Sir Thanks for the video.. Actually I'm taking help form your video to make the project on Phishing attack detection from machine learning technique. My doubt is I'm unable to show the Phishing attack testing of the URL,, What to do and sir how to do,, Reply needed!!!
Hi Mihir, This video covers domain names only. If you're looking for something a little more like a WAF then you'll need to perform anomaly detection on user input to identify malicious characteristics.
This model was created using Scikit-learn library in Python. As far as I know, you'll require the same library to execute on whatever application you're running.
Sure. "test_tfidf_X" is created in Task 3 at timestamp 7:34. If you're getting that error that means the code block was not executed correctly and there were errors. The notebook will tell you the error that occurred on that code block.
Hello Great RUclipsr , I saw your impressive video. I searched on github , on trying to understand working of machine learning.Here it is what I learned : I saw and read that Random Classifier Algorithm is the best for Phishing Detection. A data-set is used from archive.ics.uci.edu/ml/datasets/phishing+websites website. This data-set is parsed in Random Forest Classifier Algorithm. What I found is, that other features are implemented (30 in number or total). My Query is, What is the connection between the Random Forest Classifier Algorithm and these 30 Features? Because based on my understanding if I ran the features python file, And get output I should still be able to get a right output. Please put light on this problem of mine. Best Regards, Deepanshu Singh.
thanks Netsec! this video was very informative and I hope to learn more from your channel.
listening to Ken Carson's Teen X - EP right now
Well Explained Precision and Recall... Great tutorial overall
Thank you for your informative video, very useful!
Subscribed 👌✔
Nice explanations 👍
Thank you for this!
Thanks for the video =)
Thanks!
May I know which protocol you used in this regarding computer network subject
Sir
Thanks for the video..
Actually I'm taking help form your video to make the project on Phishing attack detection from machine learning technique.
My doubt is I'm unable to show the Phishing attack testing of the URL,,
What to do and sir how to do,,
Reply needed!!!
Can I apply same tokenisation algorithm for malicious xss payloads?
Yes you can! In fact, tokenization is a standard process in AI/ML tasks like this.
@@NetsecExplained great thanks!
how to try a url with custom user input
Hi Mihir,
This video covers domain names only. If you're looking for something a little more like a WAF then you'll need to perform anomaly detection on user input to identify malicious characteristics.
hello, is there any way to apply this machine learning model to a windows server?
Absolutely! The easiest way is to download and install Anaconda. It will come with Jupyter notebooks so you can clone this repo and work from there.
How can I get dataset for phishing and legitimate url with features
Hello, all datasets and workbooks are located in the GitHub linked in the description below the video.
How to deploy this model in a website or an android app ?
This model was created using Scikit-learn library in Python. As far as I know, you'll require the same library to execute on whatever application you're running.
Can this is done in R programming
It can be done in R. The algorithms are the same, I'm only demonstrating how this is done in Python.
Hello sir in task 4a I get error like "test_tfidf_X" is not defined so please help me I properly followed all steps
Sure. "test_tfidf_X" is created in Task 3 at timestamp 7:34. If you're getting that error that means the code block was not executed correctly and there were errors. The notebook will tell you the error that occurred on that code block.
@@NetsecExplained ok sir I check for it thank you
Hello Great RUclipsr ,
I saw your impressive video.
I searched on github , on trying to understand working of machine learning.Here it is what I learned :
I saw and read that Random Classifier Algorithm is the best for Phishing Detection.
A data-set is used from archive.ics.uci.edu/ml/datasets/phishing+websites website.
This data-set is parsed in Random Forest Classifier Algorithm.
What I found is, that other features are implemented (30 in number or total).
My Query is, What is the connection between the Random Forest Classifier Algorithm and these 30 Features?
Because based on my understanding if I ran the features python file, And get output I should still be able to get a right output.
Please put light on this problem of mine.
Best Regards,
Deepanshu Singh.
Why does the class imbalance not matter here????