Question Answering techniques for detecting needs of people in crisis

The use of social networks has become one of the basics of daily life to follow the news, and also in the case of inquiries and requesting assistance. In case of a crisis, information about the event begins to spread on social networks with the aim of raising awareness and knowing all the important instructions and requesting assistance. This information can be used to respond to people with the required needs depending on the type of assistance or inquiry requested. In this paper, an approach was proposed that uses Question Answering techniques based on natural language techniques and neural networks to extract the needs of affected people during a crisis and provide the proper response. The approach was tested on twitter data from different types of crises. The results showed that the suggested approach can answer with suitable guidelines with a precision of 0.81, a recall of 0.76 and an f-score of 0.78.