Typically available dialogue frameworks have adopted a semantic
representation based on dialogue-acts and slot-value pairs. Despite its
simplicity, this representation has disadvantages such as the lack of
expressivity, scalability and explainability. We present WEBDial: a dialogue
framework that relies on a graph formalism by using RDF triples instead of
slot-value pairs. We describe its overall architecture and the graph-based
semantic representation. We show its applicability from simple to complex
applications, by varying the complexity of domains and tasks: from single
domain and tasks to multiple domains and complex tasks.

WEBDial 是一个对话框架，它采用 RDF 三元组而非槽值对的图形形式进行语义表示，具有应用广泛性和可扩展性，适用于简单到复杂的应用。

WEBDial，一种多领域、多任务的统计对话框架与 RDF

WEBDial, a Multi-domain, Multitask Statistical Dialogue Framework with  RDF

The task of verbalization of RDF triples has known a growth in popularity due
to the rising ubiquity of Knowledge Bases (KBs). The formalism of RDF triples
is a simple and efficient way to store facts at a large scale. However, its
abstract representation makes it difficult for humans to interpret. For this
purpose, the WebNLG challenge aims at promoting automated RDF-to-text
generation. We propose to leverage pre-trainings from augmented data with the
Transformer model using a data augmentation strategy. Our experiment results
show a minimum relative increases of 3.73%, 126.05% and 88.16% in BLEU score
for seen categories, unseen entities and unseen categories respectively over
the standard training.

本研究旨在通过使用数据增强策略，利用 Transformer 模型的预训练技术来促进自动 RDF 到文本的生成。结果表明，相较于标准训练，已知类别、未知实体和未知类别的 BLEU 分数相对增加了 3.73%、126.05% 和 88.16%。

使用去噪预训练和数据增强策略提升 RDF 变换器的口头表述

Denoising Pre-Training and Data Augmentation Strategies for Enhanced RDF  Verbalization with Transformers

The Freebase knowledge base was a significant Semantic Web and linked data
technology during its years of operations since 2007. Following its acquisition
by Google in 2010 and its shutdown in 2016, Freebase data is contained in a
data dump of billions of RDF triples. In this research, an exploration of the
Freebase data dumps will show best practices in understanding and using the
Freebase data and also present a general methodology for parsing the linked
data. The analysis is done with limited computing resources and the use of
open-source Unix-like tools. The results showcase the efficiency of the
technique and highlight redundancies in the data, with the possibility of
restructuring nearly 60% of the original data. As an archival dataset that has
not changed since 2015, Freebase's semantic structured data has applications in
other prominent fields, such as information retrieval (IR) and knowledge-based
question answering (KBQA). Freebase can also serve as a gateway to other
structured datasets, such as DBpedia, Wikidata, and YAGO.

该研究旨在探索 Freebase 数据的最佳实践和一般方法，以便于解析链接数据，并展示在限制计算资源的情况下使用 Unix 类开源工具处理数据的效率。结果显示出数据的冗余性，其重结构的可能性高达原数据集的 60％。Freebase 的语义结构化数据用于信息检索（IR）和基于知识的问答（KBQA）等其他核心领域，并可作为其他结构化数据集（如 DBpedia，Wikidata 和 YAGO）的入口。