We present a method for automatically generating descriptions of biological events encoded in the KB Bio 101 Knowledge base. We evaluate our approach on a corpus of 336 event descriptions, provide a qualitative and quantitative analysis of the results obtained and discuss possible directions for further work.
{"title":"A Domain Agnostic Approach to Verbalizing n-ary Events without Parallel Corpora","authors":"B. Gyawali, Claire Gardent, Christophe Cerisara","doi":"10.18653/v1/W15-4703","DOIUrl":"https://doi.org/10.18653/v1/W15-4703","url":null,"abstract":"We present a method for automatically generating descriptions of biological events encoded in the KB Bio 101 Knowledge base. We evaluate our approach on a corpus of 336 event descriptions, provide a qualitative and quantitative analysis of the results obtained and discuss possible directions for further work.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132772430","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A controlled use of omnipresent data can leverage a potential of services never reached before. In this paper, we propose a user driven approach to take advantage of massive data streams. Our solution, named Stream2Text, relies on a personalized and continuous refinement of data to generate texts (in natural language) that provide a tailored synthesis of relevant data. It enables monitoring by a wide range of users as text streams can be shared on social networks or used individually on mobile devices.
{"title":"A Personal Storytelling about Your Favorite Data","authors":"C. Labbé, C. Roncancio, D. Bras","doi":"10.18653/v1/W15-4727","DOIUrl":"https://doi.org/10.18653/v1/W15-4727","url":null,"abstract":"A controlled use of omnipresent data can leverage a potential of services never reached before. In this paper, we propose a user driven approach to take advantage of massive data streams. Our solution, named Stream2Text, relies on a personalized and continuous refinement of data to generate texts (in natural language) that provide a tailored synthesis of relevant data. It enables monitoring by a wide range of users as text streams can be shared on social networks or used individually on mobile devices.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"245 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116152268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We present the sentence ordering part of a natural language generation module, used in the framework of a knowledge base of electronic navigation charts and sailing directions. Sentence ordering is done via the extraction of frequent ``interesting patterns''. The particularity of the knowledge base is that it is based on a controlled hybrid language, that is the combination of a controlled natural language and a controlled visual language. Thanks to the language's hybrid nature, the sentence ordering process is able to take into account hybrid (textual and visual) information, involving cartographic data, as well as landscape "read" by the navigator.
{"title":"Sentence Ordering in Electronic Navigational Chart Companion Text Generation","authors":"Julie Sauvage-Vincent, Y. Haralambous, J. Puentes","doi":"10.18653/v1/W15-4710","DOIUrl":"https://doi.org/10.18653/v1/W15-4710","url":null,"abstract":"We present the sentence ordering part of a natural language generation module, used in the framework of a knowledge base of electronic navigation charts and sailing directions. Sentence ordering is done via the extraction of frequent ``interesting patterns''. The particularity of the knowledge base is that it is based on a controlled hybrid language, that is the combination of a controlled natural language and a controlled visual language. Thanks to the language's hybrid nature, the sentence ordering process is able to take into account hybrid (textual and visual) information, involving cartographic data, as well as landscape \"read\" by the navigator.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122607507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We have explored how a conversational agent can introduce a selected topic in an ongoing non-task oriented interaction with a user, where the selected topic has little to do with the current topic. Based on the reasoning process of the agent we have constructed a set of transition strategies to introduce the new topic. We tested the effects of each of these strategies on the perception of the dialogue and the agent.
{"title":"Topic Transition Strategies for an Information-Giving Agent","authors":"N. Glas, C. Pelachaud","doi":"10.18653/v1/w15-4725","DOIUrl":"https://doi.org/10.18653/v1/w15-4725","url":null,"abstract":"We have explored how a conversational agent can introduce a selected topic in an ongoing non-task oriented interaction with a user, where the selected topic has little to do with the current topic. Based on the reasoning process of the agent we have constructed a set of transition strategies to introduce the new topic. We tested the effects of each of these strategies on the perception of the dialogue and the agent.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121847550","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we present the task of generating image descriptions with gold standard visual detections as input, rather than directly from an image. This allows the Natural Language Generation community to focus on the text generation process, rather than dealing with the noise and complications arising from the visual detection process. We propose a fine-grained evaluation metric specifically for evaluating the content selection capabilities of image description generation systems. To demonstrate the evaluation metric on the task, several baselines are presented using bounding box information and textual information as priors for content selection. The baselines are evaluated using the proposed metric, showing that the fine-grained metric is useful for evaluating the content selection phase of an image description generation system.
{"title":"Generating Image Descriptions with Gold Standard Visual Inputs: Motivation, Evaluation and Baselines","authors":"Josiah Wang, R. Gaizauskas","doi":"10.18653/v1/W15-4722","DOIUrl":"https://doi.org/10.18653/v1/W15-4722","url":null,"abstract":"In this paper, we present the task of generating image descriptions with gold standard visual detections as input, rather than directly from an image. This allows the Natural Language Generation community to focus on the text generation process, rather than dealing with the noise and complications arising from the visual detection process. We propose a fine-grained evaluation metric specifically for evaluating the content selection capabilities of image description generation systems. To demonstrate the evaluation metric on the task, several baselines are presented using bounding box information and textual information as priors for content selection. The baselines are evaluated using the proposed metric, showing that the fine-grained metric is useful for evaluating the content selection phase of an image description generation system.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133638009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. Sevens, Vincent Vandeghinste, Ineke Schuurman, F. V. Eynde
We present a Pictograph-to-Text translation system for people with Intellectual or Developmental Disabilities (IDD). The system translates pictograph messages, consisting of one or more pictographs, into Dutch text using WordNet links and an ngram language model. We also provide several pictograph input methods assisting the users in selecting the appropriate pictographs.
