Effective Disaster Management Through Transformer-Based Multimodal Tweet Classification
Bibliographic record
Abstract
The role of social media in crisis response and recovery is becoming increasingly prominent due to the rapid progression of information and communication technologies.This study introduces a transformative approach to extract valuable information from the enormous volume of user-generated content on social media, specifically focusing on tweets that can significantly aid emergency response and recovery efforts.The identification of informative tweets allows emergency personnel to gain a more comprehensive understanding of crisis situations, thereby facilitating the deployment of more effective recovery strategies.Previous studies have largely focused on either the textual content or the accompanying visual elements within tweets.However, evidence suggests a complementary relationship between text and visuals, offering an opportunity for synergistic insights.In response to this, a novel deep learning framework is proposed, which concurrently analyses both textual and visual components extracted from user-generated tweets.The central architecture integrates established methodologies, including RoBERTa for text analysis, Vision Transformer for image understanding, Bi-LSTM for sequence processing, and an attention mechanism for context awareness.The innovation of this approach lies in its emphasis on multimodal fusion, introducing rank fusion techniques to effectively combine the strengths of textual and visual inputs.The proposed methodology is extensively tested across seven diverse datasets, representing various natural disasters such as wildfires, hurricanes, earthquakes, and floods.The experimental results demonstrate a superior performance of the proposed system, compared to several existing methods, with accuracy levels ranging from 94% to 98%.These findings underscore the efficacy of the proposed deep learning classifier in leveraging interactions across multiple modalities.In summary, this study contributes to disaster management by promoting a comprehensive approach that exploits the potential of multimodal data, thereby enhancing decision-making processes in emergency scenarios.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".