首页 > 最新文献

Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)最新文献

英文 中文
Improving Specificity in Review Response Generation with Data-Driven Data Filtering 用数据驱动的数据过滤提高评论反应生成的特异性
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.15
Tannon Kew, M. Volk
Responding to online customer reviews has become an essential part of successfully managing and growing a business both in e-commerce and the hospitality and tourism sectors. Recently, neural text generation methods intended to assist authors in composing responses have been shown to deliver highly fluent and natural looking texts. However, they also tend to learn a strong, undesirable bias towards generating overly generic, one-size-fits-all outputs to a wide range of inputs. While this often results in ‘safe’, high-probability responses, there are many practical settings in which greater specificity is preferable. In this work we examine the task of generating more specific responses for online reviews in the hospitality domain by identifying generic responses in the training data, filtering them and fine-tuning the generation model. We experiment with a range of data-driven filtering methods and show through automatic and human evaluation that, despite a 60% reduction in the amount of training data, filtering helps to derive models that are capable of generating more specific, useful responses.
在电子商务、酒店和旅游业中,响应在线客户评论已成为成功管理和发展业务的重要组成部分。最近,神经文本生成方法旨在帮助作者组成响应已被证明提供高度流畅和自然的文本。然而,他们也倾向于学习一种强烈的、不受欢迎的偏见,即对广泛的投入产生过于通用的、一刀切的产出。虽然这通常会导致“安全”的高概率反应,但在许多实际情况下,更大的特异性是可取的。在这项工作中,我们通过识别培训数据中的一般响应,过滤它们并微调生成模型,研究为酒店领域的在线评论生成更具体的响应的任务。我们对一系列数据驱动的过滤方法进行了实验,并通过自动和人工评估表明,尽管训练数据量减少了60%,但过滤有助于推导出能够生成更具体、更有用的响应的模型。
{"title":"Improving Specificity in Review Response Generation with Data-Driven Data Filtering","authors":"Tannon Kew, M. Volk","doi":"10.18653/v1/2022.ecnlp-1.15","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.15","url":null,"abstract":"Responding to online customer reviews has become an essential part of successfully managing and growing a business both in e-commerce and the hospitality and tourism sectors. Recently, neural text generation methods intended to assist authors in composing responses have been shown to deliver highly fluent and natural looking texts. However, they also tend to learn a strong, undesirable bias towards generating overly generic, one-size-fits-all outputs to a wide range of inputs. While this often results in ‘safe’, high-probability responses, there are many practical settings in which greater specificity is preferable. In this work we examine the task of generating more specific responses for online reviews in the hospitality domain by identifying generic responses in the training data, filtering them and fine-tuning the generation model. We experiment with a range of data-driven filtering methods and show through automatic and human evaluation that, despite a 60% reduction in the amount of training data, filtering helps to derive models that are capable of generating more specific, useful responses.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132631275","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}
引用次数: 0
Clause Topic Classification in German and English Standard Form Contracts 德文和英文标准格式合同的条款主题分类
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.23
Daniel Braun, F. Matthes
So-called standard form contracts, i.e. contracts that are drafted unilaterally by one party, like terms and conditions of online shops or terms of services of social networks, are cornerstones of our modern economy. Their processing is, therefore, of significant practical value. Often, the sheer size of these contracts allows the drafting party to hide unfavourable terms from the other party. In this paper, we compare different approaches for automatically classifying the topics of clauses in standard form contracts, based on a data-set of more than 6,000 clauses from more than 170 contracts, which we collected from German and English online shops and annotated based on a taxonomy of clause topics, that we developed together with legal experts. We will show that, in our comparison of seven approaches, from simple keyword matching to transformer language models, BERT performed best with an F1-score of up to 0.91, however much simpler and computationally cheaper models like logistic regression also achieved similarly good results of up to 0.87.