{"title":"Natural Language Generation from Pictographs","authors":"L. Sevens, Vincent Vandeghinste, Ineke Schuurman, F. V. Eynde","doi":"10.18653/v1/W15-4711","DOIUrl":"https://doi.org/10.18653/v1/W15-4711","url":null,"abstract":"We present a Pictograph-to-Text translation system for people with Intellectual or Developmental Disabilities (IDD). The system translates pictograph messages, consisting of one or more pictographs, into Dutch text using WordNet links and an ngram language model. We also provide several pictograph input methods assisting the users in selecting the appropriate pictographs.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115959675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
B. B. Miranda, S. Caffiau, C. Garbay, François Portet
Automatic story generation is the subject of a growing research effort which has mainly focused on fictional stories. In this paper, we present some preliminary work to generate r´ (stories) from sensors data acquired during a ski sortie. In this approach, the story planning is performed using a task model that represents domain knowledge and sequential constraints between ski activities. To test the validity of the task model, a small-scale user evaluation was performed to compare the human perception of r´ ecit plans from hand writ...
{"title":"Generating Récit from Sensor Data: Evaluation of a Task Model for Story Planning and Preliminary Experiments with GPS Data","authors":"B. B. Miranda, S. Caffiau, C. Garbay, François Portet","doi":"10.18653/v1/W15-4714","DOIUrl":"https://doi.org/10.18653/v1/W15-4714","url":null,"abstract":"Automatic story generation is the subject of a growing research effort which has mainly focused on fictional stories. In this paper, we present some preliminary work to generate r´ (stories) from sensors data acquired during a ski sortie. In this approach, the story planning is performed using a task model that represents domain knowledge and sequential constraints between ski activities. To test the validity of the task model, a small-scale user evaluation was performed to compare the human perception of r´ ecit plans from hand writ...","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125247042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Referring to landmarks has been identified to lead to improved navigation instructions. However, a previous corpus study suggests that human “wizards” also choose to refer to street names and generate user-centric instructions. In this paper, we conduct a task-based evaluation of two systems reflecting the wizards’ behaviours and compare them against an improved version of previous landmark-based systems, which resorts to user-centric descriptions if the landmark is estimated to be invisible. We use the GRUVE virtual interactive environment for evaluation. We find that the improved system, which takes visibility into account, outperforms the corpus-based wizard strategies, however not significantly. We also show a significant effect of prior user knowledge, which suggests the usefulness of a user modelling approach.
{"title":"Generating and Evaluating Landmark-Based Navigation Instructions in Virtual Environments","authors":"A. C. Curry, Dimitra Gkatzia, Verena Rieser","doi":"10.18653/v1/W15-4715","DOIUrl":"https://doi.org/10.18653/v1/W15-4715","url":null,"abstract":"Referring to landmarks has been identified to lead to improved navigation instructions. However, a previous corpus study suggests that human “wizards” also choose to refer to street names and generate user-centric instructions. In this paper, we conduct a task-based evaluation of two systems reflecting the wizards’ behaviours and compare them against an improved version of previous landmark-based systems, which resorts to user-centric descriptions if the landmark is estimated to be invisible. We use the GRUVE virtual interactive environment for evaluation. We find that the improved system, which takes visibility into account, outperforms the corpus-based wizard strategies, however not significantly. We also show a significant effect of prior user knowledge, which suggests the usefulness of a user modelling approach.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"323 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122103047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a parsing paradigm for natural language generation task, which learns a tailored probabilistic context-free grammar for encoding meaning representation (MR) and its corresponding natural language (NL) expression, then decodes and yields natural language sentences at the leaves of the optimal parsing tree for a target meaning representation. The major advantage of our method is that it does not require any prior knowledge of the MR syntax for training. We deployed our method in response generation for a Chinese spoken dialogue system, obtaining results comparable to a strong baseline both in terms of BLEU scores and human evaluation.
{"title":"Response Generation in Dialogue Using a Tailored PCFG Parser","authors":"Caixia Yuan, Xiaojie Wang, Qianhui He","doi":"10.18653/v1/W15-4713","DOIUrl":"https://doi.org/10.18653/v1/W15-4713","url":null,"abstract":"This paper presents a parsing paradigm for natural language generation task, which learns a tailored probabilistic context-free grammar for encoding meaning representation (MR) and its corresponding natural language (NL) expression, then decodes and yields natural language sentences at the leaves of the optimal parsing tree for a target meaning representation. The major advantage of our method is that it does not require any prior knowledge of the MR syntax for training. We deployed our method in response generation for a Chinese spoken dialogue system, obtaining results comparable to a strong baseline both in terms of BLEU scores and human evaluation.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129406318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents an ongoing project about the symbolic translation from Italian to Italian Signed Language (LIS) in the rail stations domain. We describe some technical issues in the generation side of the translation, i.e. the use of XML templates for microplanning, the implementation of some LIS linguistic features in the grammar.
{"title":"Translating Italian to LIS in the Rail Stations","authors":"A. Mazzei","doi":"10.18653/v1/W15-4712","DOIUrl":"https://doi.org/10.18653/v1/W15-4712","url":null,"abstract":"This paper presents an ongoing project about the symbolic translation from Italian to Italian Signed Language (LIS) in the rail stations domain. We describe some technical issues in the generation side of the translation, i.e. the use of XML templates for microplanning, the implementation of some LIS linguistic features in the grammar.","PeriodicalId":307841,"journal":{"name":"European Workshop on Natural Language Generation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128697412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：