检索结果-内蒙古大学图书馆

Advanced text summarization Model Incorporating NLP Techniques and Feature-Based Scoring

IEEE ACCESS 2025年 13卷 19302-19319页

作者： Kadhim, Estabraq Abdulreda Feizi-Derakhshi, Mohammad-Reza Aghdasi, Hadi S. Univ Tabriz Dept Comp Engn Computerized Intelligence Syst Lab Tabriz *** Iran Islamic Azad Univ Tabriz Dept Comp Engn Tabriz *** Iran

The most common traditional approaches to summarizing large texts while retaining their importance are TF-IDF and textRank. However, these methods often fail to retain narrative coherence and accuracy. This study's improved summarization methodology overcomes these limitations by combining the linguistic and semantic resources. Moreover, although it is more computationally complex, it efficiently combines higher quality with faster summarization. Specifically, a method relies on a weighted feature score scheme. For example, various textual features such as Named Entity Counts, Noun Counts, and Sentence Position contribute to the summarization quality appropriately. This study's summarization algorithm was tested using the CNN, XSum and BBC summarization datasets, which aggregate documents from different areas. The methodology was checked against traditional methods using ROUGE-1 and ROUGE-2, ROUGE-L and BERTScore. The last one, BERTScore, evaluates the semantic similarity of the generated summaries and the references. This study shows that the proposed methodology generates summaries that are not only informative but even semantically faithfully reproduce the original textual information;it achieves high scores in terms of F1-measure across different evaluations like BERTSCORE (0.8857) and ROUGE-1(0.6388), ROUGE-2(0.5662) and ROUGE-L (0.6421). It thus suggests that the approach is applicable in real-life situations and deserves further research.

关键词： Semantics Measurement Feature extraction Data mining text summarization Computational modeling Coherence Encoding Computational efficiency Bidirectional control Extractive text summarization sentences scoring weighted features BBC-news dataset rouge metric BERTScore metric

来源：评论

学校读者我要写书评

暂无评论

Current Trends and Advances in Extractive text summarization: A Comprehensive Review

引用

IEEE ACCESS 2025年 13卷 28150-28166页

作者： Azam, Maryam Khalid, Shah Almutairi, Sulaiman Ali Khattak, Hasan Namoun, Abdallah Ali, Amjad Syed Muhammad Bilal, Hafiz Natl Univ Sci & Technol NUST Sch Elect Engn & Comp Sci Islamabad 44000 Pakistan Qassim Univ Coll Publ Hlth & Hlth Informat Dept Hlth Informat Buraydah 51452 Qassim Saudi Arabia Australian Catholic Univ Peter Faber Business Sch North Sydney NSW 2060 Australia Islamic Univ Madinah Fac Comp & Informat Syst AI Ctr Madinah 42351 Saudi Arabia Univ Swat Dept Comp & Software Technol Khyber Pakhtunkhwa 19130 Pakistan

Given the rapid increase of textual data in various fields, text summarization has become essential for efficient information handling. Over recent decades, numerous methods have been proposed to enhance summarization processes, and various review papers and books have been published to encapsulate these methodologies and discuss their implications. However, existing reviews often fail to provide a comprehensive retrospective of recent advancements, particularly concerning detailed architectural frameworks, the field's current state, evaluation methodologies, and unresolved challenges. This paper addresses this gap by presenting a detailed analysis of the extractive approaches, encompassing their inherent strengths, limitations, and underlying mechanisms. We present a detailed, multi-layered architectural framework designed to advance and develop summarization models, thereby supporting researchers in their endeavors. The text summarization framework consists mainly of text preprocessing, feature extraction, sentence scoring, use of a base model, sentence selection and output summary, and post-processing. Furthermore, this review of 145 research articles categorizes domain-specific summarization techniques, focusing on unique challenges and tailored strategies for news, scientific articles, and social media. These techniques include statistical, fuzzy logic, rule, optimization, graph, clustering-based, machine learning, and deep learning. We emphasize the impact of evaluation metrics and benchmark datasets in performance assessment, providing a detailed analysis of the commonly utilized datasets and metrics (mainly ROUGE-1, ROUGE-2, ROUGE-L, and ROUGE-S) in the current literature. This review article is a valuable resource for advancing text summarization techniques in natural language processing and machine learning by identifying future research directions and open challenges. Notable challenges include expanding summarization for complex tasks, multiple documents

关键词： text summarization Reviews Market research Data mining Transformers Semantics Feature extraction Surveys Social networking (online) Marine vehicles Survey text summarization transformer-based models domain-specific summarization generic architecture datasets and evaluation measures

来源：评论

学校读者我要写书评

暂无评论

Next-Generation text summarization: A T5-LSTM FusionNet Hybrid Approach for Psychological Data

