Document Planning & Natural Language Generation

Notes:

Automated planning is a sub-area of artificial intelligence (AI) that deals with the development of algorithms and systems that can automatically generate plans or courses of action to achieve a specific goal. Automated planning is related to automated reasoning, which involves the use of logical and computational techniques to solve problems and make decisions.

Document planning, also known as content selection, is a sub-area of automated planning that involves deciding what content should be included in a document and how it should be structured. Document planning can be seen as a form of linear theoretical proof search, in which the system generates a plan for selecting and organizing the content of the document based on a set of input data.

Macro-planning is a higher-level form of document planning that involves not only selecting the content for a document, but also structuring it in a way that conveys the desired information effectively. Macro-planning systems typically take a high-level representation of input data as input and generate a plan for selecting and organizing the final content of the document based on that input.

Natural language generation (NLG) is a subfield of artificial intelligence (AI) that focuses on the development of algorithms and systems that can automatically generate human-like text. NLG is often used to automate the creation of documents and other outputs, such as reports and presentations, by generating text based on a set of input data and a set of rules or guidelines.

In the context of document planning, NLG can be used to automate the process of selecting and structuring the content of a document. For example, an NLG system might be trained on a set of input data and a set of rules or guidelines for generating a specific type of document, such as a financial report. The NLG system would then be able to automatically generate the text for the document based on the input data and the rules or guidelines.

NLG can be used in combination with other AI techniques, such as natural language processing (NLP) and automated planning, to create more complex and sophisticated systems for automating the creation of documents and other outputs.

AI planning refers to the use of artificial intelligence (AI) techniques and algorithms to generate plans or courses of action to achieve a specific goal.
AI planning algorithms are specific algorithms or methods that are used to generate plans using AI techniques.
Automated planning refers to the development of algorithms and systems that can automatically generate plans or courses of action to achieve a specific goal. Automated planning systems are software systems that use automated planning algorithms to generate plans automatically.
Automated planning systems are related to the use of AI techniques to generate plans or courses of action to achieve specific goals. These techniques and systems are used in a wide range of applications, including robotics, manufacturing, and logistics, to name just a few examples.
Content planning refers to the process of deciding what content should be included in a document or other output and how it should be structured.
Content selection is a sub-process of content planning that involves deciding which pieces of content to include in the final output.
Document orchestration refers to the process of coordinating the different components of a document or other output to create a cohesive final product. This can involve deciding the order in which different pieces of content should be presented, as well as how they should be structured and formatted.
Document planner is a software system or algorithm that is used to automate the content planning process. A document planner may be responsible for deciding what content should be included in a document and how it should be structured, as well as coordinating the different components of the document to create a cohesive final product.
Document planning refers to the process of deciding what content should be included in a document or other output and how it should be structured. Document planning is often used to automate the creation of documents and other outputs, such as reports and presentations.
Document planning module is a component of a software system that is responsible for automating the document planning process. A document planning module may be responsible for deciding what content should be included in a document and how it should be structured.
Document structuring refers to the process of organizing the content of a document or other output in a way that conveys the desired information effectively. This can involve deciding the order in which different pieces of content should be presented, as well as how they should be formatted and structured.
Macro planning is a higher-level form of document planning that involves not only selecting the content for a document, but also structuring it in a way that conveys the desired information effectively. Macro-planning systems typically take a high-level representation of input data as input and generate a plan for selecting and organizing the final content of the document based on that input.
Narrative planning refers to the process of deciding what content should be included in a document or other output and how it should be structured in order to create a cohesive narrative or story. Narrative planning is often used in the creation of documents and other outputs that are meant to be read or presented as a narrative, such as stories, articles, and presentations.

Resources:

conf/rr .. international conference on web reasoning and rule systems proceedings *
conf/ruleml .. ruleml symposium proceedings (now merged with rr-conference.org) *

Wikipedia:

References:

Personality-dependent content selection in natural language generation systems
RM S. Ramos, DS Monteiro, I Paraboni – Journal of the Brazilian Computer …, 2020 – Springer
… Customised natural language generation NLG systems may in principle produce always the same fixed output text … the present work is to show that, in personality-dependent NLG systems, personality … more coarse-grained kind of CS performed at the document planning stage of …

Proceedings of the 13th International Conference on Natural Language Generation
B Davis, Y Graham, J Kelleher, Y Sripada – … Language Generation, 2020 – aclweb.org
… 12:00–13:00 Oral Session 5: Document Planning Schema-Guided Natural Language Generation Yuheng Du, Shereen Oraby, Vittorio Perera, Minmin Shen, Anjali Narayan-Chen … Neural NLG for Methodius: From RST Meaning Representations to Texts Symon Stevens-Guille …

