Question Answering (QA) returns concise answers or answer lists from natural language text given a context document. Many resources go into curating QA datasets to advance robust models' development. There is a surge of QA datasets for languages like English, however, this is not true for Amharic. Amharic, the official language of Ethiopia, is the second most spoken Semitic language in the world. There is no published or publicly available Amharic QA dataset. Hence, to foster the research in Amharic QA, we present the first Amharic QA (AmQA) dataset. We crowdsourced 2628 question-answer pairs over 378 Wikipedia articles. Additionally, we run an XLMR Large-based baseline model to spark open-domain QA research interest. The best-performing baseline achieves an F-score of 69.58 and 71.74 in reader-retriever QA and reading comprehension settings respectively.
Knowledge graphs are increasingly used in a plethora of downstream tasks or in the augmentation of statistical models to improve factuality. However, social biases are engraved in these representations and propagate downstream. We conducted a critical analysis of literature concerning biases at different steps of a knowledge graph lifecycle. We investigated factors introducing bias, as well as the biases that are rendered by knowledge graphs and their embedded versions afterward. Limitations of existing measurement and mitigation strategies are discussed and paths forward are proposed.
Social media platforms provide a continuous stream of real-time news regarding crisis events on a global scale. Several machine learning methods utilize the crowd-sourced data for the automated detection of crises and the characterization of their precursors and aftermaths. Early detection and localization of crisis-related events can help save lives and economies. Yet, the applied automation methods introduce ethical risks worthy of investigation - especially given their high-stakes societal context. This work identifies and critically examines ethical risk factors of social media analyses of crisis events focusing on machine learning methods. We aim to sensitize researchers and practitioners to the ethical pitfalls and promote fairer and more reliable designs.
Existing approaches on Question Answering over Knowledge Graphs (KGQA) have weak generalizability. That is often due to the standard i.i.d. assumption on the underlying dataset. Recently, three levels of generalization for KGQA were defined, namely i.i.d., compositional, zero-shot. We analyze 25 well-known KGQA datasets for 5 different Knowledge Graphs (KGs). We show that according to this definition many existing and online available KGQA datasets are either not suited to train a generalizable KGQA system or that the datasets are based on discontinued and out-dated KGs. Generating new datasets is a costly process and, thus, is not an alternative to smaller research groups and companies. In this work, we propose a mitigation method for re-splitting available KGQA datasets to enable their applicability to evaluate generalization, without any cost and manual effort. We test our hypothesis on three KGQA datasets, i.e., LC-QuAD, LC-QuAD 2.0 and QALD-9). Experiments on re-splitted KGQA datasets demonstrate its effectiveness towards generalizability. The code and a unified way to access 18 available datasets is online at https://github.com/semantic-systems/KGQA-datasets as well as https://github.com/semantic-systems/KGQA-datasets-generalization.
In this work, we focus on the task of generating SPARQL queries from natural language questions, which can then be executed on Knowledge Graphs (KGs). We assume that gold entity and relations have been provided, and the remaining task is to arrange them in the right order along with SPARQL vocabulary, and input tokens to produce the correct SPARQL query. Pre-trained Language Models (PLMs) have not been explored in depth on this task so far, so we experiment with BART, T5 and PGNs (Pointer Generator Networks) with BERT embeddings, looking for new baselines in the PLM era for this task, on DBpedia and Wikidata KGs. We show that T5 requires special input tokenisation, but produces state of the art performance on LC-QuAD 1.0 and LC-QuAD 2.0 datasets, and outperforms task-specific models from previous works. Moreover, the methods enable semantic parsing for questions where a part of the input needs to be copied to the output query, thus enabling a new paradigm in KG semantic parsing.
Task-oriented dialogue generation is challenging since the underlying knowledge is often dynamic and effectively incorporating knowledge into the learning process is hard. It is particularly challenging to generate both human-like and informative responses in this setting. Recent research primarily focused on various knowledge distillation methods where the underlying relationship between the facts in a knowledge base is not effectively captured. In this paper, we go one step further and demonstrate how the structural information of a knowledge graph can improve the system's inference capabilities. Specifically, we propose DialoKG, a novel task-oriented dialogue system that effectively incorporates knowledge into a language model. Our proposed system views relational knowledge as a knowledge graph and introduces (1) a structure-aware knowledge embedding technique, and (2) a knowledge graph-weighted attention masking strategy to facilitate the system selecting relevant information during the dialogue generation. An empirical evaluation demonstrates the effectiveness of DialoKG over state-of-the-art methods on several standard benchmark datasets.
