An AI tool in linguistic analysis leveraging natural language processing (NLP) techniques and machine learning algorithms to understand human language in preparing, during and after risk situations. Particularly interest is in developing linguistic analysis in categories like sentiment analysis to determine the sentiment or emotional tone expressed with in it; Named Entity Recognition to identify and categorize entities, and/or part-of-Speech Tagging by recognizing grammatical categories of nouns, verbs or adjectives related to the risk situation.  

We will be using Neuro Linguistic Programming models such as LDA for unsupervised topic modelling. We also intend to use LLM (Large Language Model) for sentiment analysis. In the end, we will shed policy implications towards a more crisis-resilient world. 

Using linguistic recognition of words in whatsapp messages, call centers, radio and TV news reports, the linguistic analysis tool could detect information to better facilitate urban planning after a catastrophic event. It should detect word related to urgent needs and complaints tracked by citizens and volunteers, identify words that refer to the level of service of the disaster relief staff, word to detect the functionality of systems and facilities, word to access to power and critical infrastructure, word to assess business impact analysis. 

With AI tool, planners can acces real time data of sites that have been hit a disaster and can obtain data clsiffied by: 

regulatory staff support, website, or remote office 

jurisdiction guidelines—i.e., emergency adjuster licensing rules—which can be shared with call center staff and onsite DRC volunteers 

Share jurisdiction-issued bulletins and how we are to handle them