Building a Persuasive Virtual Dietitian
L Anselma, A Mazzei – Informatics, 2020 – mdpi.com
… MADiMan consists of a numerical reasoner that takes into account users’ dietary constraints and automatically adapts the users’ diet, and of a natural language generation (NLG) system that automatically creates textual messages for explaining the results provided by the …

An explainable Link Discovery: Multilingual Link Specification verbalization and summarization
AF Ahmed, MA Sherif, D Moussallem… – Data & Knowledge …, 2021 – Elsevier
… Our approach is motivated by the pipeline architecture for natural language generation (NLG) systems performed by systems such as those introduced by … based approach underlying LS verbalization in Section 3. In Sections 3.1 and 3.2 we explain document-planner and the …

Stan: Towards describing bytecodes of smart contract
X Li, T Chen, X Luo, T Zhang, L Yu… – 2020 IEEE 20th …, 2020 – ieeexplore.ieee.org
… We analyze bytecodes through symbolic execution and generate readable descriptions following stan- dard workflow of NLG (Natural Language Generation) system … The NLG process follows the standard workflow of NLG system, ie, document planner, micro- planner …

Ontogen: A Knowledge-Based Approach to Natural Language Generation
IE Leon – 2020 – search.proquest.com
… ways in which humans can interact with agents beyond simple task automation. OntoGen is a Natural Language Generation (NLG) system that produces text from semantic meaning representations and generates grammatically correct and contextually relevant language. It …

Someone really wanted that song but it was not me!
S Najafian, O Inel, N Tintarev – 2020 – research.tudelft.nl
… We take a template-based approach, and apply a classical Natural Language Generation (NLG) pipeline [8]: Document planning. The first step is to analyze the require- ments for the content of the text that has to be generated …

Recycling a genre for news automation: The production of Valtteri the Election Bot
L Haapanen, L Leppänen – AILA Review, 2020 – jbe-platform.com
… In this paper, we define natural language generation as a process that takes non-linguistic input … neural networks, avoid this division into tasks entirely and instead view NLG as a … consider the above subtasks as grouped into three larger tasks, namely, document planning (#1–2 …

A NLG Framework for User Tailoring and Profiling in Healthcare.
S Balloccu, S Pauws, E Reiter – SmartPhil@ IUI, 2020 – ceur-ws.org
… Additional considerations: document planning and realisation The previous mockups adopt the following … Working with clinicians to improve a patient-information nlg system. In Proceedings of the Seventh International Natural Language Generation Conference, pages 100–104 …

Point at the Triple: Generation of Text Summaries from Knowledge Base Triples
P Vougiouklis, E Maddalena, J Hare… – Journal of Artificial …, 2020 – jair.org
… 1. Introduction Natural Language Generation (NLG) is the task of generating text that captures the content of structured-data records in a human-readable way (Reiter and Dale, 2000) … 2. Related Work NLG commonly has three steps: (i) document planning, (ii) micro …

Deep Learning Approaches to Text Production
Y Zhang – 2021 – direct.mit.edu
… Dale 2000; Gatt and Krahmer 2018) is also referred to as natural language generation (NLG) … I enjoyed the examples and figures demonstrating the typical NLG tasks such … components for a traditional pipeline, such as content selection, document planning, lexicalization, and …

Building natural language responses from natural language questions in the spatio-temporal context
G Landoulsi, K Mahmoudi… – International Journal of …, 2021 – inderscienceonline.com
… designed to reach this target, namely question answering systems (QAS), the natural language generation (NLG), etc … Hunter et al., 2012), are considered as the first medical NLG systems that … First of all, a document planning determines the final form of the textual descriptions to …

Table-to-Text: Generating Descriptive Text for Scientific Tables from Randomized Controlled Trials
Q Wei – 2020 – digitalcommons.library.tmc.edu
… channel for human beings. Success stories applying natural language generation (NLG) … specified a sub-problem of NLG, concept-to-text, which aims to generate language from … document planning, microplanning, and surface realization, which has been used in many …

Someone really wanted that song but it was not me! Evaluating Which Information to Disclose in Explanations for Group Recommendations
S Najafian, O Inel, N Tintarev – … of the 25th International Conference on …, 2020 – dl.acm.org
… We take a template-based approach, and apply a classical Natural Language Generation (NLG) pipeline [8]: Document planning. The first step is to analyze the require- ments for the content of the text that has to be generated …

