SaaS Platform for Context-Aware PDF Summarization using Generative NLP

  This paper presented the design, architecture, and implementation of Brevity.AI, a robust, cloud-native Software as a Service (SaaS) platform engineered to solve the computational and cognitive bottlenecks associated with long-form document comprehension. By transitioning away from purely statistical data-mining and standalone commercial APIs, this project successfully demonstrated the efficacy of a hybrid NLP pipeline.   The integration of an algorithmic Term Frequency-Inverse Document Frequency (TF-IDF) filtering layer alongside an open-source generative Transformer model (BART) proved highly successful in mitigating the strict token-limit constraints typical of deep learning models. This hybrid approach, orchestrated via a high-performance FastAPI backend and supported by scalable AWS cloud storage, ensured that complex multi-page PDFs—including those requiring Optical Character Recognition (OCR)—were summarized with high contextual accuracy and grammatical fluency. Furthermore, the inclusion of multi-modal outputs, specifically regional language translation and Text-to-Speech (TTS) capabilities, significantly enhanced the platform's accessibility. Ultimately, Brevity.AI provides an optimal, self-hosted solution for professionals and academics seeking rapid, context-aware information retrieval.

[1]  I. Awasthi, S. S. Anand, K. Gupta, P. S. Bhogal, and P. K. Soni, "Natural Language Processing (NLP) based Text Summarization A Survey," in 2021 6th International Conference on Inventive Computation Technologies (ICICT), Coimbatore, India, IEEE, 2021, pp. 1310-1317.   [2]  M. Ulker and A. B. Ozer, "Abstractive Summarization Model for Summarizing Scientific Article," IEEE Access, vol. 12, pp. 91252-91262, Jun. 2024.   [3]  A. Ramprasad and P. Sivakumar, "Context-Aware Summarization for PDF Documents using Large Language Models," in 2024 International Conference on Expert Clouds and Applications (ICOECA), Coimbatore, India, IEEE, 2024, pp. 186-191.   [4]  C. Ma et al., "An Iterative Optimizing Framework for Radiology Report Summarization with ChatGPT," IEEE Transactions on Artificial Intelligence, vol. 5, no. 5, pp. 4163-4175, May 2024.   [5]  N. Moratanch and S. Chitrakala, "A Survey on Extractive Text Summarization," in 2017 International Conference on Computer, Communication and Signal Processing (ICCCSP), Chennai, India, IEEE, 2017, pp. 1-6.   [6]  M. F. Mridha, A. A. Lima, K. Nur, S. C. Das, M. Hasan, and M. M. Kabir, "A Survey of Automatic Text Summarization: Progress, Process and Challenges," IEEE Access, vol. 9, pp. 156043-156070, Dec. 2021.   [7]  Y. Kumar, K. Kaur, and S. Kaur, "Study of automatic text summarization approaches in different languages," Artificial Intelligence Review, vol. 54, no. 8, pp. 5897-5929, Dec. 2021.   [8]  S. Munnangi, "Revolutionizing Document Workflows with AI-Powered IDP in Pega," International Journal of Intelligent Systems and Applications in Engineering, vol. 11, no. 11s, pp. 570-580, Dec. 2023.   [9]  M. Lewis et al., "BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension," in Proceedings of the 8th Annual Meeting of the Association for Computational Linguistics (ACL), Online, 2020, pp. 7871–7880.   [10] S. Tiwary and R. Kumar, "Asynchronous Performance Evaluation of FastAPI Framework for Deploying Deep Learning Models," International Journal of Computer Applications, vol. 185, no. 22, pp. 14-19, Apr. 2023.