RECORD
Toward a Period-Specific Optimized Neural Network for OCR Error Correction of Historical Hebrew Texts
Computation and Language · Submitted 30 Jul 2023
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim To overcome these challenges, we developed a new multi-phase method for generating artificial training datasets with OCR errors and hyperparameters optimization for building an effective neural network for OCR post-correction in Hebrew.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Over the past few decades, large archives of paper-based historical documents, such as books and newspapers, have been digitized using the Optical Character Recognition (OCR) technology. Unfortunately, this broadly used technology is error-prone, especially when an OCRed document was written hundreds of years ago. Neural networks have shown great success in solving various text processing tasks, including OCR post-correction. The main disadvantage of using neural networks for historical corpora is the lack of sufficiently large training datasets they require to learn from, especially for morphologically-rich languages like Hebrew. Moreover, it is not clear what are the optimal structure and values of hyperparameters (predefined parameters) of neural networks for OCR error correction in Hebrew due to its unique features. Furthermore, languages change across genres and periods. These changes may affect the accuracy of OCR post-correction neural network models. To overcome these challenges, we developed a new multi-phase method for generating artificial training datasets with OCR errors and hyperparameters optimization for building an effective neural network for OCR post-correction in Hebrew.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Over the past few decades, large archives of paper-based historical documents, such as books and newspapers, have been digitized using the Optical Character Recognition (OCR) technology. | background 1.00 | 0.05 | 0.02 | asserted 1.00 |
| s1 | Unfortunately, this broadly used technology is error-prone, especially when an OCRed document was written hundreds of years ago. | background 0.99 | 0.04 | 0.02 | asserted 1.00 |
| s2 | Neural networks have shown great success in solving various text processing tasks, including OCR post-correction. | background 1.00 | 0.04 | 0.03 | asserted 1.00 |
| s3 | The main disadvantage of using neural networks for historical corpora is the lack of sufficiently large training datasets they require to learn from, especially for morphologically-rich languages like Hebrew. | gap 0.62 | 0.07 | 0.03 | asserted 1.00 |
| s4 | Moreover, it is not clear what are the optimal structure and values of hyperparameters (predefined parameters) of neural networks for OCR error correction in Hebrew due to its unique features. | gap 0.96 | 0.06 | 0.02 | asserted 0.87 |
| s5 | Furthermore, languages change across genres and periods. | background 0.93 | 0.05 | 0.02 | asserted 1.00 |
| s6 | These changes may affect the accuracy of OCR post-correction neural network models. | background 0.82 | 0.06 | 0.03 | hedged possibility 1.00 |
| s7 | To overcome these challenges, we developed a new multi-phase method for generating artificial training datasets with OCR errors and hyperparameters optimization for building an effective neural network for OCR post-correction in Hebrew. | contribution 0.76 | 0.97 | 0.41 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Over the past few decades, large archives of paper-based historical documents, such as books and newspapers, have been digitized using the Optical Character Recognition (OCR) technology. Unfortunately, this broadly used technology is error-prone, especially when an OCRed document was written hundreds of years ago. Neural networks have shown great success in solving various text processing tasks, including OCR post-correction. The main disadvantage of using neural networks for historical corpora is the lack of sufficiently large training datasets they require to learn from, especially for morphologically-rich languages like Hebrew. Moreover, it is not clear what are the optimal structure and values of hyperparameters (predefined parameters) of neural networks for OCR error correction in Hebrew due to its unique features. Furthermore, languages change across genres and periods. These changes may affect the accuracy of OCR post-correction neural network models. To overcome these challenges, we developed a new multi-phase method for generating artificial training datasets with OCR errors and hyperparameters optimization for building an effective neural network for OCR post-correction in Hebrew.
Human review
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.