所谓的标准形式合同,即由一方单方面起草的合同,如网上商店的条款和条件或社交网络的服务条款,是我们现代经济的基石。因此,它们的处理具有重要的实用价值。通常,这些合同的庞大规模允许起草方向另一方隐瞒不利条款。在本文中,我们比较了标准形式合同中条款主题自动分类的不同方法,基于我们从德语和英语在线商店收集的170多个合同中的6,000多个条款的数据集,并根据我们与法律专家一起开发的条款主题分类法进行了注释。我们将展示,在我们对七种方法的比较中,从简单的关键字匹配到转换语言模型,BERT表现最好,f1得分高达0.91,然而更简单和计算成本更低的模型,如逻辑回归,也取得了类似的好结果,高达0.87。
{"title":"Clause Topic Classification in German and English Standard Form Contracts","authors":"Daniel Braun, F. Matthes","doi":"10.18653/v1/2022.ecnlp-1.23","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.23","url":null,"abstract":"So-called standard form contracts, i.e. contracts that are drafted unilaterally by one party, like terms and conditions of online shops or terms of services of social networks, are cornerstones of our modern economy. Their processing is, therefore, of significant practical value. Often, the sheer size of these contracts allows the drafting party to hide unfavourable terms from the other party. In this paper, we compare different approaches for automatically classifying the topics of clauses in standard form contracts, based on a data-set of more than 6,000 clauses from more than 170 contracts, which we collected from German and English online shops and annotated based on a taxonomy of clause topics, that we developed together with legal experts. We will show that, in our comparison of seven approaches, from simple keyword matching to transformer language models, BERT performed best with an F1-score of up to 0.91, however much simpler and computationally cheaper models like logistic regression also achieved similarly good results of up to 0.87.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117160306","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}
引用次数: 2
Data Quality Estimation Framework for Faster Tax Code Classification 快速税号分类的数据质量估计框架
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.4
R. Kondadadi, Allen Williams, Nicolas Nicolov
This paper describes a novel framework to estimate the data quality of a collection of product descriptions to identify required relevant information for accurate product listing classification for tax-code assignment. Our Data Quality Estimation (DQE) framework consists of a Question Answering (QA) based attribute value extraction model to identify missing attributes and a classification model to identify bad quality records. We show that our framework can accurately predict the quality of product descriptions. In addition to identifying low-quality product listings, our framework can also generate a detailed report at a category level showing missing product information resulting in a better customer experience.
本文描述了一种新的框架来估计产品描述集合的数据质量,以确定准确的产品清单分类所需的相关信息。我们的数据质量评估(DQE)框架由一个基于问答(QA)的属性值提取模型(用于识别缺失属性)和一个分类模型(用于识别不良质量记录)组成。我们证明了我们的框架可以准确地预测产品描述的质量。除了识别低质量的产品列表外,我们的框架还可以在类别级别生成详细的报告,显示缺失的产品信息,从而带来更好的客户体验。
{"title":"Data Quality Estimation Framework for Faster Tax Code Classification","authors":"R. Kondadadi, Allen Williams, Nicolas Nicolov","doi":"10.18653/v1/2022.ecnlp-1.4","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.4","url":null,"abstract":"This paper describes a novel framework to estimate the data quality of a collection of product descriptions to identify required relevant information for accurate product listing classification for tax-code assignment. Our Data Quality Estimation (DQE) framework consists of a Question Answering (QA) based attribute value extraction model to identify missing attributes and a classification model to identify bad quality records. We show that our framework can accurately predict the quality of product descriptions. In addition to identifying low-quality product listings, our framework can also generate a detailed report at a category level showing missing product information resulting in a better customer experience.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115367613","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}
引用次数: 0
Can Pretrained Language Models Generate Persuasive, Faithful, and Informative Ad Text for Product Descriptions? 预训练的语言模型能生成有说服力的、忠实的、信息丰富的产品描述广告文本吗?
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.27
Fajri Koto, Jey Han Lau, Timothy Baldwin
For any e-commerce service, persuasive, faithful, and informative product descriptions can attract shoppers and improve sales. While not all sellers are capable of providing such interesting descriptions, a language generation system can be a source of such descriptions at scale, and potentially assist sellers to improve their product descriptions. Most previous work has addressed this task based on statistical approaches (Wang et al., 2017), limited attributes such as titles (Chen et al., 2019; Chan et al., 2020), and focused on only one product type (Wang et al., 2017; Munigala et al., 2018; Hong et al., 2021). In this paper, we jointly train image features and 10 text attributes across 23 diverse product types, with two different target text types with different writing styles: bullet points and paragraph descriptions. Our findings suggest that multimodal training with modern pretrained language models can generate fluent and persuasive advertisements, but are less faithful and informative, especially out of domain.
