Search

Question Answering (QA) is a field of study addressed to develop automatic methods for answering questions expressed in natural language. Recently, the emergence of the new generation of intelligent assistants, such as Siri, Alexa, and Google Assistant, has intensified the importance of an effective and efficient QA system able to handle questions with different complexities. Regarding the type of question to be answered, QA systems have been divided into two sub-areas: (i) factoid questions that require a single fact – e.g., a name of a person or a date, and (ii) non-factoid questions that need a more complex answer – e.g., descriptions, opinions, or explanations. While factoid QA systems have overcome human performance on some benchmarks, automatic systems for answering non-factoid questions remain a challenge and an open research problem. This work provides an overview of recent research addressing non-factoid questions. It focuses on which methods have been applied in each task, the data sets available, challenges and limitations, and possible research directions. From a total of 455 recent studies, we selected 75 papers based on our quality control system and exclusion criteria for an in-depth analysis. This systematic review helped to answer what are the tasks and methods involved in non-factoid, what are the data sets available, what the limitations are, and what is the recommendations for future research.

A dramatic foreign aid policy development in the US, the proclamation of the President's Emergency Plan for AIDS Relief (PEPFAR), has been ascribed to the perception of a security threat. The securitizing scheme, however, has been debunked as a factoid not supported by empirical evidence. Despite the flaws of securitization as a causal theory, alternative explanations remain unexplored. This research seeks to address an alternative explanation, "moralization." The article begins with a brief history of changes in the US response to the global AIDS crisis since the 1980s, followed by a description of securitization theory as an alleged explanation for the aid increase in the early 2000s. Next, this research problematizes the AIDS-security nexus and proceeds to show how a humanitarian approach to the epidemic invoked a moral obligation to respond urgently to the catastrophic epidemic. The moralizing rationalization entails the explication of perception changes among significant individuals within and outside of policy circles from the theoretical framework of agenda setting, namely the garbage can model. The research closes with the theoretical and empirical implications for the international relations discipline and foreign policy analysis, as well as for the current health crisis caused by COVID-19.

Why-type non-factoid questions are complex and difficult to answer compared to factoid questions. A challenge in finding an accurate answer to a non-factoid question is to understand the intent of user as it differs with their knowledge and also the context of the question in which it is being asked. Predicting correct type of a question and its answer by a classification model is an important issue as it affects the subsequent processing of its answer. In this paper, a classification model is proposed which is trained by a combination of lexical, syntactic, and semantic features to classify open-domain why-type questions. Various supervised classifiers are trained on a featured dataset out of which support vector machine achieves the highest accuracy of 81% in determining question type and 76.8% in determining answer type which shows 14.6% improvement in predicting an answer type over a baseline why-type classifier with 62.2% accuracy.

As a member of the team that created elements 104 and 105 at Lawrence Berkeley Laboratory, James Andrew Harris [1932–2000] was the first African American credited in the discovery of an element. This factoid has been posted on social media, used in a quiz game, and repeated on numerous Web sites. The story (if any context is offered at all) is often the same—a narrative beginning with prejudice, which is overcome with perseverance, and a discovery as pay-off. It is a good story, and this article does not doubt its veracity. But there are questions that need asking. These questions can be basic: what role did Harris play in the discovery of these elements? Or, they can be complicated: what did Harris do before or after these discoveries? At present, there is a surprising lack of detailed information about James Harris’s life and career that is easily discoverable. This article is both a call to chemistry students and educators to help uncover more of Harris’s story as well as an example of how three simple guidelines can aid the discovery process.

This article offers a historiographical examination of how 20th-century ideas of assimilation and cultural purity have shaped our understanding of Bronze Age Anatolia, focusing on the canonical narrative of Assyrian presence at the site of Kültepe-Kaneš. According to this narrative, Old Assyrian merchants who lived and conducted business at Kaneš from the early 20th to the late 18th century BC left no trace in the archaeological record except for cuneiform tablets and cylinder seals, assimilating to local culture to such a degree that Kültepe’s archaeological record is entirely of Anatolian character. The accuracy of this view has met increasing circumspection in recent years. What remains to be articulated is why it remained unchallenged for so long, from its initial formulation in 1948 until the late 2000s, during which time it was widely repeated and reiterated. It is proposed here that the persistence and longevity of what is essentially a misconstrued notion of foreign (in)visibility in Kültepe’s material record can be explained by treating it as a ‘factoid’. The article first historicises the factoid’s formulation and subsequent development. This is followed by a critical evaluation of the evidentiary bases of the factoid to show how disciplinary tendencies to privilege certain categories of evidence over others have created artificial gaps in the data. Ultimately, the article seeks to highlight the epistemological implications of how one of the key sites of Bronze Age Anatolia came to represent a perceived rather than an observed case of indigenous cultural purity.

