I have an exam tomorrow and I was desperately searching for a video that matches the way our Professor taught us. It's 4.42am now, and after searching and watching several videos, finally, I got this. Now, I can go to sleep. Thank you. :D
What you covered within 00:15:17 mins long video has more weight in terms of practicalilty and knowledge than 1:02:00 hour long university video.. Truly thank you
I have an exam tomorrow and was frantic to find a video that matched the manner our Professor taught us. It's 4.42 a.m. now, and after much searching and viewing of movies, I've finally found this. I can finally sleep now. Thank you very much.
I had a one hour lecture. Our lecturer just kept on reading the presentation slides. And here, bam! I spend 15 minutes trying to understand. I understood! Thank you, ma’am.
You did an amazing job explaining this visually... I think most people get scared with Mathematical concepts and Algorithms is because they are visual, and if they can't see how it works practically and visually, they loose interest.. I wish I was taught this way in my school days. Thank you.
Great explanation! Thank you. I looked at several tutorials before this and all of them were incomprehensible. I am not a software engineer but I do understand technology and basic math so this was perfect.
This is by far the best explanation for K-Mean clustering i've seen online. I will like to see a video teaching on dimensionality reduction if there is any, please.
I have seen many videos for K means clustering as it is required for my project on Cognitive radio but none were useful. But yours video is highly helpful
Very concise and to the point. May be the best explanation for this topic. Only number and font in slides that wish it more readable. But the lecture quality is master piece !
best explanation i ever got for Kmeans- thanks prof except correction: at 11.59 where cluster2 has 3 points not 4 and hence centroid is ((8+7+6)/3,(4+5+4)/3)=(7,4.33) which is the same as you got using 4 points
Im strugling on our data mining subject since theres a pandemic and i cant attend online classes and can only rely on modules... Modules cant really explain everything so im thankful that you explained this clearly
Amazing. Thank you so much mam. Now i got to know how to solve k-means clustering problem. I really appreciate you. Thank you so much for sharing this video within us. Thankx alot mam.
very great explained...But mam can u tell me if there are multiple feature suppose 4 feature then how it work....for calculation how to take x and y....plz tell me
Thank u for the video, it is clearly explained, however there is one question. If we set for example 3 clusters at the whole beginning, could it happen so that after 'n' itterations we will have let's say only 2 clusters in the output? Because while we calculate the means and assigning clusters it might happen based on minimal value of means some clusters could be out, isn't it?
It will not happen if your initial centroids are chosen from the dataset. Because you will have at least one member (the centroid) for each cluster. however, if you start with a random start for your centroid, it is possible that one of the clusters is chosen way outside the range of data and no observation belongs to it. that's why there are methods developed to start with good starting centroids. because it also impacts the number of iterations.
Hi , This is the best K Means clustering explanation with perfect example in YT. Could you please do a video on math behind Gradient Boosting algorithm? Thanks from KM
Hi Nitin Jain, I just solved using the Euclidean distance. Both are giving the same cluster output. I think for bigger input points we can use Euclidean Distance Formula, for easy solutions.
@@sabarivishnujayanthanjaikr4653 Thank you for solving completely using Eulidean distance. For this example, the result might have been the same; but let's consider points an instance a) (10,10) and (5,5) b) (10, 10) and(1,10) . Here according to the solution provided, (5,5) would be farther to (10,10) than (1,10) but euclidean distance would clearly show that point (1,10) is farther. Just wanted to know if this is from any prescribed material or if this usage is recommended.
This s the best and simple explanation for K-means clustering. Thank you a lot Shokoufeh Mirzaei. can u plz explain abt elbow method to determine unknown K values for clustering?
