SecuritAI UI

SecuritAI UI application provides a platform to monitor for malicious requests from a stream of request logs. The prediction model is powered by Recurrent Neural Network developed with Keras as explained in Detecting Malicious Requests Using Keras & Tensorflow.

Set-up

The following sections are related to setting up the software.

Requirements

This software has the following requirements:

• Node 6.11.0+
• AWS Kinesis Permissions

AWS Kinesis Permissions

You will have to setup your local machine to be able to authenticate against your AWS account. More details/instructions can be found in the AWS documentation for CLI configuration.

Running the software

You can run the software with the following steps:

1. clone this repository
2. activate your AWS account credentials environment variables
3. cd into the project folder
4. run npm run start

Implementation details

This software is built with the following components:

• Keras (LSTM RNN)
• React
• Kinesis

Keras

An LSTM RNN model is provided under assets/ for keras-js to load up and perform predictions. This can be swapped out for any applicable model of choice

React UI

React & Redux

Kinesis Stream

Data is consumed from an AWS Kinesis stream to be processed by the RNN. A sequence number should be maintained to keep track of records consumed in the stream.

Development

The following sections are primarily related to setting up a development environment.

Start dev and watch

When you are ready to start development, you can run:

npm run start:dev