Abstract:
A natural disaster is a natural event which can cause damage to both lives and properties.
Since SM are capable of real-time nature of sharing the information, post-disaster
management can be improved to a great extent if we mine them properly. After identifying
this need and the possibility of solving them through SM, as a first step, we fetch the twitter
posts by using predefined keywords relating to the disaster from Twitter API. Those posts
were cleaned and the noise was reduced at the second stage. We followed keywords
filtering, date filtering, two-step filtering, and geo-location filtering technique for reducing
the noise. Named Entity Recognizer library was used for getting the geolocation. We did
the above two stages for news which were fetched from News API. As a final stage, we
compared the twitter with news datum to give the rating for the trueness of each Twitter
post. The rating “more accurate” was given to Twitter posts which satisfy all the three
parameters; disaster type, location details, and date with news. “Moderately accurate” rating
was given in which Twitter posts satisfy disaster type and one of other parameters. The
rating “less accurate” was given to Twitter posts which ta disaster type. “No correlation” is
the rating in which Twitter posts do not satisfy any of the parameters. We took the news as
a standard to rate Twitter posts. We believe that by using our model we can alert the
organizations to do their disaster management activities in a timely manner.
