Correcting spelling mistakes in Persian texts with rules and deep learning methods

This study aims to develop a system for automatically correcting spelling errors in Persian texts using two approaches: one that relies on rules and a common spelling mistake list and another that uses a deep neural network. The list of 700 common misspellings was compiled, and a database of 55,000 common Persian words was used to identify spelling errors in the rule-based approach. 112 rules were implemented for spelling correction, each providing suggested words for misspelled words. 2500 sentences were used for evaluation, with the word with the shortest Levenshtein distance selected for evaluation. In the deep learning approach, a deep encoder-decoder network that utilized long short-term memory (LSTM) with a word embedding layer was used as the base network, with FastText chosen as the word embedding layer. The base network was enhanced by adding convolutional and capsule layers. A database of 1.2 million sentences was created, with 800,000 for training, 200,000 for testing, and 200,000 for evaluation. The results showed that the network's performance with capsule and convolutional layers was similar to that of the base network. The network performed well in evaluation, achieving accuracy, precision, recall, F-measure, and bilingual evaluation understudy (Bleu) scores of 87%, 70%, 89%, 78%, and 84%, respectively.

Cervical Muscle Activation Characteristics and Head Kinematics in Males and Females Following Acoustic Warnings and Impulsive Head Forces

Sex, head and neck posture, and cervical muscle preparation are contributing factors in the severity of head and neck injuries. However, it is unknown how these factors modulate the head kinematics. In this study, twenty-four (16 male and 8 female) participants experienced 50 impulsive forces to their heads with and without an acoustic warning. Female participants demonstrated a 71 ms faster (p = 0.002) muscle activation onset compared to males after warning. The magnitude of muscle activation was not significant between sexes. Females exhibited 21% (p < 0.008) greater peak angular velocity in all force directions and 18% (p < 0.04) greater peak angular acceleration in sagittal plane compared to males. Females exhibited 15% (p = 0.03) greater peak linear acceleration compared to males only in sagittal flexion. Preparation attenuated head kinematics significantly (p < 0.03) in 11 out of 18 investigated head kinematics for both sexes. A warning eliciting a startle response 420 ms prior to the impact resulted in significant attenuation of all measured head kinematics in sagittal extension (p < 0.037). In conclusion, both sex and warning type were significant factors in head kinematics. These data provide insight into the complex relationship of muscle activation and sex, and may help identify innovative strategies to reduce head and neck injury risk in sports.