引用

IEEE ACCESS 2025年 13卷 37557-37571页

作者： Khan, Bilal Usman, Muhammad Khan, Inayat Khan, Jawad Hussain, Dildar Gu, Yeong Hyeon Univ Engn & Technol Dept Comp Software Engn Mardan 23200 Pakistan Univ Engn & Technol Dept Comp Sci Mardan 23200 Pakistan Gachon Univ Sch Comp Seongnam Si 13120 South Korea Sejong Univ Dept Artificial Intelligence & Data Sci Seoul 05006 South Korea

Automatic text summarization (ATS) has developed as a vital method for compressing massive amounts of textual content into concise and useful summaries, to retrieve more effective and useful information. ATS reduces textual statistics into coherent and shorter versions especially focusing on psychological text summarization to extract insights and emotional states, assisting in better analysis and understanding of psychological contents. In this context, this study proposes a new hybrid model T5-LSTM FusionNet, to enhance textual content summarization in the field of psychology. The motivation derives from the developing extent and accessibility of psychological literature online, which necessitates exceptional techniques for extracting significant findings quickly and reliably. The recommended T5-LSTM FusionNet model combines the benefits of text-to-text Transfer Transformer (T5) and Long Short-Term Memory (LSTM). The dataset with 5480 records, accumulated from numerous psychology-associated websites, has been used to verify its performance. T5-LSTM FusionNet's overall performance is evaluated against several latest models along with T5, LSTM, BERT, and DistilBERT. Measures such as accuracy, precision, recall, F1-score, and ROUGE rankings are used to evaluate the model's exceptional summarization. With T5-LSTM FusionNet accomplishing a precision of 0.72, recall of 0.72, F1-score of 0.71, and accuracy of 0.74, the effects show significant improvement over individual models like T5 and LSTM, as well as competitive models like BERT and DistilBERT, in terms of summarization effectiveness and accuracy. Furthermore, T5-LSTM FusionNet plays thoroughly in catching both unigram and bigram overlaps concerning summaries, as proven through a comparison study using ROUGE metrics. This suggests that T5-LSTM FusionNet can retain sequence integrity and relevance in summarizing tasks. This work advances ATS techniques in psychology by presenting a hybrid model that combines sequent

关键词： Psychology text summarization Long short term memory Accuracy Transformers Feature extraction Computational modeling Analytical models Mental health Law Automatic text summarization T5-LSTM FusionNet psychology natural language processing machine learning

来源：评论

学校读者我要写书评

暂无评论

text summarization for Light Weight Devices

Text Summarization for Light Weight Devices

引用

2025 IEEE International Students' Conference on Electrical, Electronics and Computer Science, SCEECS 2025

作者： Akshatha, K. Praveen, M. Taleki, Shravan Rakshith Yadav, K V Nataraj Revathi, T. Alliance University Department of Computer Science and Engineering Bangalore India

ISBN: (纸本)9798331529833

In this paper, we propose a text summarization method for low-compute devices with limited hardware capabilities like smartphones, tablets and IOT devices. Traditional methods for text summarization face significant challenges in the settings where these devices are used, given their limited computational power and memory. We present optimization algorithms that target reducing computation and memory effort whilst retaining quality in the summarization. Our solution focuses on lightweight, efficient methods to runtime-compress the summaries generated by a powerful but heavy model -in order to allow them to be computed with high mobility and little energy consumption by small devices- for example through techniques of model compression, quantization or similarity detection in datasets- thus achieving tractable abstraction of useful information from digital content [2], enhancing their accessibility experience while taking advantage of relevant mobile behaviors. Our approach enables efficient summarization with high performance and minimal resource costs, as shown by the corresponding experimental results. © 2025 IEEE.

关键词： Edge Computing Lightweight Devices Model Compression Natural Language Processing Quantization text summarization

来源：评论

学校读者我要写书评

暂无评论

Multilingual text summarization in Healthcare Using Pre-Trained Transformer-Based Language Models

引用

Computers, Materials & Continua 2025年第4期83卷 201-217页

作者： Josua Käser Thomas Nagy Patrick Stirnemann Thomas Hanne School of Business University of Applied Sciences and Arts Northwestern SwitzerlandOlten4600Switzerland Institute for Information Systems University of Applied Sciences and Arts Northwestern SwitzerlandOlten4600Switzerland

We analyze the suitability of existing pre-trained transformer-based language models(PLMs)for abstractive text summarization on German technical healthcare *** study focuses on the multilingual capabilities of these models and their ability to perform the task of abstractive text summarization in the healthcare *** research hypothesis was that large language models could perform high-quality abstractive text summarization on German technical healthcare texts,even if the model is not specifically trained in that *** experiments,the research questions explore the performance of transformer language models in dealing with complex syntax constructs,the difference in performance between models trained in English and German,and the impact of translating the source text to English before conducting the *** conducted an evaluation of four PLMs(GPT-3,a translation-based approach also utilizing GPT-3,a German language Model,and a domain-specific bio-medical model approach).The evaluation considered the informativeness using 3 types of metrics based on Recall-Oriented Understudy for Gisting Evaluation(ROUGE)and the quality of results which is manually evaluated considering 5 *** results show that text summarization models could be used in the German healthcare domain and that domain-independent language models achieved the best *** study proves that text summarization models can simplify the search for pre-existing German knowledge in various domains.