对于任何电子商务服务来说,有说服力的、忠实的、信息丰富的产品描述都能吸引顾客,提高销售额。虽然不是所有的卖家都有能力提供如此有趣的描述,但语言生成系统可以成为大规模描述的来源,并有可能帮助卖家改进他们的产品描述。之前的大多数工作都是基于统计方法(Wang et al., 2017)和有限的属性(如标题)来解决这个问题的(Chen et al., 2019;Chan et al., 2020),并且只关注一种产品类型(Wang et al., 2017;Munigala et al., 2018;Hong et al., 2021)。在本文中,我们共同训练了23种不同产品类型的图像特征和10个文本属性,使用了两种不同的写作风格的目标文本类型:项目符号和段落描述。我们的研究结果表明,使用现代预训练语言模型进行多模式训练可以生成流畅和有说服力的广告,但缺乏可信度和信息量,特别是在域外。
{"title":"Can Pretrained Language Models Generate Persuasive, Faithful, and Informative Ad Text for Product Descriptions?","authors":"Fajri Koto, Jey Han Lau, Timothy Baldwin","doi":"10.18653/v1/2022.ecnlp-1.27","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.27","url":null,"abstract":"For any e-commerce service, persuasive, faithful, and informative product descriptions can attract shoppers and improve sales. While not all sellers are capable of providing such interesting descriptions, a language generation system can be a source of such descriptions at scale, and potentially assist sellers to improve their product descriptions. Most previous work has addressed this task based on statistical approaches (Wang et al., 2017), limited attributes such as titles (Chen et al., 2019; Chan et al., 2020), and focused on only one product type (Wang et al., 2017; Munigala et al., 2018; Hong et al., 2021). In this paper, we jointly train image features and 10 text attributes across 23 diverse product types, with two different target text types with different writing styles: bullet points and paragraph descriptions. Our findings suggest that multimodal training with modern pretrained language models can generate fluent and persuasive advertisements, but are less faithful and informative, especially out of domain.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"57 11","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120921205","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}
引用次数: 7
Structured Extraction of Terms and Conditions from German and English Online Shops 从德语和英语网上商店的条款和条件的结构化提取
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.21
Tobias Schamel, Daniel Braun, F. Matthes
The automated analysis of Terms and Conditions has gained attention in recent years, mainly due to its relevance to consumer protection. Well-structured data sets are the base for every analysis. While content extraction, in general, is a well-researched field and many open source libraries are available, our evaluation shows, that existing solutions cannot extract Terms and Conditions in sufficient quality, mainly because of their special structure. In this paper, we present an approach to extract the content and hierarchy of Terms and Conditions from German and English online shops. Our evaluation shows, that the approach outperforms the current state of the art. A python implementation of the approach is made available under an open license.
