Sentiment analysis as an indicator to evaluate gender disparity on sexual violence tweets in south Africa

South Africa has one of the highest rates of sexual violence in the world, and with this technology era being the information age, it is not suprising that people are turning to social media to voice their views on this matter. As tweets on sexual violence grow continually, extracting insights from such data demands a robust, real-time, and scalable tools with flexibility in the database schema. Elasticsearch (ES) is an example of a free-license search engine written in Java and developed on Apache Lucene that meets the stated requirements. On the other hand, Kibana facilitates intuitive dashboard development, visual exploration, and real-time analysis of an index in ES through an intuitive graphical user interface. This study demonstrates how gender was inferred and evaluated through the integration of deep neural networks and Google's TensorFlow. We used the AFINN model to infer sentiment analysis as our measure of gender disparity in this instance. The Indexer, built of Node. js, and defined as the hub of the system connects with the Twitter streaming API to ingest tweets found within the boundaries of sexual violence. This system runs persistently with tweets through the ES search engine and visualised in Kibana.