Automatic text summtarization technology is a crucial component in the field of natural language processing, utilized to address the demands of processing extensive textual data by effectively extracting key informati...
详细信息
ISBN:
(纸本)9798400708305
Automatic text summtarization technology is a crucial component in the field of natural language processing, utilized to address the demands of processing extensive textual data by effectively extracting key information to enhance task efficiency. With the rise of pretrained language models, abstract text summarization has progressively become mainstream, producing fluent srummaries that encapsulate core content. Nonetheless, abstract text summarization unavoidably faces problems of inconsistency with the original text. This paper introduces a sequence tagging task to achieve multi-task learning for abstract text summarization models. In this sequence tagging task, we meticulously designed annotated datasets at both entity and sentence levels based on an analysis of the XSum dataset, aiming to enhance the factual consistency of generated summaries. Experimental results demonstrate that the optimized BART model yields favorable performance in terms of ROUGE and FactCC metrics.
textsummarization is a classic sequence-to-sequence natural language generation task. In order to improve the quality of unsupervised abstract text summarization in unsupervised mode, we propose two constraints for t...
详细信息
textsummarization is a classic sequence-to-sequence natural language generation task. In order to improve the quality of unsupervised abstract text summarization in unsupervised mode, we propose two constraints for training textsummarization model, embedding space constraint and information ratio constraint. We construct a generative adversarial network with two discriminators based on these two constraints (TC-SUM-GAN). We use unsupervised and supervised methods to train the model in the experiment. Experimental results show that the ROUGE-1 value of the unsupervised TC-SUM-GAN increases by 12.57 points compared with the basic model and at least 1.96 points compared with other comparative models. The ROUGE scores of the supervised TC-SUM-GAN are also improved. TC-SUM-GAN achieves very competitive results for the metrics of ROUGE-1 and ROUGE-2. In addition, the abstracts generated by our model are closer to those generated manually.
In the face of escalating volumes of textual data, the demand for advanced systems to efficiently manage this deluge is evident. Automatic summarization is a pivotal solution, continually evolving to meet the burgeoni...
详细信息
In the face of escalating volumes of textual data, the demand for advanced systems to efficiently manage this deluge is evident. Automatic summarization is a pivotal solution, continually evolving to meet the burgeoning data needs and user expectations. This paper delves into abstract text summarization, particularly focusing on utilizing neural networks-a recent breakthrough in the field. Through a comprehensive analysis of various models, we meticulously dissect critical components such as encoder-decoder design, mechanisms, training methodologies, dataset considerations, and evaluation metrics. Our primary objective is to illuminate current model landscapes, pinpoint existing challenges, and propose potential remedies. Notably, transformer-based encoder-decoder architectures emerge as state-of-the-art solutions. We suggest combining neural networks with pre-trained language models to improve abstractive summarization, making advanced techniques more accessible and effective. This approach aims to bolster the efficiency and efficacy of summarization systems, thereby facilitating superior information extraction and comprehension for users grappling with extensive textual datasets.
In the realm of Natural Language Processing (NLP), abstract text summarization (ATS) holds a crucial position, involving the transformation of lengthy textual content into concise summaries while retaining essential i...
详细信息
Nowadays, the Internet is a structure that people can access easily and at the same time produce content easily and without control. In parallel with this situation, the ability of extract information from the raw dat...
详细信息
Nowadays, the Internet is a structure that people can access easily and at the same time produce content easily and without control. In parallel with this situation, the ability of extract information from the raw data that makes up big data has become more complex. The fact that the headlines of the contents contain uncontrolled and misleading elements makes it difficult to reach the right information. The headlines of the contents are important for people to reach the information they want in their limited time. In this study, it is aimed to produce headlines suitable for the content instead of headlines that may be misleading for news. For this purpose, an application that produces headlines for Turkish news with deep learning method has been developed. SuDer news corpus is used as dataset. For the training of the model, it is aimed to obtain more humanoid results in the production of news headlines by using the Transformer architecture, which is frequently preferred in natural language studies today and the abstractsummarization method. In this study, in order to compare the performance of the Transformer model, models are prepared and trained with Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) architectures. At the end of 25 epochs of training with LSTM, GRU and Transformer architectures on the corpus, the values of loss are 1.03, 0.55 and 2.49 respectively. In the experiments performed on the validation data, measurements are made with ROUGE-1, ROUGE-2 and ROUGE-L metrics. As a result of the measurements, it is observed that the Transformer architecture is partially good, based on the metric values produced. In addition, when the headlines produced with these architectures are examined, it is observed that the headline obtained with the Transformer architecture produce headlines that are partially more suitable for the news content compared to other architectures.
暂无评论