Inicio  /  Applied Sciences  /  Vol: 11 Par: 12 (2021)  /  Artículo
ARTÍCULO
TITULO

Paragraph Boundary Recognition in Novels for Story Understanding

Riku Iikura    
Makoto Okada and Naoki Mori    

Resumen

The understanding of narrative stories by computer is an important task for their automatic generation. To date, high-performance neural-network technologies such as BERT have been applied to tasks such as the Story Cloze Test and Story Completion. In this study, we focus on the text segmentation of novels into paragraphs, which is an important writing technique for readers to deepen their understanding of the texts. This type of segmentation, which we call ?paragraph boundary recognition?, can be considered to be a binary classification problem in terms of the presence or absence of a boundary, such as a paragraph between target sentences. However, in this case, the data imbalance becomes a bottleneck because the number of paragraphs is generally smaller than the number of sentences. To deal with this problem, we introduced several cost-sensitive loss functions, namely. focal loss, dice loss, and anchor loss, which were robust for imbalanced classification in BERT. In addition, introducing the threshold-moving technique into the model was effective in estimating paragraph boundaries. As a result of the experiment on three newly created datasets, BERT with dice loss and threshold moving obtained a higher F1" role="presentation">??1F1 F 1 than the original BERT had using cross-entropy loss as its loss function (76% to 80%, 50% to 54%, 59% to 63%).

 Artículos similares

       
 
Bowen Xing, Xiao Wang and Zhenchong Liu    
The path planning strategy of deep-sea mining vehicles is an important factor affecting the efficiency of deep-sea mining missions. However, the current traditional path planning algorithms suffer from hose entanglement problems and small coverage in the... ver más

 
Alice Zaghini, Francesca Gagliardi, Valentina Marsili, Filippo Mazzoni, Lorenzo Tirello, Stefano Alvisi and Marco Franchini    
Providing water with adequate quality to users is one of the main concerns for water utilities. In most countries, this is ensured through the introduction of disinfectants, such as chlorine, which are subjected to decay over time, with consequent loss o... ver más
Revista: Water

 
Aaron A. Akin, Gia Nguyen and Aleksey Y. Sheshukov    
Soil erosion by water on agricultural hillslopes leads to numerous environmental problems including reservoir sedimentation, loss of agricultural land, declines in drinking water quality, and requires deep understanding of underlying physical processes f... ver más
Revista: Water

 
Cuong Ngoc Nguyen, Hing-Wah Chau and Nitin Muttil    
Green roofs (GRs) have been researched for decades, yet their implementation remains constrained due to several reasons, including their limited appeal to policymakers and the public. Biochar, a carbon-rich material, has been recently introduced as an am... ver más
Revista: Water

 
Abdullahi T. Sulaiman, Habeeb Bello-Salau, Adeiza J. Onumanyi, Muhammed B. Mu?azu, Emmanuel A. Adedokun, Ahmed T. Salawudeen and Abdulfatai D. Adekale    
The particle swarm optimization (PSO) algorithm is widely used for optimization purposes across various domains, such as in precision agriculture, vehicular ad hoc networks, path planning, and for the assessment of mathematical test functions towards ben... ver más
Revista: Algorithms