*My takeaways:* 1. DIstance 9:30 2. k-means algorithm 17:03 - How to choose k 23:57 - Unlucky initial centroids 25:56 - An example 28:58 - Scale data into the same range 37:11
The explanation level of this video is by far the best I have ever watched. Prof. Guttag does a very good job in explaining every concept more clearly.
it just struck me, after all these lecture videos, that professor Guttag is actually using a classic positive reinforcement technique to make the students more attentive and responsive in class by giving out candies for correct answer. lol! and i am not sure if its the result of this or something else but the students seem wayyy too eager to answer questions in this paticular lecture video!
29:06 Isn’t the heart attack example a case of supervised learning, since we have the labels? 1:59 At the start of the lecture, the professor mentioned clustering as an example of unsupervised learning
When we are clustering the airports, the professor only stopped to think about linkage when he arrived at Denver. Shouldn't we have thought about it since the beginning of the clustering? If so, we could have gotten (BOS, SF) instead of (BOS, NY) for the first iteration using complete linkage.
Since in the first iteration there are a number of clusters equal to the number of cities, wouldnt complete linkage be the same as single linkage, given there is only one point of measurement for each cluster? I didnt go back to check, but perhaps after the second iteration there wouldve been some different answers?
The guys who down voted this video must had nothing better to do. The lecture was nicely paced and I think he already made the problem as clear as it can get. Anyway, that was a great lecture. A big thank you to Professor Guttag and the MIT OpenCourseWare team.
Great lecture ! I attended the Clustering lecture by prof Ayan Seal today (even though I dont have the course : Introduction to Data Science) , he didn't focus a lot on code, but had similar things to share about clustering !
Pretty sure he is talking about the number of comparisons which need to occur to create the group, n-squared meaning the number of comparisons is on the order of the square of the number of objects to compare, and n-cubed on the order of the cube of the number of objects to compare. Sort of like big-O notation.
This is not always the case (depends on the code), but it might help to think of n-squared as 2 nested loops and n-cubed 3 nested loops. For instance, in a n-squared algorithm you have n items where, for each item, you make n comparisons. Imagine a really big n.
What are some some methods to evaluate the quality of the clusters, if we do not have an outcome variable? In the example they were evaluated based in part based on whether the subjects in the cluster died at a higher rate. What do I do if I don't have an outcome to look at, only characteristics? For context, I'm creating cognitive style groups based on user data for an insurance company, and these styles will be later used for morphing, churn etc. but do not have an outcome variable per se.
To show new mean is zero: mean = (sum of vals)/n, vals is then transformed (by code) such that centered_normalized_vals = new_vals = (vals - mean)/sd, so new_mean = (sum of (new_vals - mean))/n = (sum of vals - n*mean) / (sd*n) = (n*mean - n*mean) / (sd*n) = 0. To show sd becomes 1: originally sd is calculated as root ((1/n)*sum over every val of ((val - mean)^2)), vals is transformed (by code) such that centered_normalized_vals = new_vals = (vals - mean)/sd, so new_sd = root ((1/n)*sum over every new val of ((new_val - new_mean)^2)), new_mean was previously shown to be zero, so this means new_sd = root ((1/n)*sum over every new val of (((val - mean)/sd)^2)) = root ((1/n)*(1/sd^2)*sum over every val of ((val - mean)^2)), since (by our initial definition) sum over every val of ((val - mean)^2) = n*sd^2, this tells us that new_sd = root ((1/n)*(1/sd^2)*n*(sd^2)) = new_sd = 1.
Thanks for an amazing lecture! @29:35 it tries to cluster data into two groups and see if it correctly differentiated people who dies of heart attack and those that didn't. To me this is using clustering for classification task, if yes, when would someone use clustering rather than classification?
Usually you will use clustering as an unsupervised project wherein you do not have a label. Clustering can be used a first line of segmentation of your dataset that lacks the outcome label for which a classification model instead utilizes.
Main issues of K-Means : choosing the number of clusters (k) and data scaling: But what if one wants to apply weights to the features (parameters)? Should you just multiply the features with the desired coefficients?
at 46:50 the professor mentions “has pretty good specificity, or positive predictive value, but its sensitivity is lousy.” can someone explain how specificity = ppv? im assuming: ppv = tp/(tp+fp) specificity = tn/(tn+fp) doesnt ppv = precision?
