Narrator2Vec: An Efficient Narrator Representation in Hadith Literature Using Word Embedding

Our proposed method, Narrator2Vec, uses a novel approach to examine the narrators involved in transmitting Islamic hadith. This method utilizes a distributed representation of narrators through word embedding, by training a 100-dimensional vector representation based uniquely on the names of hadith narrators found in over 650,000 hadiths. These vectors can effectively capture the relatedness and internal structure among narrators’ names, such as identifying the different names of the same narrator and predicting missing or unknown narrators in incomplete transmission chains. Additionally, Narrator2Vec can be used to detect patterns and relationships within the transmission chains of hadith, providing valuable insights into the historical context and evolution of these important religious texts. Furthermore, it can be applied to other historical texts and research fields, where the identification of key actors and their relationships are important. The ability to quickly and efficiently analyze large amounts of hadith data through Narrator2Vec can greatly enhance the accuracy and speed of hadith studies, making it a valuable tool for scholars and researchers in the field.