MétaCan
Menu
Back to cohort
Record W4390271731 · doi:10.18280/ria.370630

A Hybrid CNN-LSTM and XGBoost Approach for Crime Detection in Tweets Using an Intelligent Dictionary

2023· article· en· W4390271731 on OpenAlexvenueno aff
Zainab Khyioon Abdalrdha, Abbas M. Al-Bakry, Alaa Kadhim Farhan

Bibliographic record

VenueRevue d intelligence artificielle · 2023
Typearticle
Languageen
FieldComputer Science
TopicHate Speech and Cyberbullying Detection
Canadian institutionsnot available
Fundersnot available
KeywordsComputer scienceArtificial intelligenceMachine learningPattern recognition (psychology)Natural language processing

Abstract

fetched live from OpenAlex

As social media grows, recognizing and managing illicit content, including threats, harassment, hate speech, armed robbery, drug smuggling, blackmail, and other crimes, is crucial.The present study uses machine learning and deep learning to create an intelligent lexicon for identifying crime-related material in Twitter tweets.The Aho-Corasick technique effectively creates a dictionary for extensive text corpus search, categorization, and keyword-based action execution.This strategy overcomes the evolution of language dynamics and criminal vocabulary to improve crime-related information detection.This paper aims to gather accurate data to help law enforcement identify and prevent specific crimes.For tweet preprocessing and extraction of relevant information like textual patterns and other distinctive qualities, natural language processing (NLP) technologies are prioritized.The paper describes labeling tweets into crime categories.This dataset trains supervised learning models to categorize tweets as criminal or not.XGBoost and Hybrid CNN-LSTM are combined for this.The suggested technique is assessed using precision, recall, F1-Score, accuracy, and MAP accuracy measures.These metrics measure the model's crime-related tweet identification accuracy.The Arabic tweets dataset, encompassing 18493 tweets and 10 features, is utilized for model testing.After training, the Hybrid CNN-LSTM model demonstrated an accuracy of 99.84% and a macro F1-Score of 98.20%.When the XGBoost method was employed, the traditional machine learning model achieved a peak F1 macro score of 99.36% and a maximum accuracy of 100%.The results suggest that while the deep learning models outperform machine learning models in the F1-Score, XGBoost exhibits superior accuracy.The paper presents a comprehensive strategy for crime detection in tweets, potentially offering a significant tool for law enforcement agencies.

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.728
Threshold uncertainty score0.871

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.067
GPT teacher head0.288
Teacher spread0.221 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

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".

Quick stats

Citations9
Published2023
Admission routes1
Has abstractyes

Explore more

Same venueRevue d intelligence artificielleSame topicHate Speech and Cyberbullying DetectionFrench-language works237,207