Authorship Attribution of Social Media Messages

Theóphilo, Antônio; Giot, Romain; Rocha, Anderson

doi:10.1109/tcss.2021.3123895

Cited by 15 publications

(10 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Traditional feature representations, such as Tf-Idf, suffers from data sparsity and high dimensionality in representing n-grams and have difficulty in grasping the semantic meaning of texts [15]. Studies in AI starts to employ deep learning and embeddings to enhance the performance of AI system in attributing the author of OSN text [16,48].…”

Section: Reviews Of Ai Work In Smfmentioning

confidence: 99%

See 1 more Smart Citation

Towards Curbing Cyber-Bullying in Malaysia by Author Identification of Iban and Kadazandusun Osn Text Using Deep Learning

Tarmizi

Saee

Ibrahim

2023

AEJ

View full text Add to dashboard Cite

Online Social Network (OSN) is frequently used to carry out cyber-criminal actions such as cyberbullying. As a developing country in Asia that keeps abreast of ICT advancement, Malaysia is no exception when it comes to cyberbullying. Author Identification (AI) task plays a vital role in social media forensic investigation (SMF) to unveil the genuine identity of the offender by analysing the text written in OSN by the candidate culprits. Several challenges in AI dealing with OSN text, including limited text length and informal language full of internet jargon and grammatical errors that further impact AI's performance in SMF. The traditional AI system that analyses long text documents seems inadequate to analyse short OSN text's writing style. N-gram features are proven to efficiently represent the authors' writing style for shot text. However, representing N-grams in traditional representation like Tf-IDF resulted in sparse and difficult in grasping the semantic information from text. Besides, most AI works have been done in English but receive less attention in indigenous languages. In West Malaysia, the supreme languages that transcend ethnic boundaries are Iban of Sarawak and KadazanDusun of Sabah, which both are inherently under-resourced. This paper presented a proposed workflow of AI for short OSN text using two Under-Resourced Language (U-RL), Iban and KadazanDusun tweets, to curb the cyberbullying issue in Malaysia. This paper compares Tf-Idf (sparse) and SoA embedding-based (dense) feature representations to observe which representations best represent the stylistic features of the authors’ writing. N-grams of word, character, and POS were extracted as the features. The representation models were learned by different classifiers using machine learning (Naïve Bayes, Random Forest, and SVM). The convolutional neural network (CNN), a SoA deep learning model in sentence classification, was tested against the traditional classifiers. The result was observed by combining different representation models and classifiers on three datasets (English, Iban, and KadazanDusun). The best result was achieved when CNN learned embedding-based models with a combination of all features. KadazanDusun achieved the highest accuracy with 95.76%, English with 95.02%, and Iban with 94%..

show abstract

Section: Reviews Of Ai Work In Smfmentioning

confidence: 99%

“…Random forest reported better than SVM and k-NN with a gradual reduction in accuracy when the number of authors increased. Besides prominent machine learning classifiers, deep learning models like neural networks are gaining attention in text classification [21] as well as in AI task [16,22].…”

Section: Introductionmentioning

confidence: 99%

Towards Curbing Cyber-Bullying in Malaysia by Author Identification of Iban and Kadazandusun Osn Text Using Deep Learning

Tarmizi

Saee

Ibrahim

2023

AEJ

View full text Add to dashboard Cite

show abstract

“…For Authorship Verification, we adopt two subsets of a dataset of tweets [35]. The first subset contains messages from 100 authors, which we randomly split into two sets: 50 authors for training and 50 for testing.…”

Section: A Datasetsmentioning

confidence: 99%

Reasoning for Complex Data through Ensemble-based Self-Supervised Learning

Bertocco¹,

Theophilo²,

Andaló³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

“…En primer lugar la denominada Identificación de Autoría (IA) se emplea para determinar la probabilidad de que un autor sea el responsable de un escrito dado, de entre un conjunto de autores sospechosos (se asume que el verdadero autor se encuentra entre los posibles sospechosos, lo que no es cierto en la mayoría de los escenarios del mundo real) (Okuno et al, 2014;Alhijawi et al, 2018). En segundo lugar la Verificación de Autoría (VA) ayuda a confirmar si un sospechoso es o no el autor de un texto particular (Nirkhi Smita, 2016;White and Sprague, 2021;Theophilo et al, 2021;Boenninghoff et al, 2019). Finalmente, la Caracterización de Autoría (CA) se utiliza para recopilar meta-información del autor, como el género, antecedentes lingüísticos, edad, etc.…”

unclassified

“…Finalmente, la Caracterización de Autoría (CA) se utiliza para recopilar meta-información del autor, como el género, antecedentes lingüísticos, edad, etc. a partir de un documento anónimo (Theophilo et al, 2021;Vivitha Vijayan, 2019). En este trabajo se propone una metodología para atacar el problema de la VA y potencialmente la IA.…”

unclassified

Métricas para verificación de autoría y emulación de procesos cognitivos

Pérez-Cortés

Gómez-Pozos²,

Ruíz³

2022

ICBI

View full text Add to dashboard Cite

Se proponen dos medidas cuantitativas, el parámetro de Hurst y la legibilidad, para realizar la verificación de autor. Se determinan las cantidades mencionadas para seis diferentes autores considerando seis obras de cada uno de ellos. Con estos valores se construye un espacio de dos dimensiones donde cada punto corresponde a un único autor; midiendo la distancia entre dos puntos de dicho espacio es posible decidir si un texto es atribuible a un autor o no. Adicionalmente dichas medidas proporcionan una interpretación cualitativa, es decir, en términos como la facilidad al leer un texto y si existen palabras, asociados, a pensamientos, que persisten en un texto.

show abstract

Authorship Attribution of Social Media Messages

Cited by 15 publications

References 31 publications

Towards Curbing Cyber-Bullying in Malaysia by Author Identification of Iban and Kadazandusun Osn Text Using Deep Learning

Towards Curbing Cyber-Bullying in Malaysia by Author Identification of Iban and Kadazandusun Osn Text Using Deep Learning

Reasoning for Complex Data through Ensemble-based Self-Supervised Learning

Métricas para verificación de autoría y emulación de procesos cognitivos

Contact Info

Product

Resources

About