A comparative study of abstractive and extractive summarization techniques to label subgroups on patent dataset

Souza, Cinthia Mikaela de; Meireles, Magali Rezende Gouvêa; Almeida, Paulo

doi:10.1007/s11192-020-03732-x

Cited by 18 publications

(9 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The basis of computational intelligence can also be understood as the continuous evolution of structures. Since the birth of intelligent computing, its powerful intelligence, robustness, and self-adaptability have attracted the attention of many researchers, and great breakthroughs have been made in both theoretical research and practical applications of algorithms [ 8 ]. Kolomvatsos and Anagnostopoulos proposed a simple two-way short- and long-term memory network intelligent computing IoT algorithm for recommendation classification or nonrecommendation, which predicts the outcome by the average semantic of specific phrases in home education, with an average accuracy of 74% in the final experiment [ 9 ].…”

Section: Current Status Of Researchmentioning

confidence: 99%

“…Words are connected, so the longdistance dependence feature is greatly shortened, and its maximum path length is 1, which can capture long-distance dependence. e text data within a certain sliding window is converted into a fixed-length vector B'' by the Softmax function, and then the feature data is fed into a linear SVM classifier for training and outputting the sentiment classification results, which is calculated by equation (8), where C denotes the training error of the SVM classifier, W and H denote the final obtained weight matrix and bias value, respectively, and L denotes the loss function.…”

Section: F(x)fax(sam(x)) �mentioning

confidence: 99%

See 1 more Smart Citation

A Study of Two‐Way Short‐ and Long‐Term Memory Network Intelligent Computing IoT Model‐Assisted Home Education Attention Mechanism

2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

This paper analyzes and collates the research on traditional homeschooling attention mechanism and homeschooling attention mechanism based on two-way short- and long-term memory network intelligent computing IoT model and finds the superiority of two-way short- and long-term memory network intelligent computing IoT model. The two-way short- and long-term memory network intelligent computing IoT model is improved and an improved deep neural network intelligent computing IoT is proposed, and the improved method is verified based on discrete signal homeschooling classification experiments, followed by focusing on the application research of the two-way short- and long-term memory network intelligent computing IoT model-assisted homeschooling attention mechanism. Learning based on neural network, human behavior recognition method combining spatiotemporal networks, a homeschooling method integrating bidirectional short- and long-term memory networks and attention mechanisms is designed. The visual attention mechanism is used to add weight information to the deep visual features extracted by the convolutional neural network, and a new feature sequence incorporating salient attention weights is output. This feature sequence is then decoded using an IndRNN independent recurrent neural network to finally classify and decide on the homeschooling category. Experiments on the UCF101 dataset demonstrate that the incorporation of the attention mechanism can improve the ability of the network to classify. The attention mechanism can help the intelligent computing IoT model discover key features, and the self-attention mechanism can effectively capture the internal features of homeschooling and optimize the feature vector. We propose the strategy of combining the self-attention mechanism with a bidirectional short- and long-term memory network to solve the family education classification problem and experimentally verify that the intelligent computing IoT model combined with the self-attention mechanism can more easily capture the interdependent features in family education, which can effectively solve the family education problem and further improve the family education classification accuracy.

show abstract

Section: Current Status Of Researchmentioning

confidence: 99%

Section: F(x)fax(sam(x)) �mentioning

confidence: 99%

A Study of Two‐Way Short‐ and Long‐Term Memory Network Intelligent Computing IoT Model‐Assisted Home Education Attention Mechanism

2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…The authors experiment with inputting [the purpose of the present invention is] to retrieve and extract the patent aim. Finally, de Souza et al (2021) have compared extractive and abstractive models in naming patents' subgroups. When used to "summarize" the Abstract to produce a patent Title -which should Model R-1 R-2 R-L TextRank (Mihalcea & Tarau, 2004) 35.99 11.14 29.60 LexRank (Erkan & Radev, 2004) 35.57 10.47 29.03 SumBasic (Nenkova & Vanderwende, 2005) 27.44 7.08 23.66 RNN-ext RL (Chen & Bansal, 2018b) 34.63 10.62 29.43 LSTM seq2seq (Sutskever et al, 2014) + attention 28.74 7.87 24.66 Pointer-Generator (See et al, 2017) 30.59 10.01 25.65 Pointer-Generator + coverage (See et al, 2017) 33.14 11.63 28.55 SentRewriting (Chen & Bansal, 2018a) 37.12 11.87 32.45 TLM (Pilault et al, 2020) 36 For BIGBIRD, results using RoBERTa's (MLM) and a Pegasus' (Gap Sentence Generation) pre-training are considered.…”