WHAT IF WE ONLY HAVE TWO POINTS OR TWO (A)POINTS, I MEAN A1 and A2 AND ONLY HAVE 2 CLUSTERS OR "K" AND 2 "Centroids" BECAUSE "K==Centroids". IS IT STILL THE ALGORITHM WORKS FINE EVEN IF WE ONLY HAVE 2 POINTS? BECAUSE CENTROIDS IS RANDOMLY PICK AS LONG AS THE NUMBER OF (A)POINTS IS GREATER THAN OR EQUAL TO THE NUMBER OF CLUSTERS? THANK YOU FOR IMMEDIATE RESPONSE. WHAT A GREAT TUTORIAL, WE LOVE IT.
ONE MORE THING, SINCE IN THE FIRST RUN OF PICKING WHICH (A)POINTS SHOULD BE PICK AS CENTROIDS. IS IT POSSIBLE THAT WE CAN HAVE A SAME (A)POINTS FOR A CENTROIDS? EX. NOTE: 4 (A)POINTS AND 2 "K" AS CLUSTER NUMBER AND CENTROID IS 2. => WE HAVE 4 (A)POINTS, THEN {A1, A2, A3, A4} RANDOMIZED PICKING OF CENTROID ARE {A1, A1} .
I have an exam tomorrow and I was desperately searching for a video that matches the way our Professor taught us. It's 4.42am now, and after searching and watching several videos, finally, I got this. Now, I can go to sleep. Thank you. :D
me 2 cool..... maam make more vidios
Guess what I get this in first click... Alhamdulillah :)
Man you are telling my current situation)))))))
Its 4.58 am for me.
5.32 am
The best way of K-mean Clustering explaination! Iranian Professor is amazingly good at teaching!
What you covered within 00:15:17 mins long video has more weight in terms of practicalilty and knowledge than 1:02:00 hour long university video.. Truly thank you
I have an exam tomorrow and was frantic to find a video that matched the manner our Professor taught us. It's 4.42 a.m. now, and after much searching and viewing of movies, I've finally found this. I can finally sleep now. Thank you very much.
You want me to believe that you have made me understand k-mean clustering using just basic +,-????? Well, i believe you!! Thanks a lot!!
you saved my life. i missed a class and i am preparing for an exam tomorrow. This video is very useful.
درود خدمت تان بانو شگوفه غزیز و گرامیthank you soo much dear Ms.shukofeh
I am sure this is the video everybody is looking for. Underrated
I had a one hour lecture. Our lecturer just kept on reading the presentation slides. And here, bam! I spend 15 minutes trying to understand. I understood! Thank you, ma’am.
The BEST Explanation of K-Means Clustering Algorithm on RUclips
You did an amazing job explaining this visually... I think most people get scared with Mathematical concepts and Algorithms is because they are visual, and if they can't see how it works practically and visually, they loose interest.. I wish I was taught this way in my school days. Thank you.
This is the only video available explaining the problem with Manhattan/Rectilinear distance. Thanks a lot ma'am. Love from India. :)
Excellent and easy explanation, thanks a billion Prof. Mirzaei.
Baji chaa gaey jey tusi.... You have done a fabulous job...
Great explanation! Thank you. I looked at several tutorials before this and all of them were incomprehensible. I am not a software engineer but I do understand technology and basic math so this was perfect.
Honestly, this is the only video that helped me to understand this topic.
Your explanation is so comprehensive and easy to understand. Thank you for your time and effort that you put into making this tutorial.
Thank you Ma'am! Such a beautiful way to explain everything. I am glad I didn't skipped your video because there were less likes.
This is by far the best explanation for K-Mean clustering i've seen online. I will like to see a video teaching on dimensionality reduction if there is any, please.
That was too much helpful. Thank you very much. To learn k-means, this tutorial is recommended.
Thanks for your help. I just finished my report, after this fantastic and easy to understand lecture. Have a good day.
Awesome video! I can't believe how you simplified everything for it to make this much sense. Kudos to you!