Factoid Question-Answering (QA) Systems were developed to provide an accurate answer to a factoid question expressed in a natural language. The prime knowledge resources for most of factoid QA systems are online databases. However, the unstructured information in these resources rises the complexity of the information retrieval task. In this paper, we aim to develop an Arabic QA system for factoid questions specialized in Islamic sciences as prophetic tradition (Hadith), Hadith narrator and Quran interpretation (Tafsir). In fact, many questions in the Islamic research fields are focusing on Tafsir and Hadith text. In addition, a number of those interrogations focus on the chain of narrators who transmitted the Hadith. Besides, analyzing the narrator profile is considered an important and enquired task in hadith science. Furthermore, the proposed QA system is based on a normalized database specified in Text Encoding Initiative (TEI) standard. To achieve this, we propose a method composed of three phases to retrieve an accurate answer for the user question. The first phase is the question analysis. The second is the information search. The third phase is the answer processing. Besides, we implant the proposed QA system with a graphic interface allowing the interaction with the user. Finally, we experiment our prototype on 100 questions in the theme of Hadith, narrator and Tafsir text. The QA system succeeds to generate accurate responses for 92% of the entered questions.

With this research, it is aimed to evaluate the performance of the Google search engine on the searches made with Turkish natural sentences. In this direction, Google was asked 20 questions of different types. The first 10 questions are comprised factoid questions with specific and concise answers. On the other hand, the other 10 questions are comprised of list questions with long answers or complex questions with no specific answer. Within the scope of the research, first 10 answers brought up by the Google search engine were evaluated. After the documents retrieved during the study were classified as relevant and irrelevant, precision and general normalized recall measure of the results were calculated. Whether the precision and general normalized recall measure differ depending on the question types, the difference between the first 5 results and the first 10 results, and the relevance level of the snippets and the videos brought on the result page were examined. Mann-Whitney U, Spearman and Chi-square tests were used in the analysis of the data. As a result of the research, the relevance level of the results brought by the Google search engine in regards to Turkish natural language searches were found to be average. It was revealed that the relevance levels did not differ according to the question type, but the relevance ranking did differ. Compared to other questions, relevant documents ranked higher for factoid questions. In addition, Google has been successful in showing relevant documents at the top. It turned out that mostly in the first five results relevant results can be reached. A snippet was brought for almost half of the questions and the majority of these were determined to be relevant with the query. Snippets are available not only in factoid questions but also in list questions. Videos brought up in the result pages were revealed to be mostly irrelevant to the query. Videos for factoid questions are less relevant than other questions. It is thought that the results of the research will benefit the work carried out to improve the Google search engine. It is recommended to increase the performance tests made with real user queries.

Searching for a specific and meaningful piece of information in the humongous textual data volumes found on the internet and knowledge repositories is a very challenging task. This problem is usually constrained to answering simple, factoid questions by resorting to a question answering (QA) system built on top of complex approaches such as heuristics, information retrieval, and machine learning. More precisely, deep learning methods became into sharp focus of this research field because such purposes can realize the benefits of the vast amounts of data to boost the practical results of QA systems. In this paper, we present a systematic survey on deep learning-based QA systems concerning factoid questions, with particular focus on how each existing system addresses their critical features in terms of learning end-to-end models. We also detail the evaluation process carried out on these systems and discuss how each approach differs from the others in terms of the challenges tackled and the strategies employed. Finally, we present the most prominent research problems still open in the field.

We study the problem of non-factoid QA on instructional videos. Existing work focuses either on visual or textual modality of video content, to find matching answers to the question. However, neither is flexible enough for our problem setting of non-factoid answers with varying lengths. Motivated by this, we propose a two-stage model: (a) multimodal segmentation of video into span candidates and (b) length-adaptive ranking of the candidates to the question. First, for segmentation, we propose Segmenter for generating span candidates of diverse length, considering both textual and visual modality. Second, for ranking, we propose Ranker to score the candidates, dynamically combining the two models with complementary strength for both short and long spans respectively. Experimental result demonstrates that our model achieves state-of-the-art performance.

