Disaster tweet classification: A majority voting approach using machine learning algorithms

Abstract

Nowadays, People share their opinions through social media. This information may be informative or non-informative. To filtering the informative information from the social media plays a challenging issue. Nevertheless, in social media especially when a disaster been occurs the peoples will interact more on that particular disaster event. They share their opinion through some textual information such as tweets or posts. In this work, we are proposing a generalized approach for categorizing the informative and non-informative on twitter media. We collected the seven natural disaster events from the crisisNLP. These datasets are different disaster events which contains the people’s opinions on that specific event. We preprocess the information which converts the tweet information into machine understandable vectors. These vectors been processed by the different machine learning algorithms. We consider the individual performance of each ML algorithm on different disaster datasets upon chosen the best five algorithms for voting techniques. We tested the performance with matrices such as accuracy, precision, recall and f1-score. We compared our results with existing models in which our proposed model performed better than other existing state of art models.

Optimizing Safety Stock Placement in Large Real-World Automotive Supply Networks Using the Guaranteed-Service Model

“…Computer programmers approach AI from a more technical perspective in that they define it as a field of study that relates computation to cognition, with the intention of writing programs that attempt to achieve an intelligent behavior (Barr & Feigenbaum, 2014). According to Krishna et al (2017), AI is becoming a whole theory of developing systems that are able to perform tasks that normally require human intelligence. Even if these definitions agree about the increasing abilities of machines, applications, and systems to accomplish specific roles and tasks habitually performed by humans in the workplace and the society in general (Dwivedi et al, 2019), it is worth noting that the diversity of definitions can be explained by the evolutionary nature of AI.…”

DPHFM: A Deep Parallel Hybrid Fusion Model for Disaster Tweet Classification on Twitter Data

“…However, they also acknowledge the challenges and limitations of working with Twitter data, such as data noise and false positives. The authors [39] proposed majority voting approaches in which topperforming machine learning models were chosen for voting approaches, and also the authors [42] also use a stacking ensemble based on machine learning models. In this approach, there are two layers, base learner and meta-leaner, for prediction of identification of informative tweets, and the authors proposed three-tier approaches [40] for private cloud for the performance of the machine learning models.…”

“…Imbalance occurs when some classes have significantly fewer samples than others. The equations [38][39][40]…”

“…[3] The authors propose features based mechanism with machine learning based classifier for classification) R4 Krishna, D. S., et al. [39] The authors proposed majority voting approaches in which chosen best performed top five machine learning models for prediction of damage assessment on disaster tweets R5 Dasari, S. K., et al. [42] The authors proposed stacking ensemble approach based on machine learning models for the disaster tweet classification for the prediction.…”

“…Late the following recent trends, many researchers focused on machine learning based [3][4][5], deep learning based [11] approaches. In the recent past, most researchers have focused on transfer learning approaches such as voting [31,39], stacking [10,42], and BERT [33] to improve classification tasks. This work proposes a DPHFM for Twitter text classification on social media platforms.…”