An efficient twitter data collection and analytics framework for effective disaster management

Abstract

The world has faced an exponential increase in disasters and it is very evident that losses incurred due to disasters and the risk of disasters are also increasing. The literature [1] states that India has faced 371 natural hazards resulting in considerable deaths, infrastructural, agricultural, and economic devastation. Social networking development has been expeditious and has been considered beneficial in disaster management. In culture, social media sites such as Twitter and Facebook have reached an irreplaceable stature that demonstrates the prominence of colossal data acquired. Social networking data can be gathered and used for disaster-specific data analysis, which can be used to extract disaster logistics, identify patterns, and provide public preparedness assistance. The proposed work establishes a data collection system that is capable of capturing disaster-related tweets such as landslides, rain and flood in focus, eliminating obsolete and irrelevant data, retaining related data such as media and location without breaching social media ethics privacy policies. Within the period from May 2020 to December 2021, the system was able to collect 9,21,897 tweets. The diversity of culture, language and expressions prevalent in different parts of the globe is reflected in the way knowledge is conveyed, which makes it relevant to collect tweets in Indian languages. So, for easier data analysis, this system gathers tweets that are in Indian languages and translates them into English for efficient data analysis.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 21 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.