A Roadmap to Realization Approaches in Natural Language Generation
L Kurup, M Narvekar – Ambient Communications and Computer Systems, 2020 – Springer
… Text generation has been an important aspect in the area of natural language generation, dated from … Any NLG system follows three basic processing steps to generate the text which is … The first one being document planning, the content and information needed to generate the …

Natural Language Generation
C Room – algorithms, 2020 – devopedia.org
… It’s also called Macro Planning or Document Planning. Information could come from a knowledge base … The first International Natural Language Generation Conference (INLG) is held, as a continuation of international workshops on NLG held regularly during 1982-1998 …

Explaining Habits and Changes of Activities in Smart Homes
H Banaee, G Chimamiwa, M Alirezaie… – Artificial Intelligence for …, 2020 – diva-portal.org
… Natural Language Generation (NLG) is one technique whereby natural linguistic de- scriptions are generated from … NLG systems are designed to recognise the significance of informa- tion … data analysis and content determination as well as document planning, micro-planning …

How are you? Introducing stress-based text tailoring
S Balloccu, E Reiter, A Johnstone, C Fyfe – arXiv preprint arXiv …, 2020 – arxiv.org
… This can be split into two comple- mentary document planning (length) and micro- planning (terminology) steps … 2018. Saferdrive: An nlg-based behaviour change support system for drivers … Fiorella De Rosis and Floriana Grasso. 1999. Affec- tive natural language generation …

Abstraction and Summarization of Meaning in Natural Language Processing
F Huseynova – ijels.com
… Module Content Task Structure Task Document planning Content determination Document structuring Microplanning … 335. [6] L. Danlos, F. Meunier, V. Combet, EasyText: an Operational NLG System, in: Proc. of ENLG’11, 2011, pp.139–144 …

Automated Java exceptions explanation using natural language generation techniques
FY Assiri, H Elazhary – Computer Applications in Engineering …, 2020 – Wiley Online Library
… LANGUAGE GENERATION. The NLG process typically starts with a set of communicative goals or facts that need to be communicated in the form of NL surface text to human readers or listeners, based on three main tasks [19, 26, 27]. The first task is document planning , in …

A Novel GCN Architecture for Text Generation from Knowledge Graphs: Full Node Embedded Strategy and Context Gate with Copy and Penalty Mechanism
Z Hua, W Zhangb, D Wanga, W Niub, F Mob, J Maa… – semantic-web-journal.net
… Moreover, document planning issues, such as order, coherence, and discourse markers, should be considered for gen- erating a concise and faithful … KG-to-Text, one of the important tasks in natural language generation (NLG) [15], aims to generate text from the sub-graph struc …

“Automation will save journalism”–News automation from the service providers’ point of view
M Kjellman – 2021 – helda.helsinki.fi
… structured interviews were conducted with representatives from companies that create natural language generation software used … 1). According to Reiter (2013), one key aspect that sets NLG systems apart … The general NLP process consists of three stages: document planning …

Neural generation of textual summaries from knowledge base triples
P Vougiouklis – 2020 – books.google.com
… Natural Language Generation (NLG) is concerned with the devel- opment of the textual interfaces that generate text that describes the input records of a structured data source in a fluent and sensible man- ner [Reiter and Dale, 2000] …

The CACAPO Dataset: A Multilingual, Multi-Domain Dataset for Neural Pipeline and End-to-End Data-to-Text Generation
C van der Lee, C Emmery, S Wubben… – … Language Generation, 2020 – aclweb.org
… Neural Natural Language Generation (NLG) is a promising technique, as neural NLG systems are not bound … texts and relevant data) is required for training neural NLG systems, and … Dale, 2000) that sequen- tially performs tasks related to document planning, sentence planning …

Design and Implementation of Phylotastic, a Service Architecture for Evolutionary Biology
ASM Tayeen, TH Nguyen, VD Nguyen… – International Journal of …, 2020 – World Scientific
… 3.6. Natural language generation … In this version, the NLG module is developed manually following three major processing phases: (1) document planning (content determination); (2) micro-planning; and (3) surface realization as described in [29] …

Deep learning approaches to text production
S Narayan, C Gardent – Synthesis Lectures on Human …, 2020 – morganclaypool.com
Page 1. 0 N AR A Y AN • GAR DE N T DE E PL EAR NING AP P R O A CH E ST OT E X TP R OD UCT ION M O R GAN & CL A YPOO L Page 2. Page 3. Deep Learning Approaches to Text Production Page 4. Page 5. Synthesis Lectures on Human Language Technologies …

