A Study of the Effects of Stemming Strategies on Arabic Document Classification

Alhaj, Yousif A.; Xiang, Jianwen; Zhao, Dongdong; Al-qaness, Mohammed A. A.; Elaziz, Mohamed E. Abd; Dahou, Abdelghani

doi:10.1109/access.2019.2903331

Cited by 64 publications

(43 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In addition to evaluating the accuracy of these classifiers correctly according to Sahih al-Bukhari book, which Muslims view as one of two most trusted books of hadith along with Sahih Muslim, https://www.islamicfinder.org/hadith/bukhari/ These hadiths have been collected and classified manually, therefore there is an urgency to reclassify them automatically using automated learning techniques and compared the results of the automatic classification to manual classification. Hadiths classification process consists of four main stages: the first one is text pre-processing [27]. Then the term weighting is used, known as text representation, term weighting was applied in this study using two approaches, the first is the Boolean algebra and the second is term frequencyinverse document frequency (TF-IDF).…”

Section: Introductionmentioning

confidence: 99%

Classification of Hadith According to Its Content Based on Supervised Learning Algorithms

2019

View full text Add to dashboard Cite

Given the importance of the Prophet's Hadith for Muslims all over the world, where it is the second source of Islam after the Qur'an and the fundamental resource of legislation in the Islam community. This study is focused on the Classification of hadith automatically into different categories according to its content, based on Hadith text. The objective of this study is to build a classifier model can classify and differentiate hadith categories, to predict its topic like prayer, fasting, and zakat; using data mining and machine learning techniques. In this study, many supervised learning algorithms plus combination methods such as the stacking algorithm was used to improve classification accuracy. The best three classifiers were evaluated mainly: the Decision Tree (DT), Random Forest (RF), and Naïve Bayes (NB), which achieved higher accuracy reached up to 0.965%, 0.956, and 0.951% respectively. Also, Binary (Boolean algebra) and TF-IDF methods as term weighting was applied to determine the frequency of each word in the hadith text, and identify the most significant features in training dataset using Information Gain (IG), and Chi-square (CHI). The experimental results showed that retrain these classifiers after applying IG and CHI as features selection; gave better accuracy compared to the previous results. Additional to, the best classifier gave high accuracy was DT, it has achieved higher accuracy in most test cases whether in the Boolean algebra or TF-IDF because it can deal with missing values and identifying the most essential features from the training dataset, known as features engineering.

show abstract

Section: Introductionmentioning

confidence: 99%

Classification of Hadith According to Its Content Based on Supervised Learning Algorithms

2019

View full text Add to dashboard Cite

show abstract

“…A further attempt was made to identify document using word2vec in combination with the LDA method [21], which also provided better results. In addition to text classification, word2vec has been used in many other areas of application such as improving medical knowledge through unsupervised medical corporate learning [22], answer selected from possible collection, good, poor in a question -response method [23], etc. Word2Vec is an unsupervised model of writing, writing the semantic context associated with the text.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Document Features Extraction with Clustering based Classification Framework on Large Document Sets

Devi¹,

Kumar²

2020

IJACSA

View full text Add to dashboard Cite

As the size of the document collections are increasing day-by-day, finding an essential document clusters for classification problem is one of the major problem due to high inter and intra document variations. Also, most of the conventional classification models such as SVM, neural network and Bayesian models have high true negative rate and error rate for document classification process. In order to improve the computational efficacy of the traditional document classification models, a hybrid feature extraction-based document cluster approach and classification approaches are developed on the large document sets. In the proposed work, a hybrid glove feature selection model is proposed to improve the contextual similarity of the keywords in the large document corpus. In this work, a hybrid document clustering similarity index is optimized to find the essential key document clusters based on the contextual keywords. Finally, a hybrid document classification model is used to classify the clustered documents on large corpus. Experimental results are conducted on different datasets, it is noted that the proposed document clustering-based classification model has high true positive rate, accuracy and low error rate than the conventional models.

show abstract

“…Empirically, the types of texts used for text categorization tasks can be varied and composed of various languages such as English, Chinese, and other languages [11], [12]. Unlike English text, Chinese text tasks usually require word segmentation because every character in Chinese is connected together [15].…”

Section: Chinese Radiology Text a Descriptionmentioning

confidence: 99%

“…The output gate O t will determine which outputs will be treated as current states. We update the information Z t in the LSTM by (12), and (14) can obtain the output H t of a hidden layer at time step t. Note that W and U are the network weight parameters, and is the ''Hadamard product'' operation.…”

Section: Long Short-term Memory Networkmentioning

confidence: 99%

“…However, existing methods are less efficient and incomplete in extracting this deep information and historical information in very long sequences since these methods are likely to have advantages in only one aspect. Chinese text classification methods in recent years have been based mainly on CNN and LSTM [6], [9], [10] as well as their variants and combinations, and they have achieved good results in the classification tasks of different languages [4], [11], [12]. Notably, although the machine learning algorithm has still been very popular in the last two years [13], some study results such as [39] show that deep neural networks have advantages over some machine learning methods in some classification tasks.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hierarchical Comprehensive Context Modeling for Chinese Text Classification

et al. 2019

View full text Add to dashboard Cite

The Chinese text classification task is challenging compared to tasks based on other languages such as English due to the characteristics of the Chinese text itself. In recent years, some popular methods based on deep learning have been used for text classification, such as the convolutional neural network (CNN) and the long short-term memory (LSTM) network. However, some problems are still encountered when classifying Chinese text. For example, important but obscure context information in Chinese text is not easily extracted. To improve the effect of Chinese text classification, we propose a novel classification model in this paper named the hierarchical comprehensive context modeling network (HCCMN) that can extract more comprehensive context. Our approach aims to extract contextual information and integrate it with the original input and then extract hierarchically more context, spatial information and high-weight local features from the integrated results. In addition, our method can remember long-term historical obscure information. Since Chinese radiology texts are complicated and difficult to obtain, we collected a Chinese radiology medical text dataset (CIRTEXT) containing more than 56,000 real-world data samples to verify the effect of this work. We conducted experiments on four datasets and showed that our HCCMN performs at state-of-the-art levels on three selected evaluation metrics compared to baselines. We present promising results showing that our hierarchical context modeling network extracts useful context from Chinese text more effectively and comprehensively. INDEX TERMS Chinese text, classification, contextual information, deep neural network.

show abstract

A Study of the Effects of Stemming Strategies on Arabic Document Classification

Cited by 64 publications

References 34 publications

Classification of Hadith According to Its Content Based on Supervised Learning Algorithms

Classification of Hadith According to Its Content Based on Supervised Learning Algorithms

A Hybrid Document Features Extraction with Clustering based Classification Framework on Large Document Sets

Hierarchical Comprehensive Context Modeling for Chinese Text Classification

Contact Info

Product

Resources

About