The recent explosion in question answering research produced a wealth of both factoid reading comprehension (RC) and commonsense reasoning datasets. Combining them presents a different kind of task: deciding not simply whether information is present in the text, but also whether a confident guess could be made for the missing information. We present QuAIL, the first RC dataset to combine text-based, world knowledge and unanswerable questions, and to provide question type annotation that would enable diagnostics of the reasoning strategies by a given QA system. QuAIL contains 15K multi-choice questions for 800 texts in 4 domains. Crucially, it offers both general and text-specific questions, unlikely to be found in pretraining data. We show that QuAIL poses substantial challenges to the current state-of-the-art systems, with a 30% drop in accuracy compared to the most similar existing dataset.

Biomedical text mining is becoming increasingly important as the number of biomedical documents grow rapidly. Deep learning has boosted the development of biomedical text mining models. However, as deep learning models require a large amount of training data, a hierarchical attention based transfer learning model is proposed in this paper for the question answering task in biomedical field which lacks of sufficient training data. We adopt BERT (Bidirectional Encoder Representation Transformers), which has the ability to learn from large-scale unsupervised data, to enrich the semantic representation in our model. Especially, the scaled dot-product attention mechanism captures the question interaction clues for passage encoding. The domain adaptation technique of fine-tuning is used to reinforce the performance, which penalizes the deviations from the source model’s parameters and remembers the knowledge of source domain. We evaluate the system performance on the open data set of BioASQ-Task B. The results show that our system achieves the state-of-the-art performance without any handcrafted features and outperforms the best solution for factoid questions in 2016 and 2017 BioASQ-Task B.

Considering the widespread use of mobile and voice search,answer passage retrieval for non-factoid questions plays a critical rolein modern information retrieval systems. Despite the importance of thetask, the community still feels the significant lack of large-scale non-factoid question answering collections with real questions and comprehensive relevance judgments. In this paper, we develop and release a collection of 2,626 open-domain non-factoid questions from a diverse set ofcategories. The dataset, called ANTIQUE, contains 34k manual relevanceannotations. The questions were asked by real users in a community question answering service, i.e., Yahoo! Answers. Relevance judgments for allthe answers to each question were collected through crowdsourcing. Tofacilitate further research, we also include a brief analysis of the data aswell as baseline results on both classical and neural IR models.

In this article, we propose a unified deep neural network framework for multilingual question answering (QA). The proposed network deals with the multilingual questions and answers snippets. The input to the network is a pair of factoid question and snippet in the multilingual environment (English and Hindi), and output is the relevant answer from the snippet. We begin by generating the snippet using a graph-based language-independent algorithm, which exploits the lexico-semantic similarity between the sentences. The soft alignment of the question words from the English and Hindi languages has been used to learn the shared representation of the question. The learned shared representation of question and attention-based snippet representation are passed as an input to the answer extraction layer of the network, which extracts the answer span from the snippet. Evaluation on a standard multilingual QA dataset shows the state-of-the-art performance with 39.44 Exact Match (EM) and 44.97 F1 values. Similarly, we achieve the performance of 50.11 Exact Match (EM) and 53.77 F1 values on Translated SQuAD dataset.

The task of reasoning over knowledge graph for factoid questions has received significant interest from the research community of natural language processing. Performing this task inevitably faces the issues of question complexity and reasoning efficiency. In this paper, we investigate modern reasoning approaches over knowledge graph to tackle complex factoid questions of diverse reasoning schemas with attractive speedup in computational efficiency. To this end, we propose two evidence retrieval strategies to generate concise and informative evidence graph of high semantic-relevance and factual coverage to the question. Then, we adopt DELFT, a graph neural networks based framework that takes the linguistic structure representation of a question and the evidence graph as input, to predict the answer by reasoning over the evidence graph. We evaluate the performance across several baselines in terms of effectiveness and efficiency on two real-world datasets, MOOCQA and MetaQA. The results show the superiority of message passing paradigm in delivering a robust reasoner with better answer quality and significantly improved computational efficiency.

Over the last years, the use and the need for automated question answering systems have become more important than ever. The main reasons for this relate to the constant increase of the information that is available in textual form as well as the need to facilitate users in getting information they seek in a precise, fast, and easy way. In this work, we deal with the problem of the open domain factoid-based answering of questions, where the answer to a question is in the form of a word or a small phrase. We present an efficient methodology for analyzing questions and specifying proper answers with the use of knowledge bases. Initially, given a specific question that a user sets, a high-quality linguistic analysis of the question is performed. The dependencies of the question are specified, and the main triplets of the question are created. The system creates a SPARQL query to get the right answer for the question via the API from the Wikidata Query Service. The results are quite encouraging and highlight the quite good performance of our methodology.

