{"id":4980,"date":"2025-06-02T10:37:21","date_gmt":"2025-06-02T09:37:21","guid":{"rendered":"https:\/\/blog.bham.ac.uk\/bear\/?p=4980"},"modified":"2025-06-13T11:43:49","modified_gmt":"2025-06-13T10:43:49","slug":"extracting-tender-information-with-bluebear-tackling-complex-documents-with-nlp-and-image-processing","status":"publish","type":"post","link":"https:\/\/blog.bham.ac.uk\/bear\/2025\/06\/02\/extracting-tender-information-with-bluebear-tackling-complex-documents-with-nlp-and-image-processing\/","title":{"rendered":"Extracting tender information with BlueBEAR: tackling complex documents with NLP and image processing"},"content":{"rendered":"\n

In this case study we talk to <\/b><\/span>Asim Abbas, PhD Researcher in Artificial Intelligence (School of Computer Science), who is using BlueBEAR to extract structured information from unstructured and ambiguous tender documents<\/strong><\/em>.<\/p>\n\n\n\n

\n

My research focuses on extracting structured information from unstructured and ambiguous tender documents, whereas these documents present distinctive processing challenges due to their complex structure.<\/p>\n<\/div>

\"\"<\/figure><\/div>\n\n\n\n

Unlike standard PDFs, tender documents differ because they contain various elements including unstructured texts, tables, workflows and diagrams and non-standard page sizes which traditional processing tools cannot handle effectively. My system uses Natural Language Processing (NLP) and image processing techniques to transform document pages into images, then apply advanced models for information extraction.<\/p>\n\n\n\n

\n

BlueBEAR enables me to process extensive datasets efficiently through its combination of computational strength, GPU accessibility, and developer-friendly platform<\/p>\n<\/blockquote>\n\n\n\n

The processing demands intense computational power, along with high-performance GPUs and scalable resources, which BlueBEAR has successfully supported. BlueBEAR enables me to process extensive datasets efficiently through its combination of computational strength, GPU accessibility, and developer-friendly platform. The detailed documentation<\/a> combined with easy BlueBEAR setup has enabled me to optimise my workflow, so I can concentrate on research activities.<\/p>\n\n\n\n

Thanks to BlueBEAR, I am able to address these real-world challenges in automating tender document analysis, and I look forward to seeing how this research can contribute to practical applications in industry.<\/p>\n\n\n\n

We were so pleased to hear of how Asim was able to make use of what is on offer from Advanced Research Computing, particularly to hear of how they have made use of the BEAR HPC and GPU cores \u2013 if you have any examples of how it has helped your research then do get in contact with us at bearinfo@contacts.bham.ac.uk<\/a>.<\/em><\/strong><\/p>\n\n\n\n

We are always looking for good examples of use of High Performance Computing to nominate for HPC Wire Awards \u2013 see our recent winner<\/a> for more details<\/em><\/strong>.<\/p>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

In this case study we talk to Asim Abbas, PhD Researcher in Artificial Intelligence (School of Computer Science), who is using BlueBEAR to extract structured information from unstructured and ambiguous tender documents. My research focuses on extracting structured information from unstructured and ambiguous tender documents, whereas these documents present distinctive processing challenges due to their … Continue reading “Extracting tender information with BlueBEAR: tackling complex documents with NLP and image processing”<\/span><\/a><\/p>\n","protected":false},"author":97,"featured_media":4981,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[71,244,91,90],"tags":[87,242,5,53,69,35,243,17],"class_list":["post-4980","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-case-study","category-user-case-study-bear-ai-case-study","category-bluebear","category-rds","tag-ai","tag-artificial-intelligence","tag-bear","tag-bluebear","tag-gpu","tag-hpc","tag-nlp","tag-supercomputing"],"_links":{"self":[{"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/posts\/4980","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/users\/97"}],"replies":[{"embeddable":true,"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/comments?post=4980"}],"version-history":[{"count":6,"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/posts\/4980\/revisions"}],"predecessor-version":[{"id":5045,"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/posts\/4980\/revisions\/5045"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/media\/4981"}],"wp:attachment":[{"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/media?parent=4980"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/categories?post=4980"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blog.bham.ac.uk\/bear\/wp-json\/wp\/v2\/tags?post=4980"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}