14:20 the distance from Denver to Seattle is 1307 and the distance from Denver to Boston is 1949, so why he clustered Denver to Seattle instead of Boston when using Complete linkage? should it not be clustered to the greatest distance?
No, for complete linkage you look up, as you said the greatest distance of each cluster to the datapoint and then cluster it with the smallest out of these distances.
Ok, by minute 7 my mind is wondering if there's going to be a bonus assignment to find the probability that Professor Guttag will correctly throw you the piece of candy on the first try. The odds of you catching it greatly increase the closer your sit to the front center of the room.
Hello, I come from wet lab and I am not familiar with machine learning. But I am really interested in this topic since I want to apply machine learning to my research in plant genetics. I have watched this video several times but still I have not gotten all the things the professor mentioned. I wonder if the author or anyone can share the lecuter or books in this topic. It will mean alot to me. Thank you in advance.
A late response but here we go. I would suggest you read the 100 pages machine learning book, it doesn't really really assume any background but of course having it makes things easier.
As the video description states, the course name is "Introduction to Computational Thinking and Data Science" as it was taught in the Fall of 2016 by the Massachusetts Institute of Technology. For more information, see the course on MIT OpenCourseWare at: ocw.mit.edu/6-0002F16.
"Clustering" is usually taught to "signal" "alumni" that anyone "in their *network*" can't learn and be good at some skills because some Terrorists in their "*network*" may be affected andor effected.
The textbook is: Guttag, John. Introduction to Computation and Programming Using Python: With Application to Understanding Data. 2nd ed. MIT Press, 2016. ISBN: 9780262529624. See the Readings section for more details: ocw.mit.edu/6-0002F16. Best wishes on your studies!
Take care, students, with democrat teachers in computer science classes. They don't care to play with you and call you a dumb if you are a republican and later ask you to choose who is the dumb and who is the smart. I hope you grades doesn't be influencied by you political bias.
*My takeaways:*
1. DIstance 9:30
2. k-means algorithm 17:03
- How to choose k 23:57
- Unlucky initial centroids 25:56
- An example 28:58
- Scale data into the same range 37:11
thank you
@@adiflorense1477 you’re welcome
The explanation level of this video is by far the best I have ever watched. Prof. Guttag does a very good job in explaining every concept more clearly.
Professor gifting the ones who contribute to this lecture. Loved that👏💝
it just struck me, after all these lecture videos, that professor Guttag is actually using a classic positive reinforcement technique to make the students more attentive and responsive in class by giving out candies for correct answer. lol! and i am not sure if its the result of this or something else but the students seem wayyy too eager to answer questions in this paticular lecture video!
It's Skinner all the way down ;-)
I do it in my classes too.
I bet he would reward any good try - not just correct answers
6.006 they gave out cushions for good answers cos the benches were hard.. got the carrot and stick going on at MIT XD
Sheldon would approve
K-Means at16.30: one of the very best algorithms in IA
Thanks
I wish I get the opportunity to sit in a class at MIT someday! Such brilliant minds
You're just as smart as everyone in the audience. The profs have been doing this for decades.
so enjoy your sit
Yeah, making a habit at doing something for an extended period creates excellence. Just stick at.
I cant believe I am binge watching MIT lectures. I wish I had a chance to attend MIT back then.
Same! But feels great that we get all this for free, its a privilege
same haha!!
29:06 Isn’t the heart attack example a case of supervised learning, since we have the labels?
1:59 At the start of the lecture, the professor mentioned clustering as an example of unsupervised learning
When we are clustering the airports, the professor only stopped to think about linkage when he arrived at Denver. Shouldn't we have thought about it since the beginning of the clustering?
If so, we could have gotten (BOS, SF) instead of (BOS, NY) for the first iteration using complete linkage.
Since in the first iteration there are a number of clusters equal to the number of cities, wouldnt complete linkage be the same as single linkage, given there is only one point of measurement for each cluster? I didnt go back to check, but perhaps after the second iteration there wouldve been some different answers?
