Chevron Left
返回到 Machine Learning: Classification

學生對 华盛顿大学 提供的 Machine Learning: Classification 的評價和反饋

4.7
3,470 個評分
577 條評論

課程概述

Case Studies: Analyzing Sentiment & Loan Default Prediction In our case study on analyzing sentiment, you will create models that predict a class (positive/negative sentiment) from input features (text of the reviews, user profile information,...). In our second case study for this course, loan default prediction, you will tackle financial data, and predict when a loan is likely to be risky or safe for the bank. These tasks are an examples of classification, one of the most widely used areas of machine learning, with a broad array of applications, including ad targeting, spam detection, medical diagnosis and image classification. In this course, you will create classifiers that provide state-of-the-art performance on a variety of tasks. You will become familiar with the most successful techniques, which are most widely used in practice, including logistic regression, decision trees and boosting. In addition, you will be able to design and implement the underlying algorithms that can learn these models at scale, using stochastic gradient ascent. You will implement these technique on real-world, large-scale machine learning tasks. You will also address significant tasks you will face in real-world applications of ML, including handling missing data and measuring precision and recall to evaluate a classifier. This course is hands-on, action-packed, and full of visualizations and illustrations of how these techniques will behave on real data. We've also included optional content in every module, covering advanced topics for those who want to go even deeper! Learning Objectives: By the end of this course, you will be able to: -Describe the input and output of a classification model. -Tackle both binary and multiclass classification problems. -Implement a logistic regression model for large-scale classification. -Create a non-linear model using decision trees. -Improve the performance of any model using boosting. -Scale your methods with stochastic gradient ascent. -Describe the underlying decision boundaries. -Build a classification model to predict sentiment in a product review dataset. -Analyze financial data to predict loan defaults. -Use techniques for handling missing data. -Evaluate your models using precision-recall metrics. -Implement these techniques in Python (or in the language of your choice, though Python is highly recommended)....

熱門審閱

SM
2020年6月14日

A very deep and comprehensive course for learning some of the core fundamentals of Machine Learning. Can get a bit frustrating at times because of numerous assignments :P but a fun thing overall :)

SS
2016年10月15日

Hats off to the team who put the course together! Prof Guestrin is a great teacher. The course gave me in-depth knowledge regarding classification and the math and intuition behind it. It was fun!

篩選依據:

376 - Machine Learning: Classification 的 400 個評論(共 545 個)

創建者 Shashidhar Y

2019年4月2日

Nice!!

創建者 Md. T U B

2020年9月2日

great

創建者 Subhadip P

2020年8月4日

great

創建者 Nicholas S

2016年10月7日

Great

創建者 李真

2016年3月5日

great

創建者 Saurabh A

2020年9月11日

good

創建者 SUJAY P

2020年8月21日

nice

創建者 ANKAN M

2020年8月16日

nice

創建者 Dr S J

2020年6月19日

good

創建者 AMARTHALURU N K

2019年11月24日

good

創建者 RISHI P M

2019年8月19日

Good

創建者 Akash G

2019年3月10日

good

創建者 xiaofeng y

2017年2月5日

good

創建者 Kumiko K

2016年6月5日

Fun!

創建者 Arun K P

2018年10月17日

G

創建者 Navinkumar

2017年2月22日

g

創建者 MARIANA L J

2016年8月12日

The good:

-Good examples to learn the concepts

-Good organization of the material

-The assignments were well-explained and easy to follow-up

-The good humor and attitude of the professor makes the lectures very engaging

-All videolectures are small and this makes them easy to digest and follow (optional videos were large compared with the rest of the lectures but the material covered on those was pretty advanced and its length is justifiable)

Things that can be improved:

-In some of the videos the professor seemed to cruise through some of the concepts. I understand that it is recommended to take the series of courses in certain order but sometimes I felt we were rushing through the material covered

-I may be nitpicking here but I wish the professor used a different color to write on the slides (the red he used clashed horribly with some of the slides' backgrounds and made it difficult to read his observations)

Overall, a good course to take and very easy to follow if taken together with the other courses in the series.

