- Research Article
11
- 10.1016/j.dss.2023.114038
Building a deep learning-based QA system from a CQA dataset
- Jun 15, 2023
- Decision Support Systems
- Sol Jin + 4 more +4
Building a deep learning-based QA system from a CQA dataset
Arabic factoid Question-Answering system for Islamic sciences using normalized corpora
Building a deep learning-based QA system from a CQA dataset
Building a deep learning-based QA system from a CQA dataset
Factoid question answering with web, mobile and speech interfaces
In this paper we describe the web and mobile-phone interfaces to our multi-language factoid question answering (QA) system together with a prototype speech interface to our English-language QA system. Using a statistical, data-driven approach to factoid question answering has allowed us to develop QA systems in five languages in a matter of months. In the web-based system, which is accessible at http://asked.jp, we have combined the QA system output with standard search-engine-like results by integrating it with an open-source web search engine. The prototype speech interface is based around a VoiceXML application running on the Voxeo developer platform. Recognition of the user's question is performed on a separate speech recognition server dedicated to recognizing questions. An adapted version of the Sphinx-4 recognizer is used for this purpose. Once the question has been recognized correctly it is passed to the QA system and the resulting answers read back to the user by speech synthesis. Our approach is modular and makes extensive use of open-source software. Consequently, each component can be easily and independently improved and easily extended to other languages.
Read moreAn Ontology-Driven Question Answering System For Computer Network Module
The field of "question and answering" has become a popular area of research in recent years.The main reason for this is that the search through the ”question answering" system is found to be more efficient than the normal search. The Question Answering (QA) Systems can be two types in general such as open-domain QA systems and closed-domain QA systems. Search engines are generally open-domain QA systems that are used to search and retrieve the data we need. However, instead of search engines giving accurate and precise answers to the user queries, they often returned the list of links. Then, the user clicks on each link one by one to get the answer. This method of searching can sometimes not give the precise answers to the user queries, or the user may have to spend more time searching for the answer, and hence the users may experience discomfort. This situation can be avoided by using the semantic concept. The normal web data are machine-readable and can be understood by humans, whereas the semantic web information is machine-readable and understandable. Ontology is the main component of the semantic web and it can be described as the structure of knowledge-representation of a particular domain or subject. It clearly describes concepts, roles, instances, and the relationships between them. The Question Answering (QA) system is one of the popular applications of ontology. Here, the QA system is used to extract the precise answer to the user queries from the data repository. The system can be developed using different techniques like NLP with IR, reasoning with the NLP, web-based QA System, ontology-based QA System, and more. This particular question answering system is developed using the ontology model. This ontology-driven question answering (QA) system provides the facility to the users to find accurate and concise answers for their queries in the Computer Network module of ICT Subject.The same questions that are asked in the ontology-based question answering system can be asked in the web-based system. The reason is that both of these methods are used in this system. This will make it easier for users to understand the difference between the two systems.The performance results of both these systems further strengthen the statement
Read moreAutomatic question answering using the web: Beyond the Factoid
In this paper we describe and evaluate a Question Answering (QA) system that goes beyond answering factoid questions. Our approach to QA assumes no restrictions on the type of questions that are handled, and no assumption that the answers to be provided are factoids. We present an unsupervised approach for collecting question and answer pairs from FAQ pages, which we use to collect a corpus of 1 million question/answer pairs from FAQ pages available on the Web. This corpus is used to train various statistical models employed by our QA system: a statistical chunker used to transform a natural language-posed question into a phrase-based query to be submitted for exact match to an off-the-shelf search engine; an answer/question translation model, used to assess the likelihood that a proposed answer is indeed an answer to the posed question; and an answer language model, used to assess the likelihood that a proposed answer is a well-formed answer. We evaluate our QA system in a modular fashion, by comparing the performance of baseline algorithms against our proposed algorithms for various modules in our QA system. The evaluation shows that our system achieves reasonable performance in terms of answer accuracy for a large variety of complex, non-factoid questions.
Read moreNeural Domain Adaptation for Biomedical Question Answering
Factoid question answering (QA) has recently benefited from the development of deep learning (DL) systems. Neural network models outperform traditional approaches in domains where large datasets exist, such as SQuAD (ca. 100,000 questions) for Wikipedia articles. However, these systems have not yet been applied to QA in more specific domains, such as biomedicine, because datasets are generally too small to train a DL system from scratch. For example, the BioASQ dataset for biomedical QA comprises less then 900 factoid (single answer) and list (multiple answers) QA instances. In this work, we adapt a neural QA system trained on a large open-domain dataset (SQuAD, source) to a biomedical dataset (BioASQ, target) by employing various transfer learning techniques. Our network architecture is based on a state-of-the-art QA system, extended with biomedical word embeddings and a novel mechanism to answer list questions. In contrast to existing biomedical QA systems, our system does not rely on domain-specific ontologies, parsers or entity taggers, which are expensive to create. Despite this fact, our systems achieve state-of-the-art results on factoid questions and competitive results on list questions.