关键词： text summarization pre-trained transformer-based language models large language models technical healthcare texts natural language processing

来源：评论

学校读者我要写书评

暂无评论

Towards Dataset-scale and Feature-oriented Evaluation of text summarization in Large Language Model Prompts

引用

IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS 2025年第1期31卷 481-491页

作者： Lee, Sam Yu-Te Bahukhandi, Aryaman Liu, Dongyu Ma, Kwan-Liu Univ Calif Davis Davis CA 95616 USA

Recent advancements in Large Language Models (LLMs) and Prompt Engineering have made chatbot customization more accessible, significantly reducing barriers to tasks that previously required programming skills. However, prompt evaluation, especially at the dataset scale, remains complex due to the need to assess prompts across thousands of test instances within a dataset. Our study, based on a comprehensive literature review and pilot study, summarized five critical challenges in prompt evaluation. In response, we introduce a feature-oriented workflow for systematic prompt evaluation. In the context of text summarization, our workflow advocates evaluation with summary characteristics (feature metrics) such as complexity, formality, or naturalness, instead of using traditional quality metrics like ROUGE. This design choice enables a more user-friendly evaluation of prompts, as it guides users in sorting through the ambiguity inherent in natural language. To support this workflow, we introduce Awesum, a visual analytics system that facilitates identifying optimal prompt refinements for text summarization through interactive visualizations, featuring a novel Prompt Comparator design that employs a BubbleSet-inspired design enhanced by dimensionality reduction techniques. We evaluate the effectiveness and general applicability of the system with practitioners from various domains and found that (1) our design helps overcome the learning curve for non-technical people to conduct a systematic evaluation of summarization prompts, and (2) our feature-oriented workflow has the potential to generalize to other NLG and image-generation tasks. For future works, we advocate moving towards feature-oriented evaluation of LLM prompts and discuss unsolved challenges in terms of human-agent interaction.

关键词： Visual analytics prompt engineering text summarization human-computer interaction dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

LowEST: a low resource semantic text summarization method for big data

引用

INNOVATIONS IN SYSTEMS AND SOFTWARE ENGINEERING 2025年第1期21卷 207-214页

作者： Das, Sufal North Eastern Hill Univ Shillong India

In today's world, a large collection of information is available in web due to the rapid growth in technological development. The Web platform stores huge amount of documents and is multiplying every year. The data/document that is available in the web is mostly unstructured or semi-structured. Processing of this unstructured or semi-structured data and retrieving new information from this web data is a challenge. text summarization method can be considered as major component in analysis of large collection of web data. It can be considered as dynamic text mining as well as production of knowledge. Semantic text summarization provides new direction to tackle big data where data is huge volume, dynamic/online in nature and heterogeneous. Here, a novel method is proposed for semantic analysis of web-based big text data with semantic-driven keywords frequent pattern tree to overcome the above-mentioned limitation. This proposed method considers extraction summarization approach to perform automatic text summarization. This paper represents semantic text summarization from web-based large data with the help of concept tree as hybridization of FP-tree (Frequent Pattern tree) and semantic ontology.

关键词： text summarization Semantic analysis Extraction summarization Ontology Concept tree

来源：评论

学校读者我要写书评

暂无评论

Enhancing E-Recruitment Recommendations Through text summarization Techniques

引用

INFORMATION 2025年第4期16卷 333-333页

作者： El-Deeb, Reham Hesham Abdelmoez, Walid El-Bendary, Nashwa Arab Acad Sci Technol & Maritime Transport Coll Comp & Informat Technol POB 1029 Alexandria Egypt Arab Acad Sci Technol & Maritime Transport Coll Comp & Informat Technol POB 11 Aswan Egypt

This research aims to enhance e-recruitment systems using text summarization techniques and pretrained large language models (LLMs). A job recommender system is built with integrated text summarization. The text summarization techniques that are selected are BART, T5 (text-to-text Transfer Transformer), BERT, and Pegasus. Content-based recommendation is the model chosen to be implemented. The LinkedIn Job Postings dataset is used. The evaluation of the text summarization techniques is performed using ROUGE-1, ROUGE-2, and ROUGE-L. The results of this approach deduce that the recommendation does improve after text summarization. BERT outperforms other summarization techniques. Recommendation evaluations show that, for MRR, BERT performs 256.44% better, indicating relevant recommendations at the top more effectively. For RMSE, there is a 29.29% boost, indicating recommendations closer to the actual values. For MAP, a 106.46% enhancement is achieved, presenting the highest precision in recommendations. Lastly, for NDCG, there is an 83.94% increase, signifying that the most relevant recommendations are ranked higher.

