Statistical Natural Language Generation from Tabular Non-textual Data

Mahapatra, Joy; Naskar, Sudip Kumar; Bandyopadhyay, Sivaji

doi:10.18653/v1/w16-6624

Cited by 16 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Androutsopoulos et al (2013) and Duma and Klein (2013) focused on generating descriptive language for Ontologies and RDF triples. Most recent work utilize neural networks on data-to-text generation (Mahapatra et al, 2016;Wiseman et al, 2017;Kaffee et al, 2018;Freitag and Roy, 2018;Qader et al, 2018;Dou et al, 2018;Yeh et al, 2018;Jhamtani et al, 2018;Liu et al, 2017bLiu et al, , 2019Peng et al, 2019;Dušek et al, 2019). Some closely relevant work also focused on the table-to-text generation.…”

Section: Related Workmentioning

confidence: 99%

Towards Comprehensive Description Generation from Factual Attribute-value Tables

Li¹,

Luo²,

Yang³

et al. 2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

The comprehensive descriptions for factual attribute-value tables, which should be accurate, informative and loyal, can be very helpful for end users to understand the structured data in this form. However previous neural generators might suffer from key attributes missing, less informative and groundless information problems, which impede the generation of high-quality comprehensive descriptions for tables. To relieve these problems, we first propose force attention (FA) method to encourage the generator to pay more attention to the uncovered attributes to avoid potential key attributes missing. Furthermore, we propose reinforcement learning for information richness to generate more informative as well as more loyal descriptions for tables. In our experiments, we utilize the widely used WIKIBIO dataset as a benchmark. Additionally we create WB-filter based on WIKIBIO to test our model in the simulated user-oriented scenarios, in which the generated descriptions should accord with particular user interests. Experimental results show that our model outperforms the state-of-the-art baselines on both automatic and human evaluation.

show abstract

Section: Related Workmentioning

confidence: 99%

Towards Comprehensive Description Generation from Factual Attribute-value Tables

Li¹,

Luo²,

Yang³

et al. 2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

show abstract

“…Penelitian yang berfokus pada pembangkitan representasi tekstual berdasarkan data non tekstual secara otomatis telah banyak dilakukan [2]- [8]. Penelitian tersebut secara umum menggunakan pendekatan Natural Language Generation (NLG) untuk berbagai domain masalah menggunakan berbagai pendekatan dan metode, di antaranya pembangkitan tekstual ramalan cuaca berbasis metode Case Based Reasoning (CBR) dilakukan dalam [2], kajian penerapan pengetahuan geografis dari web untuk membangkitkan deskripsi jaringan sensor hidrologi bagi publik [3], dan pembangkitan teks dari masukan nontekstual (dataset tabular) [4]. Secara khusus untuk bidang kesehatan, pembangkitan ringkasan tekstual dari data medis neonatal di ruang ICU menggunakan pendekatan datato-text dilakukan dalam [5], pembangkitan ringkasan rawat inap pasien dalam [6], pembangkitan interpretasi tekstual berbahasa Indonesia berdasarkan data grafik pengawasan kesehatan satu garis menggunakan metode bigram dalam [7], dan pembangkitan interpretasi berbahasa Indonesia berdasarkan data grafik pengawasan kesehatan dua garis menggunakan metode template dalam [8].…”

Section: Pendahuluanunclassified

Textual interpretation generation in Indonesian language based on blood chemistry tests data using r-template based approach

Aulia

Purnamawati

Junianto

2019

Jurnal Teknologi dan Sistem Komputer

View full text Add to dashboard Cite

The result of the blood chemistry tests is usually presented in the form of a table written in abbreviations, numbers, and units. Unfortunately, the young doctors often require the time and experience for interpreting the blood chemistry tests into a textual representation, which is easy to read and understand. Therefore, this research aimed at developing a system (BTISys) that can generate the textual representation in the Indonesian language automatically based on the blood chemistry test. BTISys generates the representation using Natural Language Generation (NLG) approach based on the r-template method. The reliability of BTISys is measured by considering the naturalness of generated textual representation. The naturalness can be observed by three criteria, such as readability, clarity, and general appropriateness. The reliability of BTISys is quite good to generate the textual representation automatically. It can be seen from the readability, clarity, and general appropriateness, which reach 73 %, 70 %, and 60 % respectively, that implies the naturalness of generated textual representation.

show abstract

“…For reasons discussed below, we here focus on the pipeline architecture common in earlier NLG work (Reiter, 2007;Mahapatra et al, 2016), consisting of document planning, micro planning, and surface realization stages.…”

Section: Related Workmentioning

confidence: 99%

Data-Driven News Generation for Automated Journalism

Leppänen¹,

Munezero²,

Granroth-Wilding³

et al. 2017

Proceedings of the 10th International Conference on Natural Language Generation

View full text Add to dashboard Cite

Despite increasing amounts of data and ever improving natural language generation techniques, work on automated journalism is still relatively scarce. In this paper, we explore the field and challenges associated with building a journalistic natural language generation system. We present a set of requirements that should guide system design, including transparency, accuracy, modifiability and transferability. Guided by the requirements, we present a data-driven architecture for automated journalism that is largely domain and language independent. We illustrate its practical application in the production of news articles upon a user request about the 2017 Finnish municipal elections in three languages, demonstrating the successfulness of the data-driven, modular approach of the design. We then draw some lessons for future automated journalism.

show abstract

Statistical Natural Language Generation from Tabular Non-textual Data

Cited by 16 publications

References 21 publications

Towards Comprehensive Description Generation from Factual Attribute-value Tables

Towards Comprehensive Description Generation from Factual Attribute-value Tables

Textual interpretation generation in Indonesian language based on blood chemistry tests data using r-template based approach

Data-Driven News Generation for Automated Journalism

Contact Info

Product

Resources

About