Improving Arabic Text Categorization using Normalization and Stemming Techniques

Sallam, Rouhia M.; Mousa, Hamdy M.; Hussein, Mahmoud

doi:10.5120/ijca2016908328

Cited by 15 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On PS and CV test mode, J48 had the most significant accuracy decrease from 76.3% to 64.2% in PS mode, and 69.69% to 62.6% in CV mode. Similar research conclusion was also obtained by [21] in their research. They performed Arabic text classification using three datasets that were taken from two trusted sources.…”

Section: Related Worksupporting

confidence: 89%

Stemming Impact Analysis on Indonesian Quran Translation and Their Tafsir Classification for Ontology Instances

Utomo

Suryana

Azmi

2020

IIUMEJ

View full text Add to dashboard Cite

The current gap which appears in the Quran ontology population domain is stemming impact analysis on Indonesian Quran translation and their Tafsir to develop ontology instances. The existing studies of stemming effect analysis performed in various languages, dataset, stemming method, cases, and classifier. However, there is a lack of literature that studies about stemming influence on instances classification for Quran ontology with different dataset, classifier, Quran translation, and their Tafsir on Indonesian. Based on this problem, our study aims to investigate and analyze the stemming impact on instances classification results using Indonesian Quran translation and their Tafsir as datasets with multiple supervised classifiers. Our classification framework consists of text pre-processing, feature extraction, and text classification stage. Sastrawi stemmer was used to perform stemming operation in text pre-processing stage. Based on our experiment results, it was found that Support Vector Machine (SVM) with Term Frequency-Inverse Document Frequency (TF-IDF) and stemming operation owns the best classification performance, i.e., 70.75% for accuracy and 71.55% for precision in Indonesian Quran translation dataset on 20% test data size. While in 30% test data size, SVM and TF-IDF with stemming process own the best classification performance, i.e., 67.30% for accuracy and 68.10% for precision in Ministry of Religious Affairs Indonesia dataset. Furthermore, in this study, it was also discovered that the Backpropagation Neural Network has the most precision and accuracy reduction due to the negative impact of stemming operations.

show abstract

Section: Related Worksupporting

confidence: 89%

Stemming Impact Analysis on Indonesian Quran Translation and Their Tafsir Classification for Ontology Instances

Utomo

Suryana

Azmi

2020

IIUMEJ

View full text Add to dashboard Cite

show abstract

“…The PIR is computed as formulated the Eq. (35). In case of RW-CB-Twitter, the PIRs of the STTM models LDA, BTM, PTM, GLTM, FTM, and WNTM are 1.82%, 1.96%, 2.36%, 2.51%, 3.05%, and 2.73%, respectively.…”

Section: ) Topic Coherence Evaluation Results With Smdcm and Baseline...mentioning

confidence: 96%

“…Duwairi and El-Orfali [31] studied the impact of various pre-processing techniques like n-gram models, feature correlation on Arabic text sentiment analysis. Some other works studied the impacts of stemming on the Arabic text classification performance, such as [32], [33], [34], and [35].…”

Section: ) Preprocessingmentioning

confidence: 99%

See 1 more Smart Citation

Enhancing Big Social Media Data Quality for Use in Short-Text Topic Modeling

et al. 2022

View full text Add to dashboard Cite

With the emergence of microblogging platforms and social media applications, large amounts of user-generated data in the form of comments, reviews, and brief text messages are produced every day. Microblog data is typically of poor quality; hence improving the quality of the data is a significant scientific and practical challenge. In spite of the relevance of the problem, there has been not much work so far, especially in regard to microblog data quality for Short-Text Topic Modelling (STTM) purposes. This paper addresses this problem and proposes an approach called the social media data cleansing model (SMDCM) to improve data quality for STTM. We evaluate SMDCM using six topic modelling methods, namely the Latent Dirichlet Allocation (LDA), Word-Network Topic Model (WNTM), Pseudo-document-based Topic Modelling (PTM), Biterm Topic Model (BTM), Global and Local word embedding-based Topic Modeling (GLTM), and Fuzzy Topic modelling (FTM). We used the Real-world Cyberbullying Twitter (RW-CB-Twitter) and the Cyberbullying Mendeley (CB-MNDLY) datasets in the evaluation. The results proved the efficiency of the GLTM and WNTM over the other STTM models when applying the SMDCM techniques, which achieved optimum topic coherence and high accuracy values on RW-CB-Twitter and CB-MNDLY datasets.

show abstract

“…SANAD comprises of the above three datasets, which makes it the largest, to our knowledge, available and representative corpus. In contrast with other few available datasets such as those used in [2], [3], [4], SANAD is large enough to enable researchers to implement classical and deep learning models for text classification as it is the case in [1], which used for sentiment classification. Few similar datasets already exist but are not comparable in size and have less tags.…”

Section: Experimental Design Materials and Methodsmentioning

confidence: 99%

SANAD: Single-label Arabic News Articles Dataset for automatic text categorization

2019

View full text Add to dashboard Cite

Text Classification is one of the most popular Natural Language Processing (NLP) tasks. Text classification (aka categorization) is an active research topic in recent years. However, much less attention was directed towards this task in Arabic, due to the lack of rich representative resources for training an Arabic text classifier. Therefore, we introduce a large Single-labeled Arabic News Articles Dataset (SANAD) of textual data collected from three news portals. The dataset is a large one consisting of almost 200k articles distributed into seven categories that we offer to the research community on Arabic computational linguistics. We anticipate that this rich dataset would make a great aid for a variety of NLP tasks on Modern Standard Arabic (MSA) textual data, especially for single label text classification purposes. We present the data in raw form. SANAD is composed of three main datasets scraped from three news portals, which are AlKhaleej, AlArabiya, and Akhbarona. SANAD is made public and freely available at https://data.mendeley.com/datasets/57zpx667y9 .

show abstract

Improving Arabic Text Categorization using Normalization and Stemming Techniques

Cited by 15 publications

References 16 publications

Stemming Impact Analysis on Indonesian Quran Translation and Their Tafsir Classification for Ontology Instances

Stemming Impact Analysis on Indonesian Quran Translation and Their Tafsir Classification for Ontology Instances

Enhancing Big Social Media Data Quality for Use in Short-Text Topic Modeling

SANAD: Single-label Arabic News Articles Dataset for automatic text categorization

Contact Info

Product

Resources

About