1. A few examples of biometric authentication are mouse movement authentication, keystroke authentication, handwritten password authentication, and even palm print authentication. Moving away from sensor-based biometric authentication makes it available to numerous different uses that previously required a specific sensor. Not only will this allow for more accessible biometric authentication, but it will keep the system and devices more secure as these types of biometrics are much harder to impersonate. RNNs can also open the environments in which authentication is performed.

2. Facial recognition ranges from identifying one’s identity to deciphering their emotions. Expression recognition often relies on a CNN for extraction of important features from image data before that image data can be used by the RNN [6]. Once these features are deciphered the LSTM RNN can make a prediction about the emotion perceived

3. One popular implementation of RNNs is applied to the domain of anomaly detection. Anomaly detection can range from detecting spam emails, to malicious network traffic and maritime vessel traffic

4. One such application of anomaly detection can be applied to Internet of Things (IoT) devices. An example of Anomaly Detection in IoT devices can be seen in where researchers develop an Intrusion Detection System (IDS) for IoT devices. An IDS using a RNN would rely on detecting anomalous patterns in the data to alert a user if there was anyone trying to hack into their IoT devices.

These are the four main topics that this paper will be reviewing. The goal of this paper is to analyze novel approaches in each of the four applications of RNNs.

Fernandez-Lopez, Pablo, Judith Liu-Jimenez, Kiyoshi Kiyokawa, Yang Wu, and Raul Sanchez-Reillo. 2019. "Recurrent Neural Network for Inertial Gait User Recognition in Smartphones" Sensors 19, no. 18: 4054. https://doi.org/10.3390/s19184054

2. Mouse and Keyboard Authentication Methods
S. Fu, D. Qin, D. Qiao and G. T. Amariucai, "RUMBA-Mouse: Rapid User Mouse-Behavior Authentication Using a CNN-RNN Approach," 2020 IEEE Conference on Communications and Network Security (CNS), Avignon, France, 2020, pp. 1-9, doi: 10.1109/CNS48642.2020.9162287.

3. Handwritten Authentication Methods
R. Tolosana, R. Vera-Rodriguez, J. Fierrez and J. Ortega-Garcia, "Biometric Signature Verification Using Recurrent Neural Networks," 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Kyoto, Japan, 2017, pp. 652-657, doi: 10.1109/ICDAR.2017.112

4. Model for Facial Expression Recognition Using LSTM RNN
Rajan, S., Chenniappan, P., Devaraj, S. and Madian, N. (2020), Novel deep learning model for facial expression recognition based on maximum boosted CNN and LSTM. IET Image Processing, 14: 1373-1381. https://doi.org/10.1049/iet-ipr.2019.1188

5. Multimodal Expression Recognition Implementing an RNN Approach
Shizhe Chen and Qin Jin. 2015. Multi-modal Dimensional Emotion Recognition using Recurrent Neural Networks. In Proceedings of the 5th International Workshop on Audio/Visual Emotion Challenge (AVEC '15). Association for Computing Machinery, New York, NY, USA, 49–56. https://doi.org/10.1145/2808196.2811638