The rise in popularity of mobile and voice search has led to a shift in focus from document retrieval to short answer passage retrieval for non-factoid questions. Some of the questions have multiple answers, and the aim is to retrieve a set of relevant answer passages, which covers all these alternatives. Compared to documents, answers are more specific and typically form more defined types or groups. Grouping answer passages based on strong similarity measures may provide a means of identifying these types. Typically, kNN clustering in combination with term-based representations have been used in Information Retrieval (IR) scenarios. An alternate method is to use pre-trained distributional representations such as GloVe and BERT, which capture additional semantic relationships. The recent success of trained neural models for various tasks provides the motivation for generating more task-specific representations. However, due to the absence of large datasets for incorporating passage level similarity information, a more feasible alternative is to use weak supervision based training. This information can then be used to generate a final ranked list of diversified answers using standard diversification algorithms. In this paper, we introduce a new dataset NFPassageQA_Sim, with human annotated similarity labels for pairs of answer passages corresponding to each question. These similarity labels are then processed to generate another dataset NFPassageQA_Div, which consists of answer types for these questions. Using the similarity labels, we demonstrate the effectiveness of using weak supervision signals derived from GloVe, fine-tuned and trained using a BERT model for the task of answer passage clustering. Finally, we introduce a model which incorporates these clusters into a MMR (Maximal Marginal Relevance) model, which significantly beats other diversification baselines using both diversity and relevance metrics.

Biomedical factoid question answering is an important task in biomedical question answering applications. It has attracted much attention because of its reliability. In question answering systems, better representation of words is of great importance, and proper word embedding can significantly improve the performance of the system. With the success of pretrained models in general natural language processing tasks, pretrained models have been widely used in biomedical areas, and many pretrained model-based approaches have been proven effective in biomedical question-answering tasks. In addition to proper word embedding, name entities also provide important information for biomedical question answering. Inspired by the concept of transfer learning, in this study, we developed a mechanism to fine-tune BioBERT with a named entity dataset to improve the question answering performance. Furthermore, we applied BiLSTM to encode the question text to obtain sentence-level information. To better combine the question level and token level information, we use bagging to further improve the overall performance. The proposed framework was evaluated on BioASQ 6b and 7b datasets, and the results have shown that our proposed framework can outperform all baselines.

Background Biomedical question answering (QA) is a sub-task of natural language processing in a specific domain, which aims to answer a question in the biomedical field based on one or more related passages and can provide people with accurate healthcare-related information. Recently, a lot of approaches based on the neural network and large scale pre-trained language model have largely improved its performance. However, considering the lexical characteristics of biomedical corpus and its small scale dataset, there is still much improvement room for biomedical QA tasks. Results Inspired by the importance of syntactic and lexical features in the biomedical corpus, we proposed a new framework to extract external features, such as part-of-speech and named-entity recognition, and fused them with the original text representation encoded by pre-trained language model, to enhance the biomedical question answering performance. Our model achieves an overall improvement of all three metrics on BioASQ 6b, 7b, and 8b factoid question answering tasks. Conclusions The experiments on BioASQ question answering dataset demonstrated the effectiveness of our external feature-enriched framework. It is proven by the experiments conducted that external lexical and syntactic features can improve Pre-trained Language Model’s performance in biomedical domain question answering task.

The increased availability of information on the WWW has made it more challenging to select and filter information that we require. As a result, there is a greater demand for question-answering systems. Question Answering System responds to natural language-based questions from the user. However, traditionally, ranking the answers is based on the availability of the user's query terms in the candidate documents, which results in lower performance. The goal of this paper is to address the problem of ranking answer candidates for factoid-based questions. We improved the Answer Extraction by adding weight and several features and tokens of questions to determine the candidate's answer. We also enhanced the Question Analysis phase by joining phrase queries and boolean combination queries. Based on the results of our experiments, our proposed framework had the best performance of 54%.