条款和条件的自动分析近年来引起了人们的关注,主要是因为它与消费者保护有关。结构良好的数据集是所有分析的基础。虽然内容提取通常是一个研究得很好的领域,并且有许多开源库可用,但我们的评估表明,现有的解决方案无法以足够的质量提取条款和条件,主要是因为它们的特殊结构。在本文中,我们提出了一种从德语和英语在线商店中提取条款和条件的内容和层次结构的方法。我们的评估表明,这种方法优于目前的最先进的方法。该方法的python实现在开放许可下可用。
{"title":"Structured Extraction of Terms and Conditions from German and English Online Shops","authors":"Tobias Schamel, Daniel Braun, F. Matthes","doi":"10.18653/v1/2022.ecnlp-1.21","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.21","url":null,"abstract":"The automated analysis of Terms and Conditions has gained attention in recent years, mainly due to its relevance to consumer protection. Well-structured data sets are the base for every analysis. While content extraction, in general, is a well-researched field and many open source libraries are available, our evaluation shows, that existing solutions cannot extract Terms and Conditions in sufficient quality, mainly because of their special structure. In this paper, we present an approach to extract the content and hierarchy of Terms and Conditions from German and English online shops. Our evaluation shows, that the approach outperforms the current state of the art. A python implementation of the approach is made available under an open license.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127468134","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}
引用次数: 1
Product Titles-to-Attributes As a Text-to-Text Task 作为文本到文本任务的产品标题到属性
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.12
Gilad Fuchs, Yoni Acriche
Online marketplaces use attribute-value pairs, such as brand, size, size type, color, etc. to help define important and relevant facts about a listing. These help buyers to curate their search results using attribute filtering and overall create a richer experience. Although their critical importance for listings’ discoverability, getting sellers to input tens of different attribute-value pairs per listing is costly and often results in missing information. This can later translate to the unnecessary removal of relevant listings from the search results when buyers are filtering by attribute values. In this paper we demonstrate using a Text-to-Text hierarchical multi-label ranking model framework to predict the most relevant attributes per listing, along with their expected values, using historic user behavioral data. This solution helps sellers by allowing them to focus on verifying information on attributes that are likely to be used by buyers, and thus, increase the expected recall for their listings. Specifically for eBay’s case we show that using this model can improve the relevancy of the attribute extraction process by 33.2% compared to the current highly-optimized production system. Apart from the empirical contribution, the highly generalized nature of the framework presented in this paper makes it relevant for many high-volume search-driven websites.
在线市场使用属性值对,如品牌、尺寸、尺寸类型、颜色等,来帮助定义关于商品列表的重要和相关事实。这有助于买家使用属性过滤来管理他们的搜索结果,并创造更丰富的体验。尽管它们对商品的可发现性至关重要,但让卖家为每个商品输入数十个不同的属性值对成本很高,而且往往会导致信息缺失。当买家根据属性值进行过滤时,这可能会导致从搜索结果中不必要地删除相关列表。在本文中,我们演示了使用文本到文本分层多标签排名模型框架来预测每个列表最相关的属性,以及它们的期望值,使用历史用户行为数据。这个解决方案可以帮助卖家专注于验证买家可能使用的属性信息,从而提高他们的清单的预期召回率。具体到eBay的案例,我们表明,与目前高度优化的生产系统相比,使用该模型可以将属性提取过程的相关性提高33.2%。除了经验贡献之外,本文中提出的框架的高度一般化性质使其与许多高容量搜索驱动的网站相关。
{"title":"Product Titles-to-Attributes As a Text-to-Text Task","authors":"Gilad Fuchs, Yoni Acriche","doi":"10.18653/v1/2022.ecnlp-1.12","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.12","url":null,"abstract":"Online marketplaces use attribute-value pairs, such as brand, size, size type, color, etc. to help define important and relevant facts about a listing. These help buyers to curate their search results using attribute filtering and overall create a richer experience. Although their critical importance for listings’ discoverability, getting sellers to input tens of different attribute-value pairs per listing is costly and often results in missing information. This can later translate to the unnecessary removal of relevant listings from the search results when buyers are filtering by attribute values. In this paper we demonstrate using a Text-to-Text hierarchical multi-label ranking model framework to predict the most relevant attributes per listing, along with their expected values, using historic user behavioral data. This solution helps sellers by allowing them to focus on verifying information on attributes that are likely to be used by buyers, and thus, increase the expected recall for their listings. Specifically for eBay’s case we show that using this model can improve the relevancy of the attribute extraction process by 33.2% compared to the current highly-optimized production system. Apart from the empirical contribution, the highly generalized nature of the framework presented in this paper makes it relevant for many high-volume search-driven websites.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134342905","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}
引用次数: 1
Utilizing Cross-Modal Contrastive Learning to Improve Item Categorization BERT Model 利用跨模态对比学习改进项目分类BERT模型
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.25
L. Chen, Houwei Chou
Item categorization (IC) is a core natural language processing (NLP) task in e-commerce. As a special text classification task, fine-tuning pre-trained models, e.g., BERT, has become a mainstream solution. To improve IC performance further, other product metadata, e.g., product images, have been used. Although multimodal IC (MIC) systems show higher performance, expanding from processing text to more resource-demanding images brings large engineering impacts and hinders the deployment of such dual-input MIC systems. In this paper, we proposed a new way of using product images to improve text-only IC model: leveraging cross-modal signals between products’ titles and associated images to adapt BERT models in a self-supervised learning (SSL) way. Our experiments on the three genres in the public Amazon product dataset show that the proposed method generates improved prediction accuracy and macro-F1 values than simply using the original BERT. Moreover, the proposed method is able to keep using existing text-only IC inference implementation and shows a resource advantage than the deployment of a dual-input MIC system.