Evaluating Natural Language Generation (NLG) systems is a challenging task. Firstly, the metric should ensure that the generated hypothesis reflects the reference's semantics. Secondly, it should consider the grammatical quality of the generated sentence. Thirdly, it should be robust enough to handle various surface forms of the generated sentence. Thus, an effective evaluation metric has to be multifaceted. In this paper, we propose an automatic evaluation metric incorporating several core aspects of natural language understanding (language competence, syntactic and semantic variation). Our proposed metric, RoMe, is trained on language features such as semantic similarity combined with tree edit distance and grammatical acceptability, using a self-supervised neural network to assess the overall quality of the generated sentence. Moreover, we perform an extensive robustness analysis of the state-of-the-art methods and RoMe. Empirical results suggest that RoMe has a stronger correlation to human judgment over state-of-the-art metrics in evaluating system-generated sentences across several NLG tasks.
The ability to have the same experience for different user groups (i.e., accessibility) is one of the most important characteristics of Web-based systems. The same is true for Knowledge Graph Question Answering (KGQA) systems that provide the access to Semantic Web data via natural language interface. While following our research agenda on the multilingual aspect of accessibility of KGQA systems, we identified several ongoing challenges. One of them is the lack of multilingual KGQA benchmarks. In this work, we extend one of the most popular KGQA benchmarks - QALD-9 by introducing high-quality questions' translations to 8 languages provided by native speakers, and transferring the SPARQL queries of QALD-9 from DBpedia to Wikidata, s.t., the usability and relevance of the dataset is strongly increased. Five of the languages - Armenian, Ukrainian, Lithuanian, Bashkir and Belarusian - to our best knowledge were never considered in KGQA research community before. The latter two of the languages are considered as "endangered" by UNESCO. We call the extended dataset QALD-9-plus and made it available online https://github.com/Perevalov/qald_9_plus.
Data-driven systems need to be evaluated to establish trust in the scientific approach and its applicability. In particular, this is true for Knowledge Graph (KG) Question Answering (QA), where complex data structures are made accessible via natural-language interfaces. Evaluating the capabilities of these systems has been a driver for the community for more than ten years while establishing different KGQA benchmark datasets. However, comparing different approaches is cumbersome. The lack of existing and curated leaderboards leads to a missing global view over the research field and could inject mistrust into the results. In particular, the latest and most-used datasets in the KGQA community, LC-QuAD and QALD, miss providing central and up-to-date points of trust. In this paper, we survey and analyze a wide range of evaluation results with significant coverage of 100 publications and 98 systems from the last decade. We provide a new central and open leaderboard for any KGQA benchmark dataset as a focal point for the community - https://kgqa.github.io/leaderboard. Our analysis highlights existing problems during the evaluation of KGQA systems. Thus, we will point to possible improvements for future evaluations.
Each year the International Semantic Web Conference organizes a set of Semantic Web Challenges to establish competitions that will advance state-of-the-art solutions in some problem domains. The Semantic Answer Type and Relation Prediction Task (SMART) task is one of the ISWC 2021 Semantic Web challenges. This is the second year of the challenge after a successful SMART 2020 at ISWC 2020. This year's version focuses on two sub-tasks that are very important to Knowledge Base Question Answering (KBQA): Answer Type Prediction and Relation Prediction. Question type and answer type prediction can play a key role in knowledge base question answering systems providing insights about the expected answer that are helpful to generate correct queries or rank the answer candidates. More concretely, given a question in natural language, the first task is, to predict the answer type using a target ontology (e.g., DBpedia or Wikidata. Similarly, the second task is to identify relations in the natural language query and link them to the relations in a target ontology. This paper discusses the task descriptions, benchmark datasets, and evaluation metrics. For more information, please visit https://smart-task.github.io/2021/.