The guys who down voted this video must had nothing better to do. The lecture was nicely paced and I think he already made the problem as clear as it can get.
Anyway, that was a great lecture. A big thank you to Professor Guttag and the MIT OpenCourseWare team.
Great lecture !
I attended the Clustering lecture by prof Ayan Seal today (even though I dont have the course : Introduction to Data Science) , he didn't focus a lot on code, but had similar things to share about clustering !
16:10 could anyone explain what the professor is talking about when he's mentioning n-squared and n-cubed algorithms ?
Pretty sure he is talking about the number of comparisons which need to occur to create the group, n-squared meaning the number of comparisons is on the order of the square of the number of objects to compare, and n-cubed on the order of the cube of the number of objects to compare. Sort of like big-O notation.
This is not always the case (depends on the code), but it might help to think of n-squared as 2 nested loops and n-cubed 3 nested loops. For instance, in a n-squared algorithm you have n items where, for each item, you make n comparisons. Imagine a really big n.
He was basically talking about the time complexity of both the algorithms.
At 28:00, can anyone help here ? How do we compare this dissimilarity (mentioned in IF statement), in Python. Badly need this.
39:54 I think z-scaling is the same as creating a normally distributed dataset
What are some some methods to evaluate the quality of the clusters, if we do not have an outcome variable? In the example they were evaluated based in part based on whether the subjects in the cluster died at a higher rate. What do I do if I don't have an outcome to look at, only characteristics? For context, I'm creating cognitive style groups based on user data for an insurance company, and these styles will be later used for morphing, churn etc. but do not have an outcome variable per se.
Bi Plot will suffice
40:00 why is mean 0 and standard deviation 1?
To show new mean is zero: mean = (sum of vals)/n, vals is then transformed (by code) such that centered_normalized_vals = new_vals = (vals - mean)/sd, so new_mean = (sum of (new_vals - mean))/n = (sum of vals - n*mean) / (sd*n) = (n*mean - n*mean) / (sd*n) = 0. To show sd becomes 1: originally sd is calculated as root ((1/n)*sum over every val of ((val - mean)^2)), vals is transformed (by code) such that centered_normalized_vals = new_vals = (vals - mean)/sd, so new_sd = root ((1/n)*sum over every new val of ((new_val - new_mean)^2)), new_mean was previously shown to be zero, so this means new_sd = root ((1/n)*sum over every new val of (((val - mean)/sd)^2)) = root ((1/n)*(1/sd^2)*sum over every val of ((val - mean)^2)), since (by our initial definition) sum over every val of ((val - mean)^2) = n*sd^2, this tells us that new_sd = root ((1/n)*(1/sd^2)*n*(sd^2)) = new_sd = 1.
Thanks for an amazing lecture! @29:35 it tries to cluster data into two groups and see if it correctly differentiated people who dies of heart attack and those that didn't. To me this is using clustering for classification task, if yes, when would someone use clustering rather than classification?
Usually you will use clustering as an unsupervised project wherein you do not have a label. Clustering can be used a first line of segmentation of your dataset that lacks the outcome label for which a classification model instead utilizes.
This professor is awesome!
love that prof for 4:35 - that is brilliant
Main issues of K-Means : choosing the number of clusters (k) and data scaling: But what if one wants to apply weights to the features (parameters)? Should you just multiply the features with the desired coefficients?
at 46:50 the professor mentions “has pretty good specificity, or positive predictive value, but its sensitivity is lousy.” can someone explain how specificity = ppv?
im assuming:
ppv = tp/(tp+fp)
specificity = tn/(tn+fp)
doesnt ppv = precision?
No ppv means positive predictive value.
Ur formulas are crct
thank you mit! from singapore . lots of love
14:20 the distance from Denver to Seattle is 1307 and the distance from Denver to Boston is 1949, so why he clustered Denver to Seattle instead of Boston when using Complete linkage? should it not be clustered to the greatest distance?
No, for complete linkage you look up, as you said the greatest distance of each cluster to the datapoint and then cluster it with the smallest out of these distances.
I wanna attend Professor Guttag's classes mostly for the education but also for the candies.
