Studying the Effect of Syntactic Simplification on Text Summarization

Chatterjee, Niladri; Agarwal, Raksha

doi:10.1080/02564602.2022.2055670

Cited by 7 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After developing the pairwise comparison matrix, a normalized matrix is constructed by dividing each element of the comparison matrix by the sum of its corresponding column (Chatterjee and Agarwal, 2023). The sum of each column in the normalized matrix is 1 (Table 4).…”

Section: Discussionmentioning

confidence: 99%

Prioritizing dark patterns in the e-commerce industry – an empirical investigation using analytic hierarchy process

Singh,

Vishvakarma,

Mal

et al. 2024

MBE

View full text Add to dashboard Cite

Purpose E-commerce companies use different types of dark patterns to manipulate choices and earn higher revenues. This study aims to evaluate and prioritize dark patterns used by e-commerce companies to determine which dark patterns are the most profitable and risky. Design/methodology/approach The analytic hierarchy process (AHP) prioritizes the observed categories of dark patterns based on the literature. Several corporate and academic specialists were consulted to create a comparison matrix to assess the elements of the detected dark pattern types. Findings Economic indicators are the most significant aspect of every business. Consequently, many companies use manipulative methods such as dark patterns to boost their revenue. The study revealed that the revenue generated by the types of dark patterns varies greatly. It was found that exigency, social proof, forced action and sneaking generate the highest revenues, whereas obstruction and misdirection create only marginal revenues for an e-commerce company. Research limitations/implications The limitation of the AHP study is that the rating scale used in the analysis is conceptual. Consequentially, pairwise comparisons may induce bias in the results. Practical implications This paper suggests methodical and operational techniques to choose the priority of dark patterns to drive profits with minimum tradeoffs. The dark pattern ranking technique might be carried out by companies once a year to understand the implications of any new dark patterns used. Originality/value The advantages of understanding the trade-offs of implementing dark patterns are massive. E-commerce companies can optimize their spent time and resources by implementing the most beneficial dark patterns and avoiding the ones that drive marginal profits and annoy consumers.

show abstract

Section: Discussionmentioning

confidence: 99%

Prioritizing dark patterns in the e-commerce industry – an empirical investigation using analytic hierarchy process

Singh,

Vishvakarma,

Mal

et al. 2024

MBE

View full text Add to dashboard Cite

show abstract

“…Although both the statistical and neural network models possess considerable potential for biological text simplification and summarisation, the lack of availability of training corpora continues to pose a challenge for these approaches. Related to biological text simplification, studies have been conducted in relation to the usefulness of dependency parsing [27][28][29]. Junagadh et al proposed bioSimplify, a method that uses dependency-relation classification among nodes to identify noun phrases in a sentence and normalise named entities appearing in a text [20].…”

Section: Introductionmentioning

confidence: 99%

Knowledge-Based Intelligent Text Simplification for Biological Relation Extraction

Gill,

Chetty,

Lim

et al. 2023

Informatics

View full text Add to dashboard Cite

Relation extraction from biological publications plays a pivotal role in accelerating scientific discovery and advancing medical research. While vast amounts of this knowledge is stored within the published literature, extracting it manually from this continually growing volume of documents is becoming increasingly arduous. Recently, attention has been focused towards automatically extracting such knowledge using pre-trained Large Language Models (LLM) and deep-learning algorithms for automated relation extraction. However, the complex syntactic structure of biological sentences, with nested entities and domain-specific terminology, and insufficient annotated training corpora, poses major challenges in accurately capturing entity relationships from the unstructured data. To address these issues, in this paper, we propose a Knowledge-based Intelligent Text Simplification (KITS) approach focused on the accurate extraction of biological relations. KITS is able to precisely and accurately capture the relational context among various binary relations within the sentence, alongside preventing any potential changes in meaning for those sentences being simplified by KITS. The experiments show that the proposed technique, using well-known performance metrics, resulted in a 21% increase in precision, with only 25% of sentences simplified in the Learning Language in Logic (LLL) dataset. Combining the proposed method with BioBERT, the popular pre-trained LLM was able to outperform other state-of-the-art methods.

show abstract

“…TS can increase the accessibility of information to a wider audience, including youngsters, those with little literacy, people who are not native speakers, the elderly, and people with disabilities (Inui et al, 2003;Petersen and Ostendorf, 2007;De Belder and Moens, 2010;Suominen et al, 2013). Additionally, numerous studies have also demonstrated that TS can support other NLP tasks as a preprocessing step (Chen et al, 2012;Chatterjee and Agarwal, 2022).…”

Section: Introductionmentioning

confidence: 99%

Improving Text Simplification with Factuality Error Detection

Ma,

Seneviratne,

Daskalaki

2022

Proceedings of the Workshop on Text Simplification, Accessibility, and Readability (TSAR-2022)

View full text Add to dashboard Cite

In the past few years, the field of text simplification has been dominated by supervised learning approaches thanks to the appearance of large parallel datasets such as Wikilarge and Newsela. However, these datasets suffer from sentence pairs with factuality errors which compromise the models' performance. In this study we proposed a model-independent factuality error detection mechanism, considering bad simplification and bad alignment, to refine the Wikilarge dataset through reducing the weight of these samples during training. We demonstrated that this approach improved the performance of the state-of-the-art text simplification model TST5 by an FKGL reduction of 0.33 and 0.29 on the TurkCorpus and ASSET testing datasets respectively. Our study illustrates the impact of erroneous samples in TS datasets and highlights the need for automatic methods to improve their quality.

show abstract

Studying the Effect of Syntactic Simplification on Text Summarization

Cited by 7 publications

References 27 publications

Prioritizing dark patterns in the e-commerce industry – an empirical investigation using analytic hierarchy process

Prioritizing dark patterns in the e-commerce industry – an empirical investigation using analytic hierarchy process

Knowledge-Based Intelligent Text Simplification for Biological Relation Extraction

Improving Text Simplification with Factuality Error Detection

Contact Info

Product

Resources

About