2.4 DS: Jaccard Coefficient or Index or Similarity
HTML-код
- Опубликовано: 11 сен 2024
- #JaccardCoefficient #Jaccard #similarity #MachineLearning #DataScience #DataMining #ComputingForAll
🔗 Here is my easy to grasp Data Science/ML course RUclips playlist:
• Machine Learning/ Data...
🔗 Here is my easy to grasp ML workshop series:
• Data Science/Machine L...
The high similarity between a pair of points indicates that the points are nearby. Low similarity indicates a large distance. Literature covers several similarity measures. In this video we talk about Jaccard coefficient or index or similarity.
Jaccard index/coefficient/similarity is generally computed between two sets of items. It is a ratio of commonality between the sets over all the items. If X and Y are two sets, then the Jaccard index between two sets is computed using the ratio of the size of the intersection and the size of the union of the two sets.
Jaccard index can be computed between two vectors too. Jaccard index computed between two vectors/data points/objects is called a weighted Jaccard index.
Here is a video on the Python code for computing Jaccard Similarity:
• Weighted Jaccard Simil...
For examples and more similarity measures visit the page below:
computing4all....
Visit the page with all the data science materials we have developed: computing4all....
Thank you.
Dr. Shahriar Hossain
computing4all.com
#JaccardSimilarity #JaccardIndex
Thank you for making it so simple and straight-forward (but also covering the set-based approach first for informational purposes)! Much appreciated!
I am glad to hear that you liked it. Thank you for watching and providing the feedback.
Thanks. Simple and easy to understand.
I am glad to hear that you liked the video. Have a wonderful day.
GOD BLESS you made it soooooo simple THANK YOU KEEP POSTING
Thank you for watching and the kind words. I am glad to hear that the video was helpful.
Teacher , your explanations are just amazing, thank you so much
I am glad to hear that you liked it. Thank you for watching.
Thank you for this lovely explanation, you have saved me on my project. i really needed the generalization of the jaccard index and you did justice to it sir.
I am glad to know that the video was useful for you. Thank you for watching and for the kind words.
Thank you for explaining the material so clearly.
I appreciate your feedback. Thank you for watching.
Thank you, Dr. Hossain your explanation is very simple and excellent.
Thank you for watching and commenting. I am glad to hear that the video was useful. Have a wonderful time.
an amazing, simple, and easy to understand explanation. Keep it up.
I am happy to hear that you liked the video. Thank you for watching and providing the feedback. Have a wonderful time.
This was awesome I real appreciate mr
I am glad to hear that you liked the video! Thank you so much for watching!
Thank you for clear explanation
I am glad to hear that you liked it. Thank you for watching.
Amazing tutorial. Thank you! =)
I am glad to hear that you liked it. Thank you for watching and the kind words.
Masha Allah, great explaination
Thank you for watching and the kind words.
The explanation is simple and clear. Thanks for sharing!
I am glad to hear that you liked it. Thank you for watching and commenting!
Nice intuitive explanation, thanks!
I am glad to hear that you liked it. Thank you for watching and commenting. Have a wonderful weekend ahead.
really really impressive sir.. a pretty simple explanation. I have subscribed channel in the first video.
I am glad to hear that you liked the video. Thank you for watching, commenting, and subscribing.
Thank you Dr. Hossain for this excellent video.
I am glad to know that you liked it. Thank you for watching and commenting. Have a wonderful weekend ahead.
Very good explanation !! Really helpful :) Thanks
I am glad to hear that the video was helpful. Thank you for watching and commenting.
Great explanation!
I am glad to know that you liked it. Thanks for watching and commenting.
Very nice explanation sir, you should continue posting more videos 👍🏻
Thank you for watching and the wonderful comment! I hope to produce more videos in the coming days,
@@C4A Thankyou sir, today i had my machine learning exam and your videos lecture helped me for some topics.
You are most welcome and it is great to hear that the videos were helpful! Thank you.
@@C4A sir, can you explain me K-means code on iris dataset? Its for my assignment
I have the following two videos on k-means.
ruclips.net/video/DtsYfwl4uKw/видео.html
ruclips.net/video/qSy0nXYrVRM/видео.html
You can tailor the codes discussed in the videos fir iris data.
You are great teacher! Thank you!!!
Thank you for the kind words. Have a wonderful weekend.
Thank you for your video
You are most welcome and thank you for watching.
Wow. This is a nice explanation. Thank you
Thank you for watching and commenting.
Thanks. simple explanation
I am glad to know that you liked it. Thank you for watching and commenting. Have a wonderful time.
Excellent explanation. Thanks a lot
Glad you liked it! Thanks for watching.
Thank you so much! Your video helped me a lot!