Professor Guttag: 'Dendrogram... I should write that down.'
also Professor Guttag: mispells it :D
Very good class! Thanks for the video and for the knowledge!
great professor
Pure gold... thank you so much.
I wish I had this professor, would probably love algorithms
Thank You MIT.
Why we use clustering while we have the label? Like in the medical example, we already know the label (0,1).
Ok, by minute 7 my mind is wondering if there's going to be a bonus assignment to find the probability that Professor Guttag will correctly throw you the piece of candy on the first try. The odds of you catching it greatly increase the closer your sit to the front center of the room.
i like how Dr. Guttag just throws candy at the students
From where can I get the pdf of the same. OR some notes.
The course materials are available for free on MIT OpenCourseWare at: ocw.mit.edu/6-0002F16. Best wishes on your studies!
What does he throw to the students who answers ?
Hello, I come from wet lab and I am not familiar with machine learning. But I am really interested in this topic since I want to apply machine learning to my research in plant genetics. I have watched this video several times but still I have not gotten all the things the professor mentioned. I wonder if the author or anyone can share the lecuter or books in this topic. It will mean alot to me. Thank you in advance.
A late response but here we go. I would suggest you read the 100 pages machine learning book, it doesn't really really assume any background but of course having it makes things easier.
What was the thing that John Guttag threw at the student
Is the full code of his examples accessible?
great lecture! at the speed where it is easy to understand
what is the average of examples in the same cluster?
The cluster centroid.
Thats so fucking cool. Explaining how to group data and throwing candy at your students for answering right
it's dendRogram, with an R. Comes from the word for "tree"
What’s the name of the course? And in what college ?
As the video description states, the course name is "Introduction to Computational Thinking and Data Science" as it was taught in the Fall of 2016 by the Massachusetts Institute of Technology. For more information, see the course on MIT OpenCourseWare at: ocw.mit.edu/6-0002F16.
To quote Dr. Banner: ‘Basic cluster recognition’...
Does someone know the name of the book 📚 used and where to access the code he mentioned he distributed?
Mit Open Course Ware website. Just search it by the name of the course
thank you Prof!
Great lecture! Informative AND entertaining.
MIT: 2 kinds of people.
Harvard: .........
Princeton: .........
actually 3. People like you trying to make differences at any price.
I guess that the statment that he was trying to set as True to scale the data was at line 14. Awesome lecture! Thanks.
how do we test different k values when examples are unlabeled?
Hierarchical clustering. Just stop when you like what you see?
"Clustering" is usually taught to "signal" "alumni" that anyone "in their *network*" can't learn and be good at some skills because some Terrorists in their "*network*" may be affected andor effected.
Can anyone link machine learning to digital signal processing for me?
what is the reference book ?
The textbook is: Guttag, John. Introduction to Computation and Programming Using Python: With Application to Understanding Data. 2nd ed. MIT Press, 2016. ISBN: 9780262529624. See the Readings section for more details: ocw.mit.edu/6-0002F16. Best wishes on your studies!
Thanks for the lesson professor, it's really good explanation
Thanks MIT
Data scientists actually have to think. Good one
Amazing!
That's one amazing lecture !
awesome class. I craved candy while watching it
I am feeling stress like in a class with a bunch of genius.
This is way more comfortable when at 1.25 speed.
I answered the questions give me my candies 🤣🤣
Each one choose for itself...
23:20
great lecture but the cholate did not reach me.
Prof john Guttag has banch of Candie's
1:45 Democrat/Republican... Smart/Dumb... Professor, you're being redundant!
Hahahaha
so we are getting candies for every right answer, i am 26 years old and heck yeah!! i would still love to have free candies 👍😜
He is treating you like pets. Like little hamsters.
I want that choco
This man has the mannerisms of Bill Gates
Normal Playback = 1.5x speed
if you have adhd yes
@@beepbeep767
gaps.....pauses.....deliberations = reduced
sloooooow intonaaaaation = reduced
Love how he subtly alluded to Republicans as dumb (and they are).
Take care, students, with democrat teachers in computer science classes. They don't care to play with you and call you a dumb if you are a republican and later ask you to choose who is the dumb and who is the smart. I hope you grades doesn't be influencied by you political bias.