La entrada de la sociedad en lo que se ha venido denominando como «era de la posverdad» –la cual viene alentada por la tecnología digital, los nuevos procesos comunicacionales, como las redes sociales en Internet, y el cambio del escenario político–, induce a muchos campos del conocimiento a trabajar desde los replanteamientos que esta propone. Las premisas en las que se funda conducen a reflexionar sobre la condición de lo falso como algo que transforma nuestra recepción de los mensajes que difunde ese entramado; traducción, interpretación y asimilación de una nueva realidad que, progresivamente, se inmiscuye e impone. Este movimiento inercial que, en ese sentido, se traslada al engranaje social, también modifica la tectónica epistemológica de la presente época y, en concreto, el campo de las artes como reflejo expresivo y crítico directo de las fluctuaciones que experimenta la colectividad. Profundizando en la condición de lo falso, podemos cerciorarnos de que esta supone una necesidad inherente a la representación artística. Asimismo, investigar sobre cómo el arte toma en cuenta los parámetros de posverdad que rigen la era actual, y varía su producción en relación a estos, nos lleva a poner el foco sobre la explicación de dos creaciones similares directamente relacionadas: fakes y factoids, como productos artísticos sintéticos adulterados, tergiversados, o distorsionados al extremo. Creaciones que representan la reverberación de una acelerada realidad que fluye interconectada y que enfatiza el borrado de cualquier delimitación útil para definir cada fenómeno acaecido en ese contexto. Sobre estas circunstancias, podemos señalar la importancia de los retos que, desde el espectro estudiado, lanza al público cierto arte activista. Espectadores quienes, bajo su «contrato de veridicción», serán los que establezcan un juicio crítico ante «verdades» que, posiblemente, no lo sean tanto.

We proposed the usage of dependency tree information to increase the accuracy of Indonesian factoid question answering. We employed MSTParser and Universal Dependency corpus to build the Indonesian dependency parser. The dependency tree information as the result of the Indonesian dependency parse is used in the answer finder component of Indonesian factoid question answering system. Here, we used dependency tree information in two ways: 1) as one of the features in machine learning based answer finder (classifying each term in the retrieved passage as part of a correct answer or not); 2) as an additional heuristic rule after conducting the machine learning technique. For the machine learning technique, we combined word based calculation, phrase based calculation and similarity dependency relation based calculation as the complete features. Using 203 data, we were able to enhance the accuracy for the Indonesian factoid QA system compared to related work by only using the phrase information. The best accuracy was 84.34% for the correct answer classification and the best MRR was 0.954.

Comparative questions in Chinese, as a special and complex form of question answering (QA), have their own unique sentence structure, existing methods cannot solve them well. Inspired by cognitive studies on how humans solve complex problems, we propose a hybrid framework which combines Logic Programming and attention based Bi-LSTM. This framework is decomposed into three consecutive components: 1) identify comparative questions, 2) extract comparative elements from the identified comparative questions, and 3) answer factoid questions containing the extracted comparative elements. Specifically, for the former two components, Logic Programming is adopted to filter out non-comparative questions and extract comparative elements. For the latter one, a bidirectional long and short term memory (Bi-LSTM) model with attention mechanism is utilized. Experimental results on Chinese geographical question datasets show that our proposed hybrid framework achieves outstanding performance for practical use.

Biomedical experts and bio-curators are unable to quickly find short and precise information using typical search engines as the amount of biomedical literature is increasing exponentially. The research community is focusing on biomedical question answering (QA) systems so that anyone can find precise information nuggets from the massive amount of biomedical literature. Generally, the user queries fall under different categories such as factoid, list, yes/no, or summary. The existing state-of-the-art question answering systems deal with most of these question types. However, the research to improve the performance of individual question types is also on the rise. To improve QA system performance, question classification plays a vital role for factoid and list type questions as it allows the answer processing stage to narrow down the candidate answer space and assigns a higher rank to the correct answers. A single biomedical answer or entity may be associated with more than one biomedical category or semantic type, e.g., Coenzyme Q(10) is classified under two categories in Unified Medical Language System (UMLS): organic chemical and biologically active substance. This inherent characteristic of biomedical entities makes question classification in the biomedical domain a multi-label classification problem, where one question might expect answers belonging to more than one semantic type. To the best of our knowledge, several QA systems deal with question classification as a multi-class classification problem and only one state-of-the-art system – OAQA – deals with it as a multi-label classification problem. In this paper, we analyze the pipeline of the OAQA system for factoid and list type questions, emphasizing the multi-label question classification. We use an improved question classification dataset with the copy transformation technique to improve the performance of list type questions. Moreover, we introduce a binary transformation in the pipeline of factoid questions to increase its performance. Our modified methodology enhances the performance of both list and factoid type questions by a margin of 2% and 3% evaluated on standard $F_{1}$ and Mean Reciprocal Rank measures, respectively.

