PASS: A Dutch data-to-text system for soccer, targeted towards specific audiences

@inproceedings{Lee2017PASSAD,
  title={PASS: A Dutch data-to-text system for soccer, targeted towards specific audiences},
  author={Chris van der Lee and E. Krahmer and S. Wubben},
  booktitle={INLG},
  year={2017}
}
We present PASS, a data-to-text system that generates Dutch soccer reports from match statistics. One of the novel elements of PASS is the fact that the system produces corpus-based texts tailored towards fans of one club or the other, which can most prominently be observed in the tone of voice used in the reports. Furthermore, the system is open source and uses a modular design, which makes it relatively easy for people to add extensions. Human-based evaluation shows that people are generally… Expand
Evaluating the text quality, human likeness and tailoring component of PASS: A Dutch data-to-text system for soccer
TLDR
An evaluation of PASS, a data-to-text system that generates Dutch soccer reports from match statistics which are automatically tailored towards fans of one club or the other, found that compared to human-written texts, computer-generated texts were rated slightly lower on style-related text components and slightly higher in terms of the correctness of given information. Expand
Template-based multilingual football reports generation using Wikidata as a knowledge base
TLDR
This paper describes how, in a short amount of time, PASS was extended to produce English texts, exploiting machine translation and Wikidata as a large-scale source of multilingual knowledge. Expand
MEmoFC: introducing the Multilingual Emotional Football Corpus
TLDR
It is demonstrated how the corpus can be used to investigate the influence of affect on the reports through different approaches and illustrated how game outcome influences references to the own team and the opponent, and the use of positive and negative emotion terms in the different languages. Expand
An NLG System for Constituent Correspondence: Personality, Affect, and Alignment
TLDR
PoliScribe, a pipeline-architectured NLG platform, constructs personalized responses to constituents inquiring about legislation through structurable data that can be leveraged by a natural language generation (NLG) system to create a coherent letter response. Expand
A Closer Look at Recent Results of Verb Selection for Data-to-Text NLG
TLDR
Two simple, efficient, interpretable baseline approaches for statistical selection of trend verbs are provided, which give a strong performance on both previously used evaluation metrics and the new evaluation metrics. Expand
Generating Objective Summaries of Sports Matches Using Social Media
TLDR
A function maximization method which measures the objectivity of each post based on its constituent words and a neural network-based approach, where the author uses an encoder-decoder architecture, which shows promising results that indicate the effectiveness of these methods. Expand
Automated learning of templates for data-to-text generation: comparing rule-based, statistical and neural methods
TLDR
Evaluation using BLEU and human evaluation metrics indicated that Neural Machine Translation yielded the highest quality output and that the templatization method was able to increase text quality in multiple situations. Expand
Towards Subjective Multimedia Summarization Framework for Sporting Event in the Context of Digital Twins
TLDR
This thesis introduces an event-based summarization approach that incorporates multimedia data, sentiment, and human reactions with respect to their point of view in order to generate a concise subjective multimedia summary of the event. Expand
Generating Live Soccer-Match Commentary from Play Data
TLDR
An encoder is proposed for play event data, which is enhanced with a gate mechanism and an attention mechanism on events, and placeholders and their reconstruction mechanism are introduced to enable the model to copy appropriate player names and team names from the input data. Expand
Intégration de VerbNet dans un réalisateur profond
Natural language generation’s (NLG) goal is to produce understandable text from nonlinguistic data. Generation essentially consists in two tasks : first, determine the content of a message toExpand
...
1
2
3
...

References

SHOWING 1-10 OF 28 REFERENCES
The Multilingual Affective Soccer Corpus (MASC): Compiling a biased parallel corpus on soccer reportage in English, German and Dutch
TLDR
The Multilingual Affective Soccer Corpus is a collection of soccer match reports in English, German and Dutch with the aim of investigating the role of affect in sports reportage in different languages and cultures. Expand
Learning to sportscast: a test of grounded language acquisition
TLDR
A novel commentator system that learns language from sportscasts of simulated soccer games and uses a novel algorithm, Iterative Generation Strategy Learning (IGSL), for deciding which events to comment on. Expand
An Architecture for Data-to-Text Systems
I present an architecture for data-to-text systems, that is NLG systems which produce texts from non-linguistic input data; this essentially extends the architecture of Reiter and Dale (2000) toExpand
Content Selection in Data-to-Text Systems: A Survey
TLDR
This survey initially introduces the field of data-to-text generation, describes the general data- to-text system architecture and then it reviews the state-of-the-art content selection methods. Expand
Towards more variation in text generation: Developing and evaluating variation models for choice of referential form
TLDR
A nondeterministic method for referring expression generation using a Naive Bayes model and a Recurrent Neural Network to generate referential forms in texts from the GREC-2.0 corpus is introduced. Expand
From data to speech: a general approach
TLDR
The use of syntactically enriched templates is guided by knowledge of the discourse context, while for speech generation pre-recorded phrases are combined in a prosodically sophisticated manner to achieve a better prosodic output quality than can be achieved in a plain text-to-speech system. Expand
Unsupervised Concept-to-text Generation with Hypergraphs
TLDR
This work presents a joint model that captures content selection and surface realization in an unsupervised domain-independent fashion and defines a probabilistic context-free grammar that globally describes the inherent structure of the input (a corpus of database records and text describing some of them). Expand
Collective Content Selection for Concept-to-Text Generation
TLDR
This work presents an efficient method for automatically learning content selection rules from a corpus and its related database and treats content selection as a collective classification problem, thus allowing it to capture contextual dependencies between input items. Expand
Building a System for Stock News Generation in Russian
TLDR
An implementation of an NLG system that serves for stock news generation that has two modules: analysis module and NLG module that gives relatively accurate results and can be used as a newsbot forStock news generation. Expand
Reactive Content Selection in the Generation of Real-time Soccer Commentary
TLDR
It is described how a principle of maximizing the total gain of importance scores during a game can be used to incorporate content selection into the surface generation module, thus accounting for issues such as interruption and abbreviation. Expand
...
1
2
3
...