I am glad to know that the video was helpful. Thank you for watching and sending me the note. Have a wonderful weekend ahead.
This was an awesome explanation, thank you very much.
I am glad to hear that you liked it. Thank you!
Thank you sir, very simple regarding explanantion, helps lot
I am glad to hear that you liked it. Thank you for watching and commenting. Have a wonderful day.
thank you for making it so interesting!
Glad to hear that you liked the video. Thank you for watching and commenting.
Good and easy explanation, Thank you
Thank you for your comment. Best regards.
Thanks alot, This is very clear explanation
I am glad to hear that you liked it. Thank you for watching and providing the feedback. Have a wonderful week.
Thanks sir wonderful explanation.
How to classify jackard coefficient like excellent,very good,good,poor,very poor ?
I am glad that you liked the video. Thank you for your question. It is hard to classify a similarity measure in a general context. For some datasets one similarity measure might be better than the other. Jaccard is a widely used similarity measure when the algorithm using it does not require the triangle inequality of distance to be satisfied. For measuring document-document resemblance, generally cosine similarity is used. For general commodity/count-based features in a dataset, Jaccard is an intuitive measure for computing similarity between two objects/rows of a data table.
Thanks again for this great question.
@@C4A wow
Thanks for prompt response 😍🤲🏼
Amazing..thank you Sir! It helps a lot
I am glad to hear that the video is helpful. Thank you for watching and commenting.
Thank you so much ❤️❤️
You are most welcome! Thank you for watching and commenting.
3:23 jaccard(x,y) = 0/6?
Well explained. Thanks
Kofi Amador I am happy to hear that you liked the video. Have a wonderful week ahead.
Thank you for this nice video. Pleasing music too.
Haha 😆 I am glad that you liked the video and the music. Thank you for watching. Have a wonderful weekend.
Thank you sir.
You are most welcome! Thank you for watching.
Crisp and clear
sai bhargav lanka I am glad to hear that you liked the video. Have a wonderful weekend ahead.
awesome explanation sir
CSE ROCKER I am glad to hear that you liked it. Thank you for watching. Have a wonderful week.
so far so good thank you do much
I am glad to hear that you liked it. Thank you for visiting and commenting.
Very nice video, thank you sir.
Garry Long Thank you for watching.
Sir you may make a mistake. 076 is greater than 0.323076923 that means row1 is very similary to row5 than row3.
Higher Jaccard similarity means "more similar". Since Row 1 and Row 3 have a Jaccard similarity of 0.76 (higher), but Row 1 and Row 5 have a Jaccard similarity 0.32 (lower), Row 1 is more similar to Row 3 (compared to Row 5). It seems I said it correctly but I wrote it incorrectly on the screen in the top-left corner at 8:02 timestamp. The smaller than symbol "" in timestamp 08:02 . Thank you for your comment.
@@mahmudshahriarhossain7155 Thanks sir for your kind and quick response.
@@shahariarsarkar8646 I have now blurred the text in the top left side where I have the "
Hallo, thanks for the video, but I would like to know how to write a program to calculate this?
Hello, sure. I have provided Python code for Jaccard, Cosine, and Tanimoto similarities here: computing4all.com/courses/introductory-data-science/lessons/similarity-measures/
Thank you for watching the video! Have a wonderful time.
awesome job, thanks!
I am glad to know that you liked it. Have a wonderful weekend ahead.
You look like ElectroBOOMs brother :P Nice video!
😂🤣😅 Thank you for watching! Have a wonderful week.
Merci
You are welcome! 😁
Sir in vector based jaccard similarity if we have two same values in same attributes (features) then which value will be taking as in minimum or maximum. because we have two same values.
Thank you for your question. Either one is fine as the minimum or maximum if two values are the same. Fir example, min(5, 5)=5 and max(5, 5)=5. That is minimum and the maximum will be the same number.
@@C4A i need your email
The email address is: allforcomputing at the rate gmail dot com
sir, can we use this index for multiple samples? or it is only for comparison between two samples?
Similarity or distance is generally computed between a pair of vectors (samples). If you need to compute average similarity, then you can compute similarity between all pairs, sum the similarities, and divide by number of pairs.
If you need to compute similarity between two groups of samples, you can construct a pair for which one sample comes from group1 and the other sample comes from group2, and then compute the similarity for that pair. Compute the average of all similarities computed for all such pairs. You can call this average "the average similarity between group1 and group2".
I hope that answers your question. Please feel free to ask questions you may have. Have a wonderful week.
Thanks a lot
Jose Romero You are welcome. Have a wonderful week.
Great explanation!
Thank you for watching and commenting! Have a wonderful week ahead.