In question-answering systems, question classification is a fundamental task. Identifying the accurate question type enhances the retrieval of more accurate answers. Factoid questions are the most challenging type of question to classify in which many approaches have been proposed with the objective of enhancing the classification of this type of question. In this paper, a grammar-based framework is used. The framework makes use of three main features which are, grammatical features, domain specific features and patterns. Using machine learning algorithms for the classification process, experimental results show that our approach has a good level of accuracy. Keywords: Information retrieval, Question classification, Factoid questions, Grammatical features, Machine learning

This article examines the effect of changes in sovereign credit ratings and their outlook on the stock market returns of European countries at different phases of business cycle. Using standard four-factor model, it records a significant average marginal effect of credit rating announcements on stock market returns. Both magnitude and significance of the effect vary with business cycle and across announcement types. However, we do not find evidence of pro-cyclical effect of sovereign rating and outlook changes on stock returns. Our results show that stock markets react more negatively to rating downgrades in recovery phases and more positively to rating upgrades in contractionary period. Both results are statistically significant and robust to various sensitivity tests.

We introduce DELFT, a factoid question answering system which combines the nuance and depth of knowledge graph question answering approaches with the broader coverage of free-text. DELFT builds a free-text knowledge graph from Wikipedia, with entities as nodes and sentences in which entities co-occur as edges. For each question, DELFT finds the subgraph linking question entity nodes to candidates using text sentences as edges, creating a dense and high coverage semantic graph. A novel graph neural network reasons over the free-text graph-combining evidence on the nodes via information along edge sentences-to select a final answer. Experiments on three question answering datasets show DELFT can answer entity-rich questions better than machine reading based models, bert-based answer ranking and memory networks. DELFT's advantage comes from both the high coverage of its free-text knowledge graph-more than double that of dbpedia relations-and the novel graph neural network which reasons on the rich but noisy free-text evidence., WWW2020

Taking a user-centric approach, we study the features that render an answer to a non-factoid question useful in the eyes of the person who asked that question. An editorial study, where participants assess the usefulness of the answers they received in response to their questions, as well as 12 different aspects associated with the answers, indicates considerable correlation between certain aspects such as relevance, correctness, and completeness with the user-perceived usefulness of answers. Moreover, we investigate the effectiveness of some commonly used answer quality measures, such as ROGUE, BLEU, METEOR, and BERTScore, demonstrating that these measures are limited in their ability to capture the aspects of usefulness and have room for improvement. The question answering dataset created in our work was made publicly available.

Восприятие исторических событий, основанное на доступных фактах, в условиях медиатизации социального пространства всё в большей степени дополняется и заменяется их смысловой интерпретацией. Это вызывает к жизни появление таких понятий, как «постправда» и «постреальность». В ситуации прогрессирующего роста источников и объёмов информации, «настоящая» реальность отходит на второй план и уступает место потоку интерпретаций, производимых заинтересованными субъектами в целях влияния на сознание и поведение потребителей. «Постправда» применяется для характеристики любых персон, организаций и событий, в том числе прецедентов военной истории. Одним из инструментов «постправды» становится коммеморация – историческая память. В фокусе статьи находятся вопросы о том, с помощью каких коммуникативных приёмов осуществляются коммеморативные операции; как противодействовать тем из них, которые являются противозаконными, и какую информационную политику вести в отношении механизмов допустимых. Анализируется практика конструирования информационного фона вокруг исторических ситуаций Великой Отечественной и Второй мировой войн с помощью «фактов», «фактоидов», «фейков», «дезинформации». Делается вывод о том, что коммеморативные решения со смысловыми интерпретациями имеют традиции в прошлом и становятся неизбежными в настоящем. Они выполняют роль системообразующего информационного конструктора, программирующего желаемый образ мыслей и действий целевых групп как внутри страны, так и за её пределами. Смысловое позиционирование событий военной истории предполагает ориентацию аудитории на то, как следует воспринимать исторический контекст сейчас и что он значит в перспективе. В рамках данного процесса требование «объективности» заменяется необходимостью «актуальности», и контекст сообщений переосмысливается в приоритетном для конкретного политического субъекта направлении., The perception of historical events based on available facts, in the context of the mediatization of social space, is increasingly supplemented and replaced by their semantic interpretation. This gives rise to the emergence of such concepts as "post-truth" and "post-reality". In a situation of progressive growth of sources and volumes of information, "real" reality fades into the background and gives way to a stream of interpretations made by interested parties in order to influence the consciousness and behavior of consumers. "Post-truth" is used to characterize any person, organization and event, including the precedents of military history. One of the tools of "post-truth" is commemoration –historical memory. The focus of the article is on the questions of what communication methods are used to carry out commemorative operations; how to counteract those of them that are illegal, and what information policy to conduct with respect to the mechanisms of admissible. The practice of constructing the information background around the historical situations of the Great Patriotic War and the Second World War is analyzed with the help of “facts”, “factoids”, “fakes”, “disinformation”. It is concluded that commemorative decisions with semantic interpretations have traditions in the past and become inevitable in the present. They play the role of a system-forming information constructor, programming the desired way of thinking and actions of target groups both within the country and abroad. The semantic positioning of events in military history presupposes the orientation of the audience on how the historical context should be perceived now and what it means in the future. Within the framework of this process, the requirement of "objectivity" is replaced by the need for "relevance", and the context of messages is rethought in a priority direction for a particular political subject.