Section: Abstractive Modelsmentioning

confidence: 99%

“…In the context of abstractive summarization and patent generation, some works (de Souza et al, 2021;Lee, 2020) highlight that ROUGE is unable to find semantically similar sentences expressed in different wording. In the context of Natural Language Generation, some new measures have recently been proposed to solve these issues.…”

Section: Current and Future Directionsmentioning

confidence: 99%

Summarization, Simplification, and Generation: The Case of Patents

Casola,

Lavelli

2021

Preprint

View full text Add to dashboard Cite

We survey Natural Language Processing (NLP) approaches to summarizing, simplifying, and generating patents' text. While solving these tasks has important practical applications -given patents' centrality in the R&D processpatents' idiosyncrasies open peculiar challenges to the current NLP state of the art. This survey aims at a) describing patents' characteristics and the questions they raise to the current NLP systems, b) critically presenting previous work and its evolution, and c) drawing attention to directions of research in which further work is needed. To the best of our knowledge, this is the first survey of generative approaches in the patent domain.

show abstract

“…These standard taxonomies consist of complex hierarchical structures that cover all technology areas and help maintain inter-operability among various patent offices worldwide (Gomez & Moens, 2014). Therefore, accurate automated classification of patent documents is critical and will help experts manage patent documents, facilitate reliable patent search and retrieval, reduce the risk of missing a relevant patent in preventing patent infringement, and further patent analysis tasks (Yun & Geum, 2020;Souza et al, 2020;Gomez & Moens, 2014).…”

Section: Introductionmentioning

confidence: 99%

PatentNet: multi-label classification of patent documents using deep learning based language understanding

et al. 2021

View full text Add to dashboard Cite

Patent classification is an expensive and time-consuming task that has conventionally been performed by domain experts. However, the increase in the number of filed patents and the complexity of the documents make the classification task challenging. The text used in patent documents is not always written in a way to efficiently convey knowledge. Moreover, patent classification is a multi-label classification task with a large number of labels, which makes the problem even more complicated. Hence, automating this expensive and laborious task is essential for assisting domain experts in managing patent documents, facilitating reliable search, retrieval, and further patent analysis tasks. Transfer learning and pre-trained language models have recently achieved state-of-the-art results in many Natural Language Processing tasks. In this work, we focus on investigating the effect of fine-tuning the pre-trained language models, namely, BERT, XLNet, RoBERTa, and ELECTRA, for the essential task of multi-label patent classification. We compare these models with the baseline deep-learning approaches used for patent classification. We use various word embeddings to enhance the performance of the baseline models. The publicly available USPTO-2M patent classification benchmark and M-patent datasets are used for conducting experiments. We conclude that fine-tuning the pre-trained language models on the patent text improves the multi-label patent classification performance. Our findings indicate that XLNet performs the best and achieves a new state-of-the-art classification performance with respect to precision, recall, F1 measure, as well as coverage error, and LRAP.

show abstract

A comparative study of abstractive and extractive summarization techniques to label subgroups on patent dataset

Cited by 18 publications

References 32 publications

A Study of Two‐Way Short‐ and Long‐Term Memory Network Intelligent Computing IoT Model‐Assisted Home Education Attention Mechanism

A Study of Two‐Way Short‐ and Long‐Term Memory Network Intelligent Computing IoT Model‐Assisted Home Education Attention Mechanism

Summarization, Simplification, and Generation: The Case of Patents

PatentNet: multi-label classification of patent documents using deep learning based language understanding

Contact Info

Product

Resources

About