Optimizing the performance of a server-based classification for a large business document flow
The document categorization problem in the case of a large business document flow is considered. Textual and visual embeddings were employed for classification. Textual embeddings were extracted via OCR Tesseract. The Viola and Jones method was applied to generate visual embeddings. This paper descr...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Belarusian National Technical University
2023-02-01
|
| Series: | Системный анализ и прикладная информатика |
| Subjects: | |
| Online Access: | https://sapi.bntu.by/jour/article/view/595 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832543651739205632 |
|---|---|
| author | O. A. Slavin |
| author_facet | O. A. Slavin |
| author_sort | O. A. Slavin |
| collection | DOAJ |
| description | The document categorization problem in the case of a large business document flow is considered. Textual and visual embeddings were employed for classification. Textual embeddings were extracted via OCR Tesseract. The Viola and Jones method was applied to generate visual embeddings. This paper describes the performance optimization technology for the implemented classification algorithm. Servers with Intel CPUs were used for the algorithm execution. For single-threaded implementation, high-level and low-level optimizations were performed. High-level optimization was based on the parametrization of the recognition algorithms and the employment of intermediate data. Low-level optimization was carried out via compiler tools allowing for an extended set of SIMD instructions. The implementation of parallelization with several multithreaded applications on multiple servers was also described. The proposed solution was tested using own test data sets of business documents. The proposed method can be applied in modern information systems to analyze the content of a large flow of digital document images. |
| format | Article |
| id | doaj-art-922c6c5eadff4cc58ffd0029da62d89d |
| institution | Kabale University |
| issn | 2309-4923 2414-0481 |
| language | English |
| publishDate | 2023-02-01 |
| publisher | Belarusian National Technical University |
| record_format | Article |
| series | Системный анализ и прикладная информатика |
| spelling | doaj-art-922c6c5eadff4cc58ffd0029da62d89d2025-02-03T11:37:40ZengBelarusian National Technical UniversityСистемный анализ и прикладная информатика2309-49232414-04812023-02-0104606410.21122/2309-4923-2022-4-60-64444Optimizing the performance of a server-based classification for a large business document flowO. A. Slavin0Federal Research Center “Informatics and Management” of the Russian Academy of Sciences; Smart Engines Service LLCThe document categorization problem in the case of a large business document flow is considered. Textual and visual embeddings were employed for classification. Textual embeddings were extracted via OCR Tesseract. The Viola and Jones method was applied to generate visual embeddings. This paper describes the performance optimization technology for the implemented classification algorithm. Servers with Intel CPUs were used for the algorithm execution. For single-threaded implementation, high-level and low-level optimizations were performed. High-level optimization was based on the parametrization of the recognition algorithms and the employment of intermediate data. Low-level optimization was carried out via compiler tools allowing for an extended set of SIMD instructions. The implementation of parallelization with several multithreaded applications on multiple servers was also described. The proposed solution was tested using own test data sets of business documents. The proposed method can be applied in modern information systems to analyze the content of a large flow of digital document images.https://sapi.bntu.by/jour/article/view/595text analysisdocument recognitiondocument classificationspeedup |
| spellingShingle | O. A. Slavin Optimizing the performance of a server-based classification for a large business document flow Системный анализ и прикладная информатика text analysis document recognition document classification speedup |
| title | Optimizing the performance of a server-based classification for a large business document flow |
| title_full | Optimizing the performance of a server-based classification for a large business document flow |
| title_fullStr | Optimizing the performance of a server-based classification for a large business document flow |
| title_full_unstemmed | Optimizing the performance of a server-based classification for a large business document flow |
| title_short | Optimizing the performance of a server-based classification for a large business document flow |
| title_sort | optimizing the performance of a server based classification for a large business document flow |
| topic | text analysis document recognition document classification speedup |
| url | https://sapi.bntu.by/jour/article/view/595 |
| work_keys_str_mv | AT oaslavin optimizingtheperformanceofaserverbasedclassificationforalargebusinessdocumentflow |