Knowledge base question answering (KBQA) aims at automatically answering factoid questions over knowledge bases (KBs). For complex questions that require multiple KB relations or constraints, KBQA faces many challenges including question understanding, component linking (e.g., entity, relation, and type linking), and query composition. In this paper, we propose a novel graph structure called Entity Description Graph (EDG) to represent the structure of complex questions, which can help alleviate the above issues. By leveraging the EDG structure of given questions, we implement a QA system over DBpedia, called EDGQA. Extensive experiments demonstrate that EDGQA outperforms state-of-the-art results on both LC-QuAD and QALD-9, and that EDG-based decomposition is a feasible way for complex question answering over KBs.

Non-factoid question answering system answers questions which are not fact-based and difficult to answer. The paper discusses different taxonomies of semantic relations that are utilized for answering different types of non-factoid questions. Various adverbial clauses are explored for why-type and how-type questions. Different challenges for identifying semantic relations with their approaches to address them are also discussed. Further, the paper briefs researches that used semantic relations in modules of QAS with some future directions.

Knowledge Base Question Answering (KBQA) is the problem of predicting an answer for a factoid question over a given knowledge base (KB). Answering questions typically requires reasoning over multiple links in the given KB. Humans tend to answer questions by grouping different objects to perform reasoning over acquired knowledge. Hypergraphs provide a natural tool to model group relationships. In this work, inspired by typical human intelligence, we propose a new method for KBQA based on hypergraphs. Existing methods for KBQA, though effective, do not explicitly incorporate the recursive relational group structure in the given KB. Our method, which we name RecHyperNet (Recursive Hypergraph Network), exploits a new way of modelling KBs through recursive hypergraphs to organise such group relationships in KBs. Experiments on multiple KBQA benchmarks demonstrate the effectiveness of the proposed RecHyperNet. We have released the code. © 2021 Association for Computational Linguistics

Factoid questions and definitional questions are commonly used in recent works in Question Answering (QA). These questions can be answered by a Named Entity (NE) or a short sentence. However, a more challenging and less common question to deal with is the complex question, or why-question. This issue motivated us to develop a new question-answer system for answering Arabic complex questions in the medical domain. Our system uses Natural Language Processing (NLP) and Information Retrieval (IR) techniques. Using the NooJ platform, we construct dictionaries and transducers to analyze complex questions and to extract causal sentences which represent the answers of why-questions. Experimentations of our Arabic medical question answering system show interesting results.

Users frequently ask simple factoid questions for question answering (QA) systems, attenuating the impact of myriad recent works that support more complex questions. Prompting users with automatically generated suggested questions (SQs) can improve user understanding of QA system capabilities and thus facilitate more effective use. We aim to produce self-explanatory questions that focus on main document topics and are answerable with variable length passages as appropriate. We satisfy these requirements by using a BERT-based Pointer-Generator Network trained on the Natural Questions (NQ) dataset. Our model shows SOTA performance of SQ generation on the NQ dataset (20.1 BLEU-4). We further apply our model on out-of-domain news articles, evaluating with a QA system due to the lack of gold questions and demonstrate that our model produces better SQs for news articles -- with further confirmation via a human evaluation., 17 pages

We introduce NLQuAD, the first data set with baseline methods for non-factoid long question answering, a task requiring document-level language understanding. In contrast to existing span detection question answering data sets, NLQuAD has non-factoid questions that are not answerable by a short span of text and demanding multiple-sentence descriptive answers and opinions. We show the limitation of the F1 score for evaluation of long answers and introduce Intersection over Union (IoU), which measures position-sensitive overlap between the predicted and the target answer spans. To establish baseline performances, we compare BERT, RoBERTa, and Longformer models. Experimental results and human evaluations show that Longformer outperforms the other architectures, but results are still far behind a human upper bound, leaving substantial room for improvements. NLQuAD’s samples exceed the input limitation of most pre-trained Transformer-based models, encouraging future research on long sequence language models.