One of the best videos on K-mean clustering. Thanks a ton.
its atually the best on youtube. Other people explain just to sound smart and not consider whether the person watching is a novice
I have seen many videos for K means clustering as it is required for my project on Cognitive radio but none were useful. But yours video is highly helpful
great tutorial. systematic and crystal clear explanation. many thanks. this tutorial is highly recommended to watch.
honestly the best video i have come across to clearly explain this topic!
You are a wonderful Tutor, clear and concise, thank you for sharing your material.
This was extremelyyyy helpful. Thank you so much, your explanations were crystal clear
Very concise and to the point. May be the best explanation for this topic.
Only number and font in slides that wish it more readable. But the lecture quality is master piece !
Very well explained with a basic example.Easy to understand. Good job Mirzaei!!
This is the best video I could ever find at 5:21 AM, absolutely amazing :D
Really awesome video. I understood this in one time. You teach so better than my professor, wish you were my professor.
Beautifully explained! Most simple and useful
This was the best explanation for K-means clustering. Thanks a lot!!
Your explanation is much excellent like your dutifulness. Thanks
This is the best tutorial on k means clustering
Amazing video. Also you have a great accent for a non native English speaker. Thank you so much Professor!
Thank you for the video. No one could have done a better job
best explanation i ever got for Kmeans- thanks prof
except correction: at 11.59 where cluster2 has 3 points not 4 and hence centroid is ((8+7+6)/3,(4+5+4)/3)=(7,4.33) which is the same as you got using 4 points
Im strugling on our data mining subject since theres a pandemic and i cant attend online classes and can only rely on modules... Modules cant really explain everything so im thankful that you explained this clearly
Thank you Prof. Mirzaei, your lecture is amazing !
Thanks really for helping us!!Excellent work!!
ممنونم عزیزم.. عالی توضیح دادی.. دقیقا دنبال همین مثال از کتاب میگشتم
sure! no problem :)
The way you explained was really awesome loved it ❤️ best video for kmeans clustering
The math behind the algorithm best explained! Thanks a million for your explanation!
Great video, and very well explained as I watched a number of similar topic videos, and this one is by far the best
I understand this for the first time! Thank you so much!
Awesome video! Great explanation.
thanks Shokoufeh joon, very helpful
GOOD Explanaton. Thank you for providing nice video.
Best video ever...wao. Thank you and God bless.
Mam i cant tell you how much thankful i am. You saved my life. Love you
You are too good. You have saved me alot of trouble.
I am very grateful
Wow this method is so much easier. Thank you so much. Worth every second 😀
you saved my day. at last i can hand calculate the first step of centroid of clustering
Thank you so much.. Very easy and nice understanding skill 😊 very helpful vedio
Amazing Explanation of K-Means Clustering
this is the best video on this topic
Beautiful video, it just explains exactly what I was looking for. Hopefully I get this question on my exam!
Wow this wonderful, better than my teachers
Amazing. Thank you so much mam. Now i got to know how to solve k-means clustering problem. I really appreciate you. Thank you so much for sharing this video within us. Thankx alot mam.
Thank you for this video. God bless you
Thanks a lot. Your videos helped understanding it easily and made it less ambiguous as my professor made it seem.
very great explained...But mam can u tell me if there are multiple feature suppose 4 feature then how it work....for calculation how to take x and y....plz tell me
This is best video explanation for K-means clustering, thank you a lot! Like and subscribe
I agree!
I agree too
Amazing explanation😆😍
It was very intuitive. Really cleared my understanding.
Wonderful explanation!!!
Outstanding explanation
easily understand thank you shoko
I have been searching for this type of solution for 2 two days
Thank u for the video, it is clearly explained, however there is one question. If we set for example 3 clusters at the whole beginning, could it happen so that after 'n' itterations we will have let's say only 2 clusters in the output?
Because while we calculate the means and assigning clusters it might happen based on minimal value of means some clusters could be out, isn't it?
Same question here!