商品分类是电子商务中自然语言处理(NLP)的核心任务。作为一种特殊的文本分类任务,对预训练模型(如BERT)进行微调已经成为主流的解决方案。为了进一步提高集成电路的性能,已经使用了其他产品元数据,例如产品图像。虽然多模态集成电路(MIC)系统表现出更高的性能,但从处理文本扩展到处理资源要求更高的图像会带来巨大的工程影响,并阻碍这种双输入MIC系统的部署。在本文中,我们提出了一种使用产品图像来改进纯文本IC模型的新方法:利用产品标题和相关图像之间的跨模态信号以自监督学习(SSL)的方式适应BERT模型。我们在亚马逊公共产品数据集中对三种类型进行的实验表明,与简单使用原始BERT相比,提出的方法产生了更高的预测精度和宏观f1值。此外,该方法能够继续使用现有的纯文本集成电路推理实现,并且比部署双输入集成电路系统具有资源优势。
{"title":"Utilizing Cross-Modal Contrastive Learning to Improve Item Categorization BERT Model","authors":"L. Chen, Houwei Chou","doi":"10.18653/v1/2022.ecnlp-1.25","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.25","url":null,"abstract":"Item categorization (IC) is a core natural language processing (NLP) task in e-commerce. As a special text classification task, fine-tuning pre-trained models, e.g., BERT, has become a mainstream solution. To improve IC performance further, other product metadata, e.g., product images, have been used. Although multimodal IC (MIC) systems show higher performance, expanding from processing text to more resource-demanding images brings large engineering impacts and hinders the deployment of such dual-input MIC systems. In this paper, we proposed a new way of using product images to improve text-only IC model: leveraging cross-modal signals between products’ titles and associated images to adapt BERT models in a self-supervised learning (SSL) way. Our experiments on the three genres in the public Amazon product dataset show that the proposed method generates improved prediction accuracy and macro-F1 values than simply using the original BERT. Moreover, the proposed method is able to keep using existing text-only IC inference implementation and shows a resource advantage than the deployment of a dual-input MIC system.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123465234","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}
引用次数: 1
Product Answer Generation from Heterogeneous Sources: A New Benchmark and Best Practices 从异构来源生成产品答案:一个新的基准和最佳实践
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.13
Xiaoyu Shen, Gianni Barlacchi, Marco Del Tredici, Weiwei Cheng, B. Byrne, A. Gispert
It is of great value to answer product questions based on heterogeneous information sources available on web product pages, e.g., semi-structured attributes, text descriptions, user-provided contents, etc. However, these sources have different structures and writing styles, which poses challenges for (1) evidence ranking, (2) source selection, and (3) answer generation. In this paper, we build a benchmark with annotations for both evidence selection and answer generation covering 6 information sources. Based on this benchmark, we conduct a comprehensive study and present a set of best practices. We show that all sources are important and contribute to answering questions. Handling all sources within one single model can produce comparable confidence scores across sources and combining multiple sources for training always helps, even for sources with totally different structures. We further propose a novel data augmentation method to iteratively create training samples for answer generation, which achieves close-to-human performance with only a few thousandannotations. Finally, we perform an in-depth error analysis of model predictions and highlight the challenges for future research.
