Classification of Hadiths using LVQ based on VSM Considering Words Order
| Autor: | Mohamed Ghanem, Mohammed Mourchid, Abdelaaziz Mouloudi |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Root (linguistics)
Learning vector quantization business.industry Arabic Computer science 05 social sciences 0507 social and economic geography Islam 02 engineering and technology Term (logic) computer.software_genre language.human_language Categorization 0202 electrical engineering electronic engineering information engineering Vector space model language 020201 artificial intelligence & image processing Artificial intelligence business tf–idf 050703 geography computer Natural language processing Word order |
| Zdroj: | International Journal of Computer Applications. 148:25-28 |
| ISSN: | 0975-8887 |
| DOI: | 10.5120/ijca2016911077 |
| Popis: | The religion of Islam is based on a sacred text called Qur‟an, a divine speech expressed in Arabic language. Qur‟an constitutes the main root of Islam jurisprudence which has a second source of inspiration known as Hadiths. As the Muslim‟s life is governed by those holy texts, need of their authenticity is required. Using VSM (Vector Space Model), we can represent Hadiths as a vector of words. The Term Weighting obtained by multiplying term frequency by the inverse document frequency does not take into account the word order, however, order of narrators is critical to classify Hadith. In this paper we propose a new method considering the words order (in our case the narrator‟s order), to classify Hadiths into four categories: Sahih, Hasan, Da‟if and Maudu‟. We use in this purpose LVQ (Learning Vector Quantization). We got good results for classifying Sahih and Maudu‟ categories. General Terms Hadith categorization, Algorithms. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|