It will not happen if your initial centroids are chosen from the dataset. Because you will have at least one member (the centroid) for each cluster. however, if you start with a random start for your centroid, it is possible that one of the clusters is chosen way outside the range of data and no observation belongs to it. that's why there are methods developed to start with good starting centroids. because it also impacts the number of iterations.
Excellent explanation.
this is an excellent explanation
Very easily explained madam.
Obligatory "I am here the night before my exam trying to save my ass"
very, very, very good explanation. Thaks
great explanation mam!!! It was really helpful to me. Finally I solved my problem. Thanks!
Awesome.. I am glad that you helped my problem in understanding what is really happening :)
Hi ,
This is the best K Means clustering explanation with perfect example in YT. Could you please do a video on math behind Gradient Boosting algorithm? Thanks from KM
This is the best video. Thumbs up!
Many Thanks for a great video and clear explanation.
Thank you so much for this. Please, can you do a video on k-means cluster evaluation; Internal measure, hand calculated
Thank youuu !!! The only video that made me really understand it !
Excellent and simple explanation
Amazingly explained, thanks!
I can understand by you are this movie.
Thank you very much !! ;-)
Another correction as well,
Correctoin: at 11:53, In cluster 1: /3 instead of /2
Great explanation anyways
Thanks for solving an entire problem. One doubt here, shouldn't we consider the euclidean distance between the points?
Hi Nitin Jain,
I just solved using the Euclidean distance. Both are giving the same cluster output. I think for bigger input points we can use Euclidean Distance Formula, for easy solutions.
@@sabarivishnujayanthanjaikr4653 Thank you for solving completely using Eulidean distance. For this example, the result might have been the same; but let's consider points an instance a) (10,10) and (5,5) b) (10, 10) and(1,10) . Here according to the solution provided, (5,5) would be farther to (10,10) than (1,10) but euclidean distance would clearly show that point (1,10) is farther. Just wanted to know if this is from any prescribed material or if this usage is recommended.
@@nitinneo7 you have a point bro. Let us wait for someone else's comment..
Choosing the distance method depends on your data and is rather arbitrary. However, Euclidian distance is more commonly used.
This is very very help thanks a lot ... I was searching for this classifier like whole yesterday
Excellent 👍✌✌👌
This s the best and simple explanation for K-means clustering. Thank you a lot Shokoufeh Mirzaei. can u plz explain abt elbow method to determine unknown K values for clustering?
Very Clear, many thanks
I appreciate your help. Im also from iran
Simple and clear explain, thanks!
this saved my life thanks
Thank You For making this amazing tutorial. Much appreciate!
Nice explanation for each individual steps! thanks
WHAT IF WE ONLY HAVE TWO POINTS OR TWO (A)POINTS, I MEAN A1 and A2 AND ONLY HAVE 2 CLUSTERS OR "K" AND 2 "Centroids" BECAUSE "K==Centroids".
IS IT STILL THE ALGORITHM WORKS FINE EVEN IF WE ONLY HAVE 2 POINTS? BECAUSE CENTROIDS IS RANDOMLY PICK AS LONG AS THE NUMBER OF (A)POINTS IS GREATER THAN OR EQUAL TO THE NUMBER OF CLUSTERS?
THANK YOU FOR IMMEDIATE RESPONSE.
WHAT A GREAT TUTORIAL, WE LOVE IT.
ONE MORE THING, SINCE IN THE FIRST RUN OF PICKING WHICH (A)POINTS SHOULD BE PICK AS CENTROIDS.
IS IT POSSIBLE THAT WE CAN HAVE A SAME (A)POINTS FOR A CENTROIDS?
EX.
NOTE: 4 (A)POINTS AND 2 "K" AS CLUSTER NUMBER AND CENTROID IS 2.
=> WE HAVE 4 (A)POINTS, THEN {A1, A2, A3, A4} RANDOMIZED PICKING OF CENTROID ARE {A1, A1} .
12:17, shouldnt it be /3 for cluster 1?