Effect of the Text Size on Stylometry—Application on Arabic Religious Texts

Ouamour, Siham; Khennouf, Salah; Bourib, S.; Hadjadj, Hassina; Sayoud, Halim

doi:10.1007/978-3-319-38884-7_16

Cited by 8 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The best performance they obtained with the 653 second chunk was around 87% accuracy. The same findings were reported in [5] in which the authors examined the performance of several classifiers, namely, SVM, MLP, Linear Regression, Stamatatos distance and Manhattan distance. The results confirmed the findings by Eder in 2013 for the English language [6] and the minimum size of textual data is 2500 words per documents.…”

Section: Introductionsupporting

confidence: 73%

“…In the literature, several features can be found. The interested reader is referred to [1], [5], [7] and [8]. These features can be categorized into seven main groups: lexical, character, syntactic, semantic, content-specific, structural and language-specific.…”

Section: Authorship Attribution 21 the Selected Writing Style Featuresmentioning

confidence: 99%

“…It presents accuracy of the used classifiers of those works mentioned throughout this paper regardless the text size, whilst Table 6 present the accuracy taking in consideration the same text size. [2] 87.63% Arabic books obtained from the website of the Arab Writers Union Abbasi and Chen [3] 85.4% Arabic web forum messages from Yahoo groups Stamatatos [4] 93.4% Arabic newspaper report of Alhayat Ouamour et al [5] 100% Ancient historical books in Arabic Altheneyan and El Bachir Menai [12] 97,84% Arabic books collected from Alwaraq website [8] 59.8% Up to 140 words per tweets…”

Section: Comparison With Other Workmentioning

confidence: 99%

See 2 more Smart Citations

The effect of training set size in authorship attribution: application on short arabic texts

Al-Sarem¹,

Emara²

2019

IJECE

View full text Add to dashboard Cite

<span lang="EN-US">Authorship attribution (AA) is a subfield of linguistics analysis, aiming to identify the original author among a set of candidate authors. Several research papers were published and several methods and models were developed for many languages. However, the number of related works for Arabic is limited. Moreover, investigating the impact of short words length and training set size is not well addressed. To the best of our knowledge, no published works or researches, in this direction or even in other languages, are available. Therefore, we propose to investigate this effect, taking into account different stylomatric combination. The Mahalanobis distance (MD), Linear Regression (LR), and Multilayer Perceptron (MP) are selected as AA classifiers. During the experiment, the training dataset size is increased and the accuracy of the classifiers is recorded. The results are quite interesting and show different classifiers behaviours. Combining word-based stylomatric features with n-grams provides the best accuracy reached in average 93%.</span>

show abstract

Section: Introductionsupporting

confidence: 73%

Section: Authorship Attribution 21 the Selected Writing Style Featuresmentioning

confidence: 99%

Section: Comparison With Other Workmentioning

confidence: 99%

See 1 more Smart Citation

The effect of training set size in authorship attribution: application on short arabic texts

Al-Sarem¹,

Emara²

2019

IJECE

View full text Add to dashboard Cite

show abstract

“…Such subjects as Stylometry Analysis and examining the discrimination between the hadith content and Quran or non-hadith content are among other studies addressed in some of the articles. [54]- [56] Analysis of Hadith Words is another subject of this subcategory. In the articles of this group, in addition to machine parsing, morphological and syntactical analysis of hadiths literatures that is considered as infrastructural task, certain efforts have also been made with the purpose of Word Sense Disambiguation base on the adjacent words analysis after preparation of required dataset.…”

Section: B Hadith Contentmentioning

confidence: 99%

Artificial Intelligence and Modern Information Technologies Applications in Islamic Sciences: A Survey

Rabiei Zadeh

2023

IJPCC

View full text Add to dashboard Cite

Considering the progress of Artificial Intelligence (AI) and the Information Technology (IT) we witness, during recent years, the spread of the application of these technologies in various fields. The research workflows, and in particular, the researches on Islamic sciences are not excepted from this issue. Several works have been carried out in order to exploit the AI and modern information technologies in the researches on Islamic sciences during recent years all over the Islamic regions and beyond them. It is very important to be aware of the latest developments in this field from different aspects like: 1) Benefiting from the advantages of modern technologies in the Islamic researches, 2) Reorganizing the educational plans in accordance with these developments, and 3) Introducing the new applications of AI in Islamic studies to the academics of computer sciences who may be interested in this field. In this paper, in the first step, a systematic review was conducted concerning more than four thousand international scientific articles related to applying AI and modern IT in Islamic studies, out of which 975 ones were chosen. At the same time, major institutions in this field were identified. In the next step the selected articles were classified in five thematic fields of 1) the Holy Qur’an, Tafsir and other related issues, 2) Hadith and Rijal Sciences, 3) Islamic Law and Jurisprudence, 4) the General Islamic Content in Social Media, 5) Other Subjects related to Islamic Sciences like Linguistics, History, Geography, etc. In the third step, the articles of each category were classified in a number of major subcategories that amount to 73 in total. Finally, in the last step, the distinctive articles in each field were introduced briefly.

show abstract

“…Most of this scholarship comes from humanities specialists interested in computers Feener et al, 2018;Keshani, 2012;Lange et al, 2021;Micklewright, 2021;Miller et al, 2018;Muhanna, 2016;Romanov, 2017;Sadeghi, 2011;Svensson, 2019;Syed et al, 2019;van Lit, 2019; also see Chaney, 2013Chaney, , 2016. However, contributions have also been made by computer scientists interested in Islamic texts (see e.g., Atwell et al, 2011;Azmi et al, 2019;Harrag et al, 2020;Ouamour et al, 2016;Sayoud, 2012Sayoud, , 2014Saloot et al, 2016). Some DH scholarship in Islamic studies engages with visual material like images of artifacts/buildings or maps (e.g., Feener et al, 2018;Keshani, 2012;Micklewright, 2021).…”

Section: Islamic Studies and Digital-cultural Statisticsmentioning

confidence: 99%

Cognitive and quantitative approaches to Islamic studies: Integrating psychological, socioeconomic, and digital‐cultural statistics

Nakissa¹

2021

Religion Compass

View full text Add to dashboard Cite

The past 3 decades have witnessed growing efforts to rethink areas of inquiry traditionally dominated by the humanities. One such area is religion, and, by extension, “Islamic studies” (i.e., the study of Islam and Muslim societies). Recent efforts to rethink religion and Islamic studies draw insights from cognitive science, and also frequently employ quantitative approaches. Such approaches make use of psychological statistics, socioeconomic statistics, and statistical data extracted from massive collections of digitized cultural artifacts (e.g., texts, photos, sound recordings). These types of statistical data make it possible to quantitively describe long‐term global psychological, socioeconomic, and cultural trends; explain how these trends relate to one another; and explain how these trends relate to Muslim societies. This article introduces the emerging lines of research mentioned above. Furthermore, it suggests that these lines of research open the way for a new more ambitious and more interdisciplinary Islamic studies – one which builds on valuable forms of humanities expertise while integrating the best insights from non‐humanities fields.

show abstract

Effect of the Text Size on Stylometry—Application on Arabic Religious Texts

Cited by 8 publications

References 9 publications

The effect of training set size in authorship attribution: application on short arabic texts

The effect of training set size in authorship attribution: application on short arabic texts

Artificial Intelligence and Modern Information Technologies Applications in Islamic Sciences: A Survey

Cognitive and quantitative approaches to Islamic studies: Integrating psychological, socioeconomic, and digital‐cultural statistics

Contact Info

Product

Resources

About