Summarizing Business News: Evaluating BART, T5, and PEGASUS for Effective Information Extraction
Bibliographic record
Abstract
Nowadays, deep learning models are used to summarize the large volume of text data to understand its intent effectively.Processing huge amounts of data can lead to an Information overload, where the models may generate text summaries that miss out on the important information of actual text content.Such problems in business document summaries can impact progressive business growth.This study employs a dataset comprising business articles sourced from BBC News to conduct an extensive comparative analysis of three prominent text summarization algorithms: Bidirectional and Auto-Regressive Transformers, Text-to-Text Transfer Transformer, and Pre-training with Extracted Gap-sentences for Abstractive Summarization, within the domain of business news summarization.The primary objective is to assess the accuracy of these models in generating concise and coherent summaries, utilizing ROUGE and METEOR scores as the benchmark for evaluation.Each model's proficiency in distilling business narratives while retaining crucial insights is carefully examined.This study analyzes the summaries generated and compares them with the already existing summaries.From the result analysis it observed that BART and PEGASUS show ROUGE-I score of 0.308 and 0.245, and METEOR score 0.28 and 0.25 respectively.The outcomes of this study show that the T5 excelled in the ROUGE-1 and METEOR scores which were 0.354 and 0.35 respectively.outcomes of this research offer significant implications for both researchers and practitioners, equipping them with advanced summarization techniques for extracting information effectively from business-related content.In an age where information overload is prevalent, the findings from this study can guide the selection and deployment of text summarization models to enhance information extraction processes, ultimately facilitating more efficient decision-making and information dissemination in the business domain.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.013 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".