How it works

Data is extracted from Social Media

Data is cleaned and processed

CV used for image classification

Infographics generated after analysis

Data Extraction

Scraping the bulk of data from affected people and eyewitnesses is a vital first step – text, timestamp, profile location, and media links. Implemented using the Twitter API.

Try Demo

Text Processing

Discarding irrelevant text based on keywords used in unrelated contexts, and categorizing the resources that need to be allocated.

See Demo

Image Processing

Deep Neural Networks are used to extract information from images. The Relevance, Type of Damage and the Severity is identified.

How it Works

Classified as Fire

Classified Severe

Classified Irrelevant

Visualisation

We present statistics as simple visualisations at a glance to summarise the extracted information and resource allocation needed.

Dashboard

Big Data Integration

Since we will be dealing with huge amounts of data, we use multiple Apache frameworks to collect and perform computations efficiently.

Future Prospects

Language

More research into Indian language processing, pre-trained models, and corpora collection will help us expand our project to a diverse range of localities.

More Data Sources

Higher level access to public APIs, and access to APIs that are not currently public (ex. Facebook, Koo) could provide us with more information to improve our accuracy, redundancy checks, and account for outliers.

Location Data

With better documentation of local landmarks, we can refine the search space and improve map visualisation.

Cloud Computing Resources

MLaaS and PaaS will increase processing power, reduce model training time, make it easier to manage storage and updation.

Fast, reliable, & efficient Disaster Response.