基于web产品页面上可用的异构信息源(如半结构化属性、文本描述、用户提供的内容等)来回答产品问题具有很大的价值。然而,这些来源具有不同的结构和写作风格,这对(1)证据排序,(2)来源选择和(3)答案生成提出了挑战。在本文中,我们建立了一个包含6个信息源的证据选择和答案生成的带有注释的基准。基于这个基准,我们进行了全面的研究,并提出了一套最佳实践。我们表明,所有的来源都是重要的,有助于回答问题。处理单个模型中的所有源可以跨源产生可比较的置信度分数,并且组合多个源进行训练总是有帮助的,即使对于具有完全不同结构的源也是如此。我们进一步提出了一种新的数据增强方法来迭代地创建用于答案生成的训练样本,该方法只需要几千个注释就可以达到接近人类的性能。最后,我们对模型预测进行了深入的误差分析,并强调了未来研究的挑战。
{"title":"Product Answer Generation from Heterogeneous Sources: A New Benchmark and Best Practices","authors":"Xiaoyu Shen, Gianni Barlacchi, Marco Del Tredici, Weiwei Cheng, B. Byrne, A. Gispert","doi":"10.18653/v1/2022.ecnlp-1.13","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.13","url":null,"abstract":"It is of great value to answer product questions based on heterogeneous information sources available on web product pages, e.g., semi-structured attributes, text descriptions, user-provided contents, etc. However, these sources have different structures and writing styles, which poses challenges for (1) evidence ranking, (2) source selection, and (3) answer generation. In this paper, we build a benchmark with annotations for both evidence selection and answer generation covering 6 information sources. Based on this benchmark, we conduct a comprehensive study and present a set of best practices. We show that all sources are important and contribute to answering questions. Handling all sources within one single model can produce comparable confidence scores across sources and combining multiple sources for training always helps, even for sources with totally different structures. We further propose a novel data augmentation method to iteratively create training samples for answer generation, which achieves close-to-human performance with only a few thousandannotations. Finally, we perform an in-depth error analysis of model predictions and highlight the challenges for future research.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124157004","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}
引用次数: 10
OpenBrand: Open Brand Value Extraction from Product Descriptions OpenBrand:从产品描述中提取开放式品牌价值
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.19
Kassem Sabeh, Mouna Kacimi, J. Gamper
Extracting attribute-value information from unstructured product descriptions continue to be of a vital importance in e-commerce applications. One of the most important product attributes is the brand which highly influences costumers’ purchasing behaviour. Thus, it is crucial to accurately extract brand information dealing with the main challenge of discovering new brand names. Under the open world assumption, several approaches have adopted deep learning models to extract attribute-values using sequence tagging paradigm. However, they did not employ finer grained data representations such as character level embeddings which improve generalizability. In this paper, we introduce OpenBrand, a novel approach for discovering brand names. OpenBrand is a BiLSTM-CRF-Attention model with embeddings at different granularities. Such embeddings are learned using CNN and LSTM architectures to provide more accurate representations. We further propose a new dataset for brand value extraction, with a very challenging task on zero-shot extraction. We have tested our approach, through extensive experiments, and shown that it outperforms state-of-the-art models in brand name discovery.
