Inspiration
The idea was inspired by prediction techniques in NLP. Since we wanted a model that will have some memory, the LSTM cells were the obvious choice.
What it does
The model learns to predict the logs of a given computer. Usually it does only small errors, however the hard ones are a great tool to detect anomalies in the dataset.
How we built it
We built it using Python and TensorFlow
Accomplishments that we're proud of
The model is very successful in detecting the anomalies. It pointed out to all the hints needed to solve the case.
Built With
- python
- tensorflow
Log in or sign up for Devpost to join the conversation.