Boosting Arabic Named Entity Recognition with K-Fold Cross Validation on LSTM and Bi-LSTM Models

Named-Entity-Recognition(NER) is one of the most important Information-Extraction (IE) use cases, whichis used to improve performance Natural Languages Processing (NLP) tasks,such as Relation-Extraction (RE), Question-Answering (QA). Recently, Arabic NER tackled in differentways by researchers. In this study, we assess two widelyused models, namely, LSTM and Bi-LSTM on task languageand perform a comparative study between these models. contrast thetraditional data partition technique widely during training, employthe k-fold cross-validation eachmodel. The experimental results reveal that all models isimproved when applied. Additionally, according toour experiment results, model outperforms termsof our evaluation metric. We achieve best F1 score 94.17% withCNN-Bi-LSTM-CRF. An ablation demonstrates thatthe increased from 87.28 94.17%.