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

Kasmaiee, Sa.; Kasmaiee, Si.; Homayounpour, M.

doi:10.1038/s41598-023-47295-2

Cited by 15 publications

(4 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Artificial neural networks (ANNs), as introduced by McCulloch and Pitts [69], emerged as a result of studying brain functionality and subsequently found application in computer programs [6,70,71]. In addition, it is important to note that any ANN comprises numerous individual units, commonly referred to as neurons or processing elements (PE).…”

Section: Classification Viaartificial Neural Network (Ann)mentioning

confidence: 99%

See 1 more Smart Citation

A Novel Feature Selection Approach-Based Sampling Theory on Grapevine Images Using Convolutional Neural Networks

Özaltın,

Koyuncu

2024

Arab J Sci Eng

View full text Add to dashboard Cite

Feature selection, reducing number of input variables to develop classification model, is an important process to reduce computational and modeling complexity and affects the performance of image process. In this paper, we have proposed new statistical approaches for feature selection based on sample selection. We have applied our new approaches to grapevine leaves data that possess properties of shape, thickness, featheriness, and slickness that are investigated in images. To analyze such kind of data by using image process, thousands of features are created and selection of features plays important role to predict the outcome properly. In our numerical study, convolutional neural networks have been used as feature extractors and then obtained features from the last average pooling layer to detect the type of grapevine leaves from images. These features have been reduced by using our suggested four statistical methods: simple random sampling, ranked set sampling, extreme ranked set sampling, moving extreme ranked set sampling. Then, selected features have been classified with artificial neural network and we obtained the best accuracy of 97.33% with our proposed approaches. Based on our empirical analysis, it has been determined that the proposed approach exhibits efficacy in the classification of grapevine leaf types. Furthermore, it possesses the potential for integration into various computational devices.

show abstract

Section: Classification Viaartificial Neural Network (Ann)mentioning

confidence: 99%

“…G − Mean Sensitivity × Specificity (5) where TP: True positive, FP: false positive, TN: true negative, and FN: false negative are shown [8,71,74,75].…”

Section: Performance Metricsmentioning

confidence: 99%

A Novel Feature Selection Approach-Based Sampling Theory on Grapevine Images Using Convolutional Neural Networks

Özaltın,

Koyuncu

2024

Arab J Sci Eng

View full text Add to dashboard Cite

show abstract

“…The proposed model consists of a hybrid of CNNs and LSTM. Hybrid CNN and LSTM model is widely used in many fields [29,38]. After setting parameters and preparing the dataset with the designed simulator, features are extracted using CNN and LSTM.…”

Section: Proposed Mobile Molecular Communication (Mmc) Modelmentioning

confidence: 99%

Classification of Diffusion Constants of Transmitter and Receiver and Distance Between Them Using Mobile Molecular Communication via Diffusion Model

Er,

Isik,

Kuran

et al. 2024

Arab J Sci Eng

View full text Add to dashboard Cite

Molecular communication (MC) holds promise for enabling communication in scenarios where traditional wireless methods may be impractical or ineffective, offering unique capabilities for a range of applications in both natural and engineered systems. In this research, a novel approach to MC is explored, diverging from the standard use of stationary transmitter and receiver models typically found in the field. The study introduces a dynamic MC model, where both the transmitter and receiver are mobile within a diffusion environment. This model operates using a 5-bit system. The key finding is that the mobility of these nanodevices alters their distance, which in turn impacts the likelihood of molecule reception at the receiver. The study employs deep learning techniques, specifically a combination of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks, to categorize the mobility patterns of the receiver (Rx) and transmitter (Tx). By analyzing various mobility rates (Drx and Dtx) and distances between the Tx and Rx, the research successfully identifies the most efficient mobile MC model in terms of molecule reception rates. The use of Linear Support Vector Machine alongside the CNN and LSTM hybrid feature vector resulted in an 87.68% accuracy in predicting diffusion coefficients. Moreover, using a Cubic Support Vector with the same hybrid feature vector, the study achieved an 88.09% accuracy in estimating the distance between the transmitter and receiver. The study concludes that an increase in the mobilities of Rx and Tx correlates with a higher rate of molecule reception.

show abstract

“…We propose a self-adaptive speech feature fusion network SSFF-Net that dynamically integrates enhanced features with original noisy features, fully integrating enhanced features and original speech information. In recent years, there has been a number of research into textual spelling error correction, which has used methods including rule-based and deep learning-based methods [12]. Error correction techniques have been widely used to refine the sentences output of ASR models [13]- [14], including error detection and correction.…”

Section: Introductionmentioning

confidence: 99%

NRSC-KG: Noise-Robust Speech Comprehension Based on Knowledge Graph for HRI

Gun,

Deng,

Xie

2024

Preprint

View full text Add to dashboard Cite

Speech comprehension is the key to Human-Robot Interaction (HRI), however, noise energetic masking can greatly reduce the performance of speech comprehension. Simply using Speech Enhancement (SE) methods can easily lead to suboptimal solutions. To alleviate these issues, this paper proposes a Noise-Robust Speech Comprehension system based on Knowledge Graph (NRSC-KG), which designs robust methods in Automatic Speech Recognition (ASR) and ASR error correction. In ASR, we propose a Self-adaptive Speech Feature Fusion Network (SSFF-Net) to dynamically fuse enhanced speech features with the original speech features. In ASR error correction, we propose a divergent apperceive error correction method based on domain-specific Knowledge Graph (KG). The experiment on AISHELL-1 speech corpus shows that compared to the traditional joint training of SE and ASR, our method achieves a relative Character Error Rate (CER) reduction of 12.7%. We validated the KG based error correction method on our customized Herbal Cuisine ordering Dialogue speech dataset (HCDialogue), and the results showed that our method achieved a relative CER reduction of 5.0% and 2.1%, respectively, compared to the language model based and end-to-end neural network based methods. It’s indicated that the KG based method has great potential in improving speech comprehension performance.

show abstract

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

Cited by 15 publications

References 63 publications

A Novel Feature Selection Approach-Based Sampling Theory on Grapevine Images Using Convolutional Neural Networks

A Novel Feature Selection Approach-Based Sampling Theory on Grapevine Images Using Convolutional Neural Networks

Classification of Diffusion Constants of Transmitter and Receiver and Distance Between Them Using Mobile Molecular Communication via Diffusion Model

NRSC-KG: Noise-Robust Speech Comprehension Based on Knowledge Graph for HRI

Contact Info

Product

Resources

About