Visual Question Answering (VQA) is a multi disciplinary challenging problem involving various fields such as Natural Language Processing, Computer Vision, Deep Learning and Knowledge Representation, which has garnered much interest among researchers, especially with the recent advancements in machine perception. The problem at hand not only involves reasoning over visual elements present in the image or natural language understanding of the input query, but also may involve outside world knowledge in order to infer the answer. In this paper, we convert the VQA problem to a factoid question answering task over a set of natural language facts extracted from images in the Visual Genome Dataset [1]. Recent literatures in textual question answering have established the effectiveness of End to End Memory Networks (MemN2N) over the standard LSTMs. Inspired by the approaches incorporated by researchers in this direction, as a first step to create an explainable VQA model, this paper proposes the incorporation of MemN2N with soft attention for inferring the answer from a set of regional facts extracted from the image. We also experiment with the addition of a Bayesian Neural layer for posterior reasoning of the answer from a fixed vocabulary, as proposed in [2] which shows a significant improvement in the accuracy score compared to the other models tested.

Answer re-ranking and quality tagging are two major techniques used to address irrelevant answers in web-based community question answering systems (CQA). However, answer re-ranking is memory inefficient, and quality tagging lacks the ability to predict partially relevant responses. The reported precisions of both mechanisms are also low. Those facts emphasize the importance of finding alternative techniques for identifying irrelevant answers. In this paper, we have analyzed the capability of three widely popular deep learning models (CNN, LSTM and CLSTM) in the NLP literature to identify irrelevant answers in factoid and non-factoid systems. Further, we studied the ability of the same deep learning models to detect partially relevant answers in non-factoid systems. According to the results, the CLSTM model performed over CNN and LSTM in detecting irrelevant answers.

Question Answering (QA) framework is a data recovery framework in which an explicit reply is required because of a submitted query, as opposed to a lot of references that may contain the appropriate responses. It is a man-machine specialized gadget. The essential thought of QA frameworks in Natural Language Processing (NLP) is to give the right responses to the queries. This paper shows a step towards building a question answering framework in Bangla. A significant initial phase in building up a QA framework is the dataset. As the Bangla QA framework is at the beginning phase of advancement, we have prepared our own Bangla question answering dataset. The proposed model for Bangla-QA contains careful stacking of several deep learning-based sub-models to grasp the context from a comprehension and generate appropriate answers for diverse question types. We have accomplished up to 92.16% F1 score considering partial match and 76.8% exact match on test dataset utilizing LSTM.

To study the impact of providing direct answers in search results on user behavior, we conducted a controlled user study to analyze factors including reading time, eye-tracked attention, and the influence of the quality of answer module content. We also studied a more advanced answer interface, where multiple answers are shown on the search engine results page (SERP). Our results show that users focus more extensively than normal on the top items in the result list when answers are provided. The existence of the answer module helps to improve user engagement on SERPs, reduces user effort, and promotes user satisfaction during the search process. Furthermore, we investigate how the question type -- factoid or non-factoid -- affects user interaction patterns. This work provides insight into the design of SERPs that includes direct answers to queries, including when answers should be shown.

We investigated how users evaluate passage-length answers for non-factoid questions. We conduct a study where answers were presented to users, sometimes shown with automatic word highlighting. Users were tasked with evaluating answer quality, correctness, completeness, and conciseness. Words in the answer were also annotated, both explicitly through user mark up and implicitly through user gaze data obtained from eye-tracking. Our results show that the correctness of an answer strongly depends on its completeness, conciseness is less important. Analysis of the annotated words showed correct and incorrect answers were assessed differently. Automatic highlighting helped users to evaluate answers quicker while maintaining accuracy, particularly when highlighting was similar to annotation. We fine-tuned a BERT model on a non-factoid QA task to examine if the model attends to words similar to those annotated. Similarity was found, consequently, we propose a method to exploit the BERT attention map to generate suggestions that simulate eye gaze during user evaluation.

Voice-based assistants have become a popular tool for conducting web search, particularly for factoid question answering. However, for more complex web searches, their functionality remains limited, as does our understanding of the ways in which users can best interact with audio-based search results. In this paper, we compare and contrast user behaviour through the representation of search results over two mediums: text and audio. We begin by conducting a crowdsourced study exposing the differences in user selection of search results when those are presented in text and audio formats. We further confirm these differences and investigate the reasons behind them through a mixed-methods laboratory study. Through a qualitative analysis of the collected data, we produce a list of guidelines for an audio-based presentation of search results.