創建者 Hanif S

2016年6月2日

Highly recommended course, looking under the hood to examine how popular ML algorithms like decision trees and boosting are actually implemented. I'm surprised at how intuitive the idea of boosting really is. Also interesting that random forests are dismissed as not as powerful as boosting, but I would love to know why! Both methods appear to expose more data to the learner, and a heuristic comparison between RF and boosting would have been greatly appreciated.

One can immediately notice the difference between statistician Emily, who took us through the mathematical derivation of the derivative (ha.ha.) function for linear regression (much appreciated Emily!), and computer scientist Carlos, who skipped this bit for logistic regression but provided lots of verbose code to track the running of algorithms during assignments (helps to see what is actually happening under the hood). Excellent lecturers both, thank you!

創建者 Amilkar A H M

2017年11月27日

It's a great course, but the programming assignments are a little too guided. That is good, to some extend, as it allows you to focus on the concepts, but at the same time, it leave little space for actually practicing your coding skills. I know they said from the beginning that this course was not focused on the implementation of the algorithms, however, how are you going to be able to use what you've learned without knowing how to implement the algorithms on your own.

When it comes to coding, nothing replace implementing the algorithms yourself. That is my only complaint. Other than that, it's great. I loved it. The concepts were well explained and they covered a lot of material. I wish they had spent more time in certain topics, but I guess this is just an introduction. Anyway, take this course by any means if you have some programming experience and have little to no machine learning knowledge.

創建者 Daniel C

2016年4月24日

This series is taught by Emily and Carlos. Course 2 was Emily and this course 3 is Carlos. Carlos takes a more practical approach by showing how things are related using pictures, trial and error, what happens when we do this vs. that. Emily on the other hand dives down into the math and actual facts. I feel Emily is more difficult overall - but once I got through it, I had a better foundation and intuition as to how things work and better overall understanding. So - giving this class 4 stars as compared to Emily's class that is 5 stars. I feel if they would mix it with Emily doing the math immediately followed by Carlos explanations it would be best. Finally - I don't feel this course on classification had as much content. We could've done more.

創建者 Jaiyam S

2016年4月24日

Thank you Prof. Carlos for this amazing course. You covered the topics in a very easy to understand way and the course was full of cool applications and humor! The only downside that I felt was that the programming assignments sometimes felt too easy. Even as a complete Python novice (I started learning Python with the first course), I felt the programming assignments could have been made more interesting. But in the larger scheme of things it doesn't matter because the course was really well taught and easy to understand. I'm really looking forward to the next course! :)

創建者 Lech G

2016年4月26日

Not as good as the Regression Course, but still very good.

While I appreciate prof Guestrin's enthusiasm, I missed a little rigor and mathematical depth of the Regression's course by prof. Fox.

I learned a lot, but I feel that regression clicked with me a little better than classification.

But that's probably me.

In either case, the whole series are awesome so far, better, in my opinion, than Anrdrew Ng's ML course on coursera,

A small suggestion would be to switch the main toolset from the Graphlab to something more common, like Sci-kit learn and Pandas.

創建者 Alessio D M

2016年4月17日

The course is definitely high-quality and the topics are covered in a good way. I'm not giving 5 stars because I would have expected SVMs and neural networks. Mentioning the many different algorithms for learning decision trees would have been nice, without necessarily focusing on each of them in depth. An entire week spent on precision/recall seems a little bit too much, without touching other metrics like F-score. Overall though a very nice course for beginners, and it definitely gives a good sense of classification challenges and approaches.

創建者 SUBIKESH P S

2020年6月11日

This course was very useful for learning machine learning, as this describes classification models deeply and also about other important ML techniques like Online Learning, handling missing data, precision-recall, etc. The weekly programming assignments were elaborate and explained all the topics nicely. The classes were also made interesting by Mr. Carlos by cracking puns in between.

The only problem I face is that using turicreate over sklearn. Since turicreate is depreciated for windows, it's hard to complete programming assignments.

創建者 Anjan P

2016年4月29日

Excellent course that details important concepts in supervised classification. The programming assignments can be a little easy to complete (and consequently easy to forget later), but I believe it's a well paced course and the lecture material is at any incredibly accessible pace, with options for more advanced material.

One suggestion would be to include more papers for additional technical details in the lecture or programming assignments as you did with dealing with unbalanced data.