Chevron Left
Volver a Specialized Models: Time Series and Survival Analysis

Opiniones y comentarios de aprendices correspondientes a Specialized Models: Time Series and Survival Analysis por parte de IBM

4.5
estrellas
71 calificaciones
21 reseña

Acerca del Curso

This course introduces you to additional topics in Machine Learning that complement essential tasks, including forecasting and analyzing censored data. You will learn how to find analyze data with a time component and censored data that needs outcome inference. You will learn a few techniques for Time Series Analysis and Survival Analysis. The hands-on section of this course focuses on using best practices and verifying assumptions derived from Statistical Learning. By the end of this course you should be able to: Identify common modeling challenges with time series data Explain how to decompose Time Series data: trend, seasonality, and residuals Explain how autoregressive, moving average, and ARIMA models work Understand how to select and implement various Time Series models Describe hazard and survival modeling approaches Identify types of problems suitable for survival analysis Who should take this course? This course targets aspiring data scientists interested in acquiring hands-on experience with Time Series Analysis and Survival Analysis.   What skills should you have? To make the most out of this course, you should have familiarity with programming on a Python development environment, as well as fundamental understanding of Data Cleaning, Exploratory Data Analysis, Calculus, Linear Algebra, Supervised Machine Learning, Unsupervised Machine Learning, Probability, and Statistics....

Principales reseñas

MB
6 de may. de 2021

A very well-structured course with useful techniques and detail guidelines. The Python code templates are also really useful when bringing into real-life problems.

GS
15 de may. de 2021

It is a good course to build foundation on the modeling of timerseries data. It will be good to add other lessons for anomaly detection on timeseries.

Filtrar por:

1 - 21 de 21 revisiones para Specialized Models: Time Series and Survival Analysis

por Lam C V D

10 de oct. de 2020

The problem with this course is they use simulated data which cannot cut it. They need to use real life datasets and students given chance on how to do it properly.

por Ashish P

9 de abr. de 2021

Interesting course with a whole bunch of new algorithms! Although great work from the tutor in explaining all those slides and the codes, still sadly, I would again point out that the Accent is really really hard to comprehend, inspite of the fact that English is like my native language.

Secondly, in the latter half of the course, specially in the labs for Arima, Sarima, FB prophet etc. where there is a whole bunch of complex new information to be digested, the pace in the labs feels to be apparently very rushed and haphazard.

There are too many concepts presented together but in the end it remains still quite unclear the sequence in which these methods could be applied to solve real world problems.

Helpful would be to use more real world Data Sets than Toy sets and show the sequence in which all these different Algorithms could be applied together on the same data set, to compare their performances.

Nevertheless, owing to the complexity of the subject, I appreciate the hard work put in by the tutors and the team at coursera and IBM!

Thank you!

por Mohamed G H

26 de feb. de 2021

Not much details but good as an overview on the topic

por Keyur U

24 de dic. de 2020

Toughest of all the 6 courses in the bunch.

por R W

26 de jul. de 2021

This course was added to the Intro to ML certificate. The material is useful for a data analyst/ML practitioner, but the presentation is not at the level of the other courses. The introductory labs introduce the concepts of time series analysis well, with hands-on examples, but the discussion of AR, MA, and ARIMA models is muddled and the labs for these models are not well constructed (this is the only course in this series where I felt I had to go to other sources in order to understand some of the basic concepts) . The course would be improved with a more detailed walk thru of the steps in building ARIMA models (the Box-Jenkins criteria were not covered in lecture?). The prophet module and the DL lessons seem sort of tacked on -- I would have benefitted from more explanation of how to design a DL model to handle a time series analysis. Overall, I think this topic is a good addition to the corpus, but the specific design and presentation of the material is ineffective.

por Rufus T

8 de abr. de 2021

Good course with some useful tips, the Survival part of the course was particularly interesting.

por Adam L

19 de sep. de 2021

1/5 starts

TLDR: instructors do not explain how models work very well, just give ways to apply them

Notebooks are good material however the instructor does not do a good job at all ramping the explanation of model complexity from the lectures to the applications.

A major problem with this course is that the instructors promote a "black box" mentality, that is, do not explain to many lengths how the models work and gloss over many mathematical concepts and tell the users to just trust that it works and implement the API. I do not agree with this method of teaching is it cultivates a dangerous environment for data scientists/ analystics etc. To understand how to implement a model without having a high level understanding of the inner workings is not a practical approach and will lead to catastrophes when being rolled into production in industry.

I would encourage the instructors to fully audit the course material especially for the last 2 weeks of the course and provide more comprehensive material on the math behind the models rather than just referring to wikipedia pages.

por Mehul D S

1 de jul. de 2021

Really great course to start and enhance your ML and Time series analysis. This course will touch base to all different aspects of Time series analysis. Also if you work on project work will help to acquire additional knowledge.

por My B

7 de may. de 2021

A very well-structured course with useful techniques and detail guidelines. The Python code templates are also really useful when bringing into real-life problems.

por Ghada S

16 de may. de 2021

It is a good course to build foundation on the modeling of timerseries data. It will be good to add other lessons for anomaly detection on timeseries.

por SMRUTI R D

24 de nov. de 2021

This is an excellent course covering large areas of Time Series analysis and is a must for any one intending to learn the topics with some detail.

por Altemur Ç

27 de nov. de 2021

Clearly explaind. I am currently working on time series forecasting and predictions. This course helped me a lot about the details of the topics.

por Pavuluri V C

24 de sep. de 2021

this is one the great course i learned. both theoritical and practical went parrallely that made the course much more reliable.

por george s

16 de sep. de 2021

Everything perfect, just content of 3rd week could have better examples or be more explained.

por Juan M

24 de jul. de 2021

Great course, very well taught and topics are useful for future applications

por Luis P S

17 de jul. de 2021

E​xcelente! Recomendable para iniciar en el mundo del Machine Learning.

por Jose M

16 de feb. de 2021

Again, thanks to the instructor in the videos

por Fernandes M R

19 de jun. de 2021

very good. It could be better, but it ok.

por vikas v

22 de nov. de 2020

Amazing Concepts explanations

por Zizhi W

27 de sep. de 2021

This course is good but not as good as the previous five courses in this series. I think the points need more explanation and it might be more suitable to extend this course to more weeks.

Moreover, one has to go through peer review. I really wait for a long time to get reviewed. Moreover, after I put 'my course is going to expire soon', someone just gave all zero to my assignment and I had to reloaded the work again and found new people reviewing it. This is really a terrible experience.

por krysten z

16 de oct. de 2020

not able to cancel the course.