RDFJSREALB: a Symbolic Approach for Generating Text from RDF Triples
G Lapalme – 2020 – rali.iro.umontreal.ca
… 2003) were pioneers in determining relevant content for NLG using statistical … propose a system that can automatically learn sentence templates and document planning from parallel … Dong and Holder (2014) present Natural Language Generation from Graphs (NLGG) with three …

The 2020 bilingual, bi-directional webnlg+ shared task overview and evaluation results (webnlg+ 2020)
T Ferreira, C Gardent, N Ilinykh… – … Natural Language …, 2020 – hal.archives-ouvertes.fr
… to provide a common benchmark on which to evaluate and compare “micro-planners”, ie, Natural Language Generation (NLG) systems which can … OSU Neural NLG … are also examined to analyse how different ways of integrating generation with document planning (triples order …

Data-to-text Generation with Macro Planning
R Puduppully, M Lapata – arXiv preprint arXiv:2102.02723, 2021 – arxiv.org
… selves to document planning as there is no explicit link between the summary and the content of the game (which is encoded in tabular form) … 2 Related Work Content planning has been traditionally consid- ered a fundamental component in natural language generation …

Organization of museum exhibit information for ubiquitous visitors
P Khanwalkar, P Venkataram – EAI Endorsed Transactions on …, 2020 – eprints.eudl.eu
… M- PIRO has discussed four stages within natural language generation for the exhibit information descriptions: (i) Content selection specifies the facts ie, relations or attributes that the exhibit information should describe; (ii) Document planning which specifies the sequence of …

Modeling global and local node contexts for text generation from knowledge graphs
LFR Ribeiro, Y Zhang, C Gardent… – Transactions of the …, 2020 – MIT Press
… In this context, in addition to sentence generation, document planning needs to be handled: The input needs to be mapped into several sentences; sentences need to be ordered and connected using appropriate discourse markers; and inter-sentential anaphora and ellipsis …

Conversational AI: Dialogue Systems, Conversational Agents, and Chatbots
M McTear – Synthesis Lectures on Human Language …, 2020 – morganclaypool.com
Page 1. MCT EAR C ONVE R SA T ION AL AI M O R GAN & CL A YPOO L Page 2. Page 3. Conversational AI Dialogue Systems, Conversational Agents, and Chatbots Page 4. Page 5. Synthesis Lectures on Human Language Technologies …

Contribution to Natural Language Generation for Spanish
S Garcia Mendez – 2021 – investigo.biblioteca.uvigo.es
… v Page 13. Page 14. Abstract In this thesis, we present our research aligned with the field of Natural Language Generation (NLG). Our work represents an effort to bring NLG capabilities to the research community for Spanish language …

Verbal explanations by collaborating robot teams
AK Singh, N Baranwal, KF Richter… – Paladyn, Journal of …, 2020 – degruyter.com
Jump to Content Jump to Main Navigation Publications. Subjects. Architecture and Design Arts Asian and Pacific Studies Business and Economics Chemistry Classical and Ancient Near Eastern Studies Computer Sciences Cultural …

The 2020 Bilingual, Bi-Directional WebNLG+ Shared Task Overview and Evaluation Results (WebNLG
T Ferreira, C Gardent, N Ilinykh, C van der Lee, S Mille… – pure.uvt.nl
… to provide a common benchmark on which to evaluate and compare “micro-planners”, ie, Natural Language Generation (NLG) systems which can … OSU Neural NLG … are also examined to analyse how different ways of integrating generation with document planning (triples order …

Evaluation of text generation: A survey
A Celikyilmaz, E Clark, J Gao – arXiv preprint arXiv:2006.14799, 2020 – arxiv.org
… The first stage is document planning, in which the content and its order are determined and a text plan that outlines … 1.2 Why a Survey on Evaluation on Natural Language Generation … are interested in in this paper is how to measure the quality of text generated from NLG models …

Knowledge Graphs for Multilingual Language Translation and Generation
D Moussallem – arXiv preprint arXiv:2009.07715, 2020 – arxiv.org
… presented herein began. Additionally, few works had studied the contribution of KGs to Natural Language Generation (NLG) tasks. Moreover, the multilinguality also remained an open research area in these respective tasks (Young et al., 2018) …

Linguistically Informed Language Generation: A Multi-faceted Approach
D Kang – 2020 – lti.cs.cmu.edu
… Natural language generation (NLG) is a key component of many language tech- nology applications such as dialogue systems, like Amazon’s Alexa; question an- swering systems, like IBM Watson; automatic email replies, like Google’s SmartRe- ply; and story generation …