIn
Twitter
Instagram
YouTube
Tech Blog

Mais
Termos
Privacidade
Ajuda
Acessibilidade
Imprensa
Contato
Diretório
Afiliados