关键词： information retrieval recommender systems artificial intelligence natural language processing pretrained large language models text summarization transformers

来源：评论

学校读者我要写书评

暂无评论

A Generative Adversarial Network-Based Extractive text summarization Using Transductive and Reinforcement Learning

引用

IEEE ACCESS 2025年 13卷 65490-65509页

作者： Yang, Jing Qin, Haoshen Sun, Yiping Wang, Hao Ayub Khan, Abdullah Yee Por, Lip Alizadehsani, Roohallah Plawiak, Pawel Univ Malaya Fac Comp Sci & Informat Technol Dept Comp Syst & Technol Kuala Lumpur 50603 Malaysia Univ Florida Herbert Wertheim Coll Engn Dept Comp & Informat Sci & Engn Gainesville FL 32611 USA Shanghai Jiao Tong Univ Sch Elect Informat & Elect Engn Shanghai 200240 Peoples R China Carnegie Mellon Univ Sch Comp Sci Pittsburgh PA 15213 USA Bahria Univ Dept Comp Sci Karachi Campus Karachi 75260 Pakistan Deakin Univ Inst Intelligent Syst Res & Innovat IISRI Waurn Ponds Vic 3216 Australia Cracow Univ Technol Fac Comp Sci & Telecommun Dept Comp Sci PL-31155 Krakow Poland Polish Acad Sci Inst Theoret & Appl Informat PL-44100 Gliwice Poland

text summarization is crucial in various sectors, such as engineering and healthcare, because it enhances efficiency in terms of time and costs. Current extractive text summarization methods struggle with challenges such as greedy selection, model generalization limitations, and high computational demands. To solve these problems, this research introduces a novel extractive text summarization method that uniquely integrates a Generative Adversarial Network (GAN), Transductive Long Short-Term Memory (TLSTM), and DistilBERT for sentence embedding. Our technique uses GANs, which include generator and discriminator components, with the core design based on TLSTM. TLSTM utilizes transductive learning to improve accuracy by focusing on samples geographically closer to the test data. In our model, the generator considers whether to include a sentence in the summary while the discriminator critically reviews the generated summary. This GAN model reduces greedy sentence selection, enhancing summary coherence and quality. We implement a Reinforcement Learning (RL)-based strategy to address an imbalance caused by more fake than real samples in the discriminator. This RL approach, novel in the context of GANs for summarization, views training as a sequence of interconnected decisions, treating each sample as a unique scenario. The network, acting as the decision-making agent, assigns greater rewards or penalties to the minority class to correct the imbalance. The effectiveness of our model was evaluated using the well-regarded CNN/Daily Mail dataset, achieving ROUGE-1, ROUGE-2, and ROUGE-L scores of 52.45, 26.46, and 44.85, respectively. Compared to existing methods, our results demonstrate a significant improvement in summarization quality and operational efficiency, as measured by the ROUGE metric.

关键词： text summarization Generators Long short term memory Encoding Training Reinforcement learning Feature extraction Computer science Bidirectional control Accuracy transductive learning generative adversarial network reinforcement learning DistilBERT

来源：评论

学校读者我要写书评

暂无评论

TiLTS:Tibetan Long text summarization Dataset 13th

TiLTS:Tibetan Long Text Summarization Dataset

引用

13th International Conference on Natural Language Processing and Chinese Computing

作者： Hao, Yanrong Chen, Bo Zhou, Xiaobing Minzu Univ China Sch Informat Engn Beijing Peoples R China Natl Language Resources Monitoring & Res Ctr Mino Beijing Peoples R China

ISBN: (纸本)9789819794393;9789819794409

High-quality datasets are crucial for advancing research in automatic text summarization. At present, summarization models for resource-rich languages like Chinese and English have made significant progress. However, for low-resource languages such as Tibetan, the lack of large-scale publicly available summarization datasets means that related research is still in its early stages. To address this gap, this paper constructs an open Tibetan summarization dataset, TiLTS. By collecting extensive texts from Tibetan news websites and leveraging resources from other languages, it obtains 36,507 (document, summary) pairs. Compared to the only publicly available Tibetan summarization dataset, TiSUM, TiLTS has clear advantages in both data size and challenge. This paper also conducts experiments and analyses using several summarization algorithms on this dataset.

关键词： text summarization Dataset Tibetan Low-resource

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：