This respository contains the models created to predict social unrest from ACLED data.
In this project, the goal was to develop a system for monitoring and predicting social unrest using natural language processing (NLP) techniques and transformer-based neural network models. The system was designed to analyze ACLED data in real-time and identify patterns and trends that could indicate potential unrest or conflict.
To achieve this goal, the team implemented a number of NLP techniques, including data preprocessing, feature extraction, and model training. The models used were transformer-based, which allowed the system to effectively process and analyze long sequences of data and identify subtle patterns and trends.
The results of the project demonstrated that the system was able to accurately predict social unrest with a high degree of accuracy. The team also identified a number of key factors that were important for predicting unrest, including sentiment analysis, topic modeling, and the use of external data sources such as news articles and government reports.
Overall, the project demonstrated the potential of NLP and transformer-based models for monitoring and predicting social unrest, and the results of the project could have significant implications for governments and organizations looking to proactively address potential conflicts and unrest.