从非结构化产品描述中提取属性值信息在电子商务应用中仍然是至关重要的。品牌是产品最重要的属性之一,它对消费者的购买行为有很大的影响。因此,准确提取品牌信息处理发现新品牌名称的主要挑战是至关重要的。在开放世界假设下,有几种方法采用深度学习模型,利用序列标记范式提取属性值。然而,他们没有采用更细粒度的数据表示,如字符级嵌入,这可以提高通用性。在本文中,我们介绍了OpenBrand,一种发现品牌名称的新方法。OpenBrand是一个具有不同粒度嵌入的BiLSTM-CRF-Attention模型。这种嵌入是使用CNN和LSTM架构来学习的,以提供更准确的表示。我们进一步提出了一个新的品牌价值提取数据集,其中零采样提取是一个非常具有挑战性的任务。我们已经通过大量的实验测试了我们的方法,并表明它在品牌发现方面优于最先进的模型。
{"title":"OpenBrand: Open Brand Value Extraction from Product Descriptions","authors":"Kassem Sabeh, Mouna Kacimi, J. Gamper","doi":"10.18653/v1/2022.ecnlp-1.19","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.19","url":null,"abstract":"Extracting attribute-value information from unstructured product descriptions continue to be of a vital importance in e-commerce applications. One of the most important product attributes is the brand which highly influences costumers’ purchasing behaviour. Thus, it is crucial to accurately extract brand information dealing with the main challenge of discovering new brand names. Under the open world assumption, several approaches have adopted deep learning models to extract attribute-values using sequence tagging paradigm. However, they did not employ finer grained data representations such as character level embeddings which improve generalizability. In this paper, we introduce OpenBrand, a novel approach for discovering brand names. OpenBrand is a BiLSTM-CRF-Attention model with embeddings at different granularities. Such embeddings are learned using CNN and LSTM architectures to provide more accurate representations. We further propose a new dataset for brand value extraction, with a very challenging task on zero-shot extraction. We have tested our approach, through extensive experiments, and shown that it outperforms state-of-the-art models in brand name discovery.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124200142","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}
引用次数: 1
Domain-specific knowledge distillation yields smaller and better models for conversational commerce 特定领域的知识精馏为会话式商务产生更小、更好的模型
Pub Date : 1900-01-01 DOI: 10.18653/v1/2022.ecnlp-1.18
Kristen Howell, Jian Wang, Akshay Hazare, Joe Bradley, Chris Brew, Xi Chen, Matthew Dunn, Beth-Ann Hockey, Andrew Maurer, D. Widdows
We demonstrate that knowledge distillation can be used not only to reduce model size, but to simultaneously adapt a contextual language model to a specific domain. We use Multilingual BERT (mBERT; Devlin et al., 2019) as a starting point and follow the knowledge distillation approach of (Sahn et al., 2019) to train a smaller multilingual BERT model that is adapted to the domain at hand. We show that for in-domain tasks, the domain-specific model shows on average 2.3% improvement in F1 score, relative to a model distilled on domain-general data. Whereas much previous work with BERT has fine-tuned the encoder weights during task training, we show that the model improvements from distillation on in-domain data persist even when the encoder weights are frozen during task training, allowing a single encoder to support classifiers for multiple tasks and languages.
我们证明了知识蒸馏不仅可以用来减少模型的大小,而且可以同时使上下文语言模型适应特定的领域。我们使用多语言BERT (mBERT;Devlin等人,2019)作为起点,并遵循(Sahn等人,2019)的知识蒸馏方法来训练适应手头领域的较小的多语言BERT模型。我们表明,对于领域内任务,领域特定模型的F1分数平均提高了2.3%,相对于在领域通用数据上提炼的模型。尽管BERT之前的许多工作在任务训练期间对编码器权重进行了微调,但我们表明,即使在任务训练期间编码器权重冻结时,对域内数据进行蒸馏的模型改进仍然存在,从而允许单个编码器支持多个任务和语言的分类器。
{"title":"Domain-specific knowledge distillation yields smaller and better models for conversational commerce","authors":"Kristen Howell, Jian Wang, Akshay Hazare, Joe Bradley, Chris Brew, Xi Chen, Matthew Dunn, Beth-Ann Hockey, Andrew Maurer, D. Widdows","doi":"10.18653/v1/2022.ecnlp-1.18","DOIUrl":"https://doi.org/10.18653/v1/2022.ecnlp-1.18","url":null,"abstract":"We demonstrate that knowledge distillation can be used not only to reduce model size, but to simultaneously adapt a contextual language model to a specific domain. We use Multilingual BERT (mBERT; Devlin et al., 2019) as a starting point and follow the knowledge distillation approach of (Sahn et al., 2019) to train a smaller multilingual BERT model that is adapted to the domain at hand. We show that for in-domain tasks, the domain-specific model shows on average 2.3% improvement in F1 score, relative to a model distilled on domain-general data. Whereas much previous work with BERT has fine-tuned the encoder weights during task training, we show that the model improvements from distillation on in-domain data persist even when the encoder weights are frozen during task training, allowing a single encoder to support classifiers for multiple tasks and languages.","PeriodicalId":384006,"journal":{"name":"Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127545313","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}
引用次数: 2
期刊
Proceedings of The Fifth Workshop on e-Commerce and NLP (ECNLP 5)
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1