Disaster detection technique based on the key word relevance and the method for managing the disaster using the same

摘要

The present invention relates to a disaster detection technique for each disaster issue based on key word relevance and to a method for managing a disaster by using the same. The disaster detection technique for each disaster issue based on key word relevance according to the present invention comprises the steps of: evaluating how relevant each key word is for a particular type of disaster at a key work level; then obtaining the relevance value of a message by summing the relevance of key words included in each message at a message level; and finally summing the relevance values of relevant disaster messages that have occurred for a particular type of disaster for one hour, and detecting a disaster for each disaster issue if the sum is higher than a critical value made with respect to past data. The disaster detection technique for each disaster issue based on key word relevance and the method for managing a disaster by using the same according to the present invention, as configured above, enable a user to effectively detect the precursor of a disaster for each disaster issue by constructing a probability model based on quantified relevance instead of the number of messages as a relevance evaluation technique depending on the content of the message is introduced to solve the problem of disaster issue detection using a conventional probability model based on the number of messages per hour.