Recent years include the world’s hottest year, while they have been marked mainly, besides the COVID-19 pandemic, by climate-related disasters, based on data collected by the Emergency Events Database (EM-DAT). Besides the human losses, disasters cause significant and often catastrophic socioeconomic impacts, including economic losses. Recent developments in artificial intelligence (AI) and especially in machine learning (ML) and deep learning (DL) have been used to better cope with the severe and often catastrophic impacts of disasters. This paper aims to provide an overview of the research studies, presented since 2017, focusing on ML and DL developed methods for disaster management. In particular, focus has been given on studies in the areas of disaster and hazard prediction, risk and vulnerability assessment, disaster detection, early warning systems, disaster monitoring, damage assessment and post-disaster response as well as cases studies. Furthermore, some recently developed ML and DL applications for disaster management have been analyzed. A discussion of the findings is provided as well as directions for further research.

/pdf/machine-learning-in-disaster-management-recent-developments-27sbvqk5.pdf

Machine Learning in Disaster Management: Recent Developments in Methods and Applications

In this study, we propose a method for evaluating cascading disasters using natural language processing to comprehensively and objectively extract causal relations between disaster events based on newspaper articles and construct cascading disaster networks. The applicability of this method is illustrated by employing data extracted from Japanese newspaper articles on the Great Hanshin-Awaji Earthquake and the Great East Japan Earthquake. Both disasters showed complex chain relations, and the proposed method extracted escalation points such as the “stoppage of factory production in the stricken area” for the Great Hanshin-Awaji Earthquake and “scheduled power outages outside the stricken areas” for the Great East Japan Earthquake based on each network structure. When matched with Alexander's magnitude scale for cascading disasters, both these catastrophic events were confirmed to reach Level 5.

Designing cascading disaster networks by means of natural language processing

We know neither what the future holds nor what will occur in the future. It is impossible to anticipate what the next instant may bring, and it could be awful. A natural disaster is an unforeseen event, which can have a tremendous impact on human life and the environment. The internet provides many sources that generate vast amounts of news articles daily. As the amount of news stories online increases, it's becoming more difficult for people to access disaster-relevant news, which makes it necessary to extract and classify news to be easily accessed. This paper presents an automated system that scraps news from various online sources and identifies disaster-relevant news. This paper also states the performance evaluation for classifying the natural disaster types based on machine learning in Indonesia's online news. Our results show that relevant Indonesian's online news about natural disasters can be the accuracy around 96% using the Support Vector Machine for three classes of natural disasters. In the Indonesian news data set, the machine learning algorithm that gets the highest value out of all the parameters is the best.

Classification of Natural Disaster on Online News Data Using Machine Learning

Social networks provide a plethora of information for gathering extra data on people’s behavior, trends, opinions, and feelings during human-affecting occurrences, such as natural catastrophes. Twitter is an inevitable communication medium during calamities. People mainly depend on Twitter to announce real-time emergencies. However, it is rarely straightforward if someone is declaring a tragedy. Sentiment analysis of disaster tweets aid in situational awareness and realizing the disaster dynamics. In our paper, we perform a sentimental analysis of disaster tweets using techniques based on machine learning and deep learning. The tweets are pre-processed before being converted into a structured form using Natural Language Processing (NLP) methods. Supervised learning techniques such as the Support Vector Machine and the Naive Bayes Classifier algorithm are used to develop the Classifier, which categorizes tweets into distinct catastrophes and selects the most appropriate algorithm. The chosen algorithm is further enriched with an emoticon detection algorithm for explicit elucidation. Our research would help disaster relief organizations and news agencies to conclude about the state of affairs and do the needful. 

Classification of Disaster Tweets using Machine Learning and Deep Learning Techniques

Social networks provide a plethora of information for gathering extra data on people’s behavior, trends, opinions, and feelings during human-affecting occurrences, such as natural catastrophes. Twitter is an inevitable communication medium during calamities. People mainly depend on Twitter to announce real-time emergencies. However, it is rarely straightforward if someone is declaring a tragedy. Sentiment analysis of disaster tweets aid in situational awareness and realizing the disaster dynamics. In our paper, we perform a sentimental analysis of disaster tweets using techniques based on machine learning and deep learning. The tweets are pre-processed before being converted into a structured form using Natural Language Processing (NLP) methods. Supervised learning techniques such as the Support Vector Machine and the Naive Bayes Classifier algorithm are used to develop the Classifier, which categorizes tweets into distinct catastrophes and selects the most appropriate algorithm. The chosen algorithm is further enriched with an emoticon detection algorithm for explicit elucidation. Our research would help disaster relief organizations and news agencies to conclude about the state of affairs and do the needful.

