1
$\begingroup$

I want to implement ML to monitor log file, classify them as normal and abnormal. The model must learn via this process of classification and after time be able to classify the log files itself. This classification of the log files is based on time of login and place of login. Any hint / advice of ready made model for the scenario would be of great help.

$\endgroup$
1
$\begingroup$

If you want to use only time of login and place of login, there are several models that would do the work. Take a look at Python's Sklearn. It offer lots of different options. An example one is random forest: https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html

Normally you don't event need to understand the model under the hood to be able to use it, so no need to feel intimidated by the names

$\endgroup$
  • $\begingroup$ Thank you Pedro, I real appreciate. I just want to increase security of my normal systems by introducing this feature of AI to the systems. First I thought of only time and place of login to the system. What other features I can add to my system to make it more secure? I am new to this AI world but I am committed and ready to make it happens. Please can you guide me? I am from tanzania and I don't have direct tuitor for this $\endgroup$ – Noel Nov 22 '18 at 5:55
  • $\begingroup$ Hi Noel, I haven't dealt much with applications related with security and AI. So I am not sure what the best practices are. But maybe you could look into sites like this: auth0.com/learn/anomaly-detection $\endgroup$ – Pedro Torres Nov 22 '18 at 12:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.