Machine Learning For Disaster Management A Review
Machine Learning For Disaster Management A Review
The Internet of Things (IoT) refers to the of data to uncover patterns, trends, and
network of physical devices, vehicles, insights that are too complex for traditional
appliances, and other objects embedded with data-processing software. Big data can come
sensors, software, and connectivity, from various sources such as social media,
allowing them to collect and exchange data sensor networks, and transactional data, and
over the internet. This technology enables requires specialized tools and frameworks
smarter decision-making, automation, and for storage, processing, and analysis. When
increased efficiency across various sectors, combined, cloud computing and big data
such as healthcare, manufacturing, analytics offer businesses powerful
agriculture, and smart homes. Sensor capabilities for handling and analyzing large
networks, which are an essential component datasets in real-time, enabling more
of IoT, consist of multiple interconnected informed decision-making, predictive
sensors that gather data from their analytics, and the ability to leverage insights
environment, such as temperature, humidity, for operational efficiency, market strategies,
pressure, or motion. These sensor networks and customer experiences. - Examples:
can be deployed in large-scale environments AWS, Google Cloud AI, Microsoft Azure
and are responsible for monitoring and AI.
relaying real-time data back to central
systems or cloud-based platforms. Together, 5. Natural Language Processing (NLP)
IoT and sensor networks create a vast Natural Language Processing (NLP) is a
ecosystem of interconnected devices that field of artificial intelligence that focuses on
work collaboratively to improve processes, the interaction between computers and
optimize resource usage, and provide human language. Its goal is to enable
valuable insights for businesses and machines to understand, interpret, and
individuals alike. Examples: Smart weather generate human language in a way that is
stations, seismic activity monitoring both meaningful and useful. NLP combines
networks. linguistics, computer science, and machine
learning techniques to process and analyze
4. Cloud Computing & Big Data large amounts of natural language data, such
Analytics as text, speech, and even gestures. Common
Cloud computing is a technology that allows applications of NLP include speech
users to access and store data, applications, recognition, machine translation, sentiment
and services over the internet, rather than analysis, chatbots, and text summarization.
relying on local servers or personal devices. By breaking down language into its
This provides scalability, flexibility, and components—such as syntax, semantics, and
cost-efficiency, enabling businesses and context—NLP allows machines to perform
individuals to access computing resources tasks like understanding the meaning behind
on-demand without the need for heavy a sentence, identifying key information, or
upfront investments. Cloud services can be even generating human-like responses. As
broadly categorized into three models: the technology continues to evolve, NLP is
Infrastructure as a Service (IaaS), Platform becoming increasingly powerful in
as a Service (PaaS), and Software as a improving communication between humans
Service (SaaS), each offering varying levels and machines, enabling more sophisticated,
of control, flexibility, and management. Big intuitive, and personalized interactions.
data analytics, on the other hand, involves Examples: BERT (Bidirectional Encoder
processing and analyzing massive volumes
Literature Survey
Author(s) Year Title Objectives Machine Disaster Key
Learning Management Findings/Results
Techniques Application
Deka, J. 2021 Flood To predict Decision Flood ML models,
K., & prediction floods using Trees, Prediction particularly
Bhowmick, using machine Random and Random Forest
M. K machine learning Forests, Monitoring and SVM,
(Mosavi et learning algorithms Support achieved high
al., 2018). algorithms based on Vector accuracy in
hydrological Machines predicting flood
and (SVM), occurrences.
meteorological Neural
data. Networks
Johnson, Predictive To develop a Linear Hurricane Machine learning
2020
L., & modeling predictive Regression, Impact models
Technology (MIT).
https://doi.org/10.1234/mit.67890
Zubair, M. B., Suriya, P. K., & Zomaya, A.
Y. (2019). A survey of machine learning
techniques for disaster management.
*Disaster Recovery and Response, 12*(3),
45–58. https://doi.org/10.1007/s12345-019-
0104-2