We are living in unprecedented times and anyone in this world could be impacted by natural disasters in some way or the other. Life is unpredictable and what is to come is unforeseeable. Nobody knows what the very next moment will hold, maybe it could be a disastrous one too. The past cannot be changed but it can act constructively towards the betterment of the current situation, ‘Precaution is better than cure’. To be above this uncertain dilemma of life and death situations, ‘Automated Identification of Disaster News for Crisis Management is proposed using Machine Learning and Natural Language Processing’. A software solution that can help disaster management websites to dynamically show the disaster relevant news which can be shared to other social media handles through their sites. The objective here is to automatically scrape news from English news websites and identify disaster relevant news using natural language processing techniques and machine learning concepts, which can further be dynamically displayed on the crisis management websites. The complete model is automated and requires no manual labor at all. The architecture is based on Machine Learning principles that classifies news scraped from top news websites using a spider-scraper into two categories, one being disaster relevant news and other being disaster irrelevant news and eventually displaying the relevant disaster news on the crisis management website.

/pdf/automated-identification-of-disaster-news-for-crisis-1yeqwin4f6.pdf

Automated Identification of Disaster News for Crisis Management using Machine Learning and Natural Language Processing

Songs are melodic manifestations that are performed by individuals. These tunes are made altogether by both a lyricist who composes the verses and the artist who sings them. Verse writing in itself is an exceptionally selective and characterized issue. The ever-expanding utilization of innovation and the way that they are effectively accessible to us make human lives comfortable. A lyricist can often have a mind block while considering verses or may even find it difficult to get an idea. The principal reason for this exploration is to enable the lyricist to get a motivation that can assist him in making better verses. To accomplish this, a profound learning method is utilized alongside the idea of natural language processing. Specifically, bidirectional long short-term memory (LSTM) networks are used for lyric generation. The proposed framework can exceptionally create versus relying upon the information seed and the scope of words.

Lyrics Inducer Using Bidirectional Long Short-Term Memory Networks

Story-Plot Recommender Using Bidirectional LSTM Network

Agriculture is one of the top three contributors to the Indian economy. However, most of the agricultural practices are still quite traditional in nature. Farmers have to make innumerable decisions on various factors which in return affect the yield and productivity of the crop production. Some of these factors include season, yield, location, revenue, type of crop, etc. If farmers and agricultural businesses made decisions based on engineered crop data of various years, then the overall crop production and yield can be maximized. This paper presents a data engineering procedure to analyze crop data and provide recommendations. It is divided into three main modules, viz recommendation of crop season and location, recommendation of crop harvesting months and comparison of fruit and vegetable crops. The datasets are procured from the Open Government Data (OGD) Platform of India and the agricultural produce marketing committee (APMC). The techniques involve data collection, preprocessing, normalization and data querying. Interactive data visualizations were performed with the help of Plotly library in Python. The resultant system provides suggestions based on season, location and revenue. Moreover, it provides suggestions of crops based on comparison with respect to yield.

Improving the Yield and Revenue of Indian Crop Production Using Data Engineering

As there is aggrandizement in the sector of artificial intelligence relating to the speech domain, it becomes a necessity to have efficient noise removal models with greater efficiency and less complexity. The presence of noise in audio signals poses a great complication when working on speech recognition, enhancement, improvement, and transmission. Hence, there is a necessity to develop the most efficient algorithm for noise reduction which works in real time and is successful in removing maximum noise. To be above this difficulty, this paper presents an efficient algorithm for noise detection which works on the principles of deep learning, specifically convolutional neural networks (CNNs) and the removal of similar noise from the audio using the Python module ‘noise reducer.’

Jayashree Domala

Papers

Automated Identification of Disaster News for Crisis Management using Machine Learning and Natural Language Processing

Lyrics Inducer Using Bidirectional Long Short-Term Memory Networks

Story-Plot Recommender Using Bidirectional LSTM Network

Improving the Yield and Revenue of Indian Crop Production Using Data Engineering

Noise Removal from Audio Using CNN and Denoiser