Stemming as a feature reduction technique for Arabic Text Categorization

Harrag, Fouzi; El-Qawasmah, Eyas; Al-Salman, AbdulMalik S.

doi:10.1109/isps.2011.5898874

Cited by 42 publications

(23 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Harrag et al [9] conducted a comparative study of three pretreatment techniques: light stemming, root-based stemming and dictionary lookup stemming in order to reduce the feature space. Then two classifiers were tested, Artificial Neural Networks and SVM.…”

Section: Related Workmentioning

confidence: 99%

Features based approach for indexation and representation of unstructured Arabic documents

Bazzi¹,

Mammass²,

Ennaji³

et al. 2017

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Features based approach for indexation and representation of unstructured Arabic documents

Bazzi¹,

Mammass²,

Ennaji³

et al. 2017

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

show abstract

“…Root-based stemming, light stemming, and dictionarylookup stemming are three different types of stemming [6]. Root-based stemmers are based on a pattern-matching technique to find the root of the word.…”

Section: Stemming Techniquementioning

confidence: 99%

“…Early research in this field was performed using small collections until the TREC 2001 Arabic track became available [5]. Root-based stemming, light stemming, and dictionary-lookup stemming are three different types of stemming [6].…”

Section: Introductionmentioning

confidence: 99%

Stemmer Impact on Quranic Mobile Information Retrieval Performance

Aljaloud¹,

Dahab²,

Kamal³

2016

ijacsa

View full text Add to dashboard Cite

Abstract-Stemming algorithms are employed in information retrieval (IR) to reduce verity variants of the same word with several endings to a standard stem. Stemmers can also help IR systems by unifying vocabulary, reducing term variants, reducing storage space, and increasing the likelihood of matching documents, all of which make stemming very attractive for use in IR. This paper aims to study the impact of using stemming techniques in mobile effectiveness. Two-word extraction stemming techniques will be used: a light stemmer and a dictionary-lookup stemmer. Also, three sets of experiments were conducted in this research in order to raise the efficiency of mobile aapplications. Implementing the two stemming approaches and assessing their accuracy by calculating the precision, recall, MAP, and f-measure, produced results which show that the light10 stemmer outperforms the dictionarylookup stemmer in precision and MAP. Furthermore, the mobile performance of the light10 stemmer exceeds that of the dictionary-based stemmer.

show abstract

“…Figure 2-b, explains this abstracted view, where we check the length of a given word and a pattern of that word, if they are matched, we apply the prefix and suffix removing process, then we check the candidate stem using a root dictionary. The root is extracted after removing the affixes (suffixes, prefixes, and infixes) attached to the given word [8,14].…”

Section: Stemmer Based On Affixes and Patternsmentioning

confidence: 99%

“…Light stemming aims to enhance feature/keyword reduction while keeping the word's meanings unaffected [6,8,9]. Light stemming refers to removing some defined prefixes and suffixes from the word without trying to deal with infixes instead of extracting the original root or recognizing patterns, it is not dictionary driven, so it is not required to have an Arabic word after removing suffixes [3,9,14]. Many algorithms have been developed for this approach [3,6,9].…”

Section: Stemmer Based On Suffix and Prefixmentioning

confidence: 99%

Arabic stemming techniques: Comparisons and new vision

Al-Zyoud

Al-Rabayah

2015

2015 IEEE 8th GCC Conference &Amp; Exhibition

View full text Add to dashboard Cite

Arabic information extraction processes have become a popular area of research. Many methods and approaches have designed and introduced algorithms to solve the problem of morphology and stemming of Arabic language. Each researcher proposed his own standards, testing methodology and accuracy measurements to test his algorithm. Therefore, we cannot make an exact comparison between these algorithms. However, this research goes over stemming processes by explaining and discussing Arabic language characteristics and difficulties of stemming it, comparing root-based stemming, suffix and prefix -based stemming, and translation base stemming against each other, representing a modified stemming algorithm which helps go over some missing words in other algorithms. And finally, representing a new vision of Arabic stemming techniques.

show abstract

Stemming as a feature reduction technique for Arabic Text Categorization

Cited by 42 publications

References 18 publications

Features based approach for indexation and representation of unstructured Arabic documents

Features based approach for indexation and representation of unstructured Arabic documents

Stemmer Impact on Quranic Mobile Information Retrieval Performance

Arabic stemming techniques: Comparisons and new vision

Contact Info

Product

Resources

About