|
| Research | Scope of research | Source of data | Results |
|
| Harrag and Hamdi-Cherif [10] | Classification of Hadiths depending on the query of the user. Vector space model, cosine, TF-IDF are used in the study. | 60 Hadiths | Precision = 66% |
| Recall = 80% |
|
| Ghazizadeh et al. [11] | Classification of Hadith authenticity depending on fuzzy technique | Al-Kafi book, but they did not mention how many Hadiths they used in their study | Accuracy = 94% |
| Harrag and El-Qawasmah [12] | Classification of Hadiths depending on the artificial neural network (ANN) and singular value decomposition (SVD) | 453 Hadiths | F-measures: |
| ANN = 85.75% |
| ANN + SVD = 88.33% |
|
| Bounhas et al. [13] | Naïve Bayes (NB) classifier for narration chain reliability | 1000 Hadiths | F-measure = 89.01% |
| Azmi and Bin Badia [14] | Graph construction for narration series depending on context-free grammar (CFG) and semantic web ontology | 90 Hadiths from Sahih Muslim and Sahih Al-Bukhari | Success rate = 86.7% |
| Azmi and Bin Badia [15], | Graph construction for narration series depending on context-free grammar (CFG) and memory-based learning | 90 Hadiths from Sahih Muslim and Sahih Al-Bukhari | Success rate = 86.7% |
| Azmi and AlBadia [16] |
|
| Aldhaln et al. [17] | Decision tree (DT) approach to classify the authenticity of hadiths | 999 Hadiths from Sahih Al-Bukhari, Sunan Al-Tirmzi, and the compilation of Al-Albani | Accuracy = 97.60% |
| Harrag et al. [18] | Association rule-based approach to build Hadith’s ontology | Sahih Al-Bukhari book, but they did not mention how many Hadiths they used in their study | The study did not mention any results |
| Siddiqui et al. [19] | Graph construction for narration series by the k-nearest neighbor, Naïve Bayes, and decision tree | Sahih Al-Bukhari book and Musnad Ibn Hanbal book, but they did not mention how many Hadiths they used in their study | F-measures: k-nearest neighbor = 85% |
| Naïve Bayes = 80%, |
| Decision tree = 86% |
|
| Azmi and AlOfaidly [20] | Heuristic rule-based approach to classify the authenticity of Hadiths | 752 from Sunan Al-Tirmzi | Success rate: |
| 2,180 Hadiths from Sahih Al-Bukhari |
| Sunan Al-Tirmzi = 93.6% |
| Sahih Al-Bukhari = 99.6% |
|
| Harrag [21] | Information extractor using FST | Sahih Al-Bukhari: complete set of Hadiths | F-measure = 52% |
| Abd Rahman et al. [22] | Recognizing narrators’ names depending on rule-based approach | 150 Hadiths from Sahih Al-Bukhari (in the Malay language) | The study did not mention any results |
| Alhawarat [23] | Extraction of narrators depending on rules and n-grams model | The six books of Hadith | F-measures: n-grams model = 65.11% |
| n-grams and rule-based approach = 70.76 |
|
| Faidi et al. [24] | Combine numerous classifiers and stemmers to compare Hadith classification tools | 795 Hadiths from Sahih Al-Bukhari | The best accuracy is 57.50% (stemmer of Khoja combined with SVM) |
| Balgasem and Zakaria [25] | Using log-likelihood ratio (LLR) with rule-based approach to identifying narrators’ names | 235 Hadiths from Sahih Al-Bukhari | F-measure = 82% |
| Najib et al. [26] | Classification of Hadiths (in the Malay language) using k-nearest neighbor (k-NN), Naïve Bayes (NB), and support vector machine (SVM) | 50 Hadiths from Sunan Al-Tirmzi and 50 Hadiths from Sahih Al-Bukhari (in the Malay language) | Accuracy: K-NN = 62% |
| NB = 81%, |
| SVM = 82%, |
|
| Mahmood et al. [27] | Construction of IR system depending on the conditional random field (CRF) and finite-state transducers (FSTs) | 7,563 Hadiths (in the Urdu language) from Sahih Al-Bukhari | F-measure = 92.41% |
| Sari et al. [28] | Hidden Markov model approach to index the names in Hadiths | 38,102 Hadiths (in Indonesian language) from the books of Tirmzi, Nasai, Malik, Ibnu Majah, Darimi, Ahmad, Al-Bukhari, Muslim, and Abu Dawud | F-measure = 86% |
| Najeeb [4] | Predicting narrators’ names and the other Isnad parts using genetic algorithm (GA) | 3,033 from Sahih Muslim book | Accuracy = 81.44%. |
|