Read moreA Study of Deep Learning for Factoid Question Answering System
End-to-end question answering system has attracted considerable attention in the artificial intelligence research community in recent years. In this paper, we proposed an integrated deep learning model for factoid question answering system. This study uses the Delta Reading Comprehension Dataset (DRCD) to build a model to implement a factoid question answering system and to combine the classification of question and answer to evaluate with exact match (EM) and F1 score. The study determines whether the comparison can increase the proportion of EM and whether the expected answer type can effectively increase the answer accuracy rate. To perfect the transformation, a question-and-answer system that uses the BERT pre-training model is applied to the DRCD dataset together with the expected answer type analysis and comparison. The contribution of this paper is that we proposed a system architecture of factoid question answering (QA) system using BERT with question expected answer type (Q-EAT) and answer type classification (AT) models. Findings confirm that the classification of question and answer can improve the EM ratio. When the question sentence and the answer classification are the same, the prediction accuracy EM of the question answering system will be improved.
Read moreDesign of Question Answering System with Automated Question Generation
One of the most difficult problems in developing question-answering (QA) system is that it is so hard to generate natural language questions and to find an answer to a query question. In order to avoid a number of difficulties of developing QA systems, we propose a new style of question-answering system architecture that actively uses sentences within a document as a source of question/answer. Basically, our proposed QA system gives user a set of candidate query question for user information needs, and the candidate questions are automatically generated from significant sentences that are expected to contain meaningful facts or events. The QA system builds a complete database of (question, answer) pairs after analyzing a whole collection of documents. For this, we need to perform the following steps: sentence split, named-entity recognition, question generation, question filtering, question/ answer indexing. The important things in the process are question generation and question filtering. For the first thing, we can generate questions that ask the entities extracted from a given sentence. The question filtering is to isolate significant sentences that have meaningful information that users want.
Read moreA Fast and Reliable Question-Answering System Based on Predictive Answer Indexing and Lexico-Syntactic Pattern Matching
We propose a Question-Answering (QA) system in Korean that uses a predictive answer indexer. The predictive answer indexer, first, extracts all answer candidates in a document in indexing time. Then, it gives scores to the adjacent content words that are closely related to each answer candidate. Next, it stores the weighted content words with each candidate into a database. Using this technique, along with a complementary analysis of questions, the proposed QA system can save response time because it is not necessary for the QA system to extract answer candidates with scores on retrieval time. If the QA system is combined with a traditional Information Retrieval system, it can improve the document retrieval precision for closed-class questions after minimum loss of retrieval time.
Read moreMAYA
We propose a Question-answering (QA) system in Korean that uses a predictive answer indexer. The predictive answer indexer, first, extracts all answer candidates in a document in indexing time. Then, it gives scores to the adjacent content words that are closely related with each answer candidate. Next, it stores the weighted content words with each candidate into a database. Using this technique, along with a complementary analysis of questions, the proposed QA system can save response time because it is not necessary for the QA system to extract answer candidates with scores on retrieval time. If the QA system is combined with a traditional Information Retrieval system, it can improve the document retrieval precision for closed-class questions after minimum loss of retrieval time.
Read moreCookingQA: A Question Answering System Based on Cooking Ontology
We present an approach to develop a Question Answering (QA) system over cooking recipes that makes use of Cooking Ontology management. QA systems are designed to satisfy the user’s specific information need whereas ontology is the conceptualization of knowledge and it exhibits the hierarchical structure. The system is an Information retrieval (IR) based system where the various tasks to be handled like question classification, answer pattern recognition, indexing, final answer generation. Our proposed QA System use Apache Lucene for document retrieval. All cooking related documents are indexed using Apache Lucene. Stop words are removed from each cooking related question and formed the query words which are identified to retrieve the most relevant document using Lucene. Relevant paragraphs are selected from the retrieved documents based on the tf-idf of the matching query words along with n-gram overlap of the paragraph with the original question. This paper also presents a way to develop an ontology model in such a way that the queries can be processed with the help of the ontology knowledge base and generate the exact answer.
Read moreFact-based question decomposition in DeepQA
Factoid questions often contain more than one fact or assertion about their answers. Question-answering (QA) systems, however, typically do not use such fine-grained distinctions because of the need for deep understanding of the question in order to identify and separate the facts. We argue that decomposing complex factoid questions is beneficial to QA systems, because the more facts that support an answer candidate, the more likely it is to be the correct answer. We broadly categorize decomposable questions into two types: parallel and nested. Parallel decomposable questions contain subquestions that can be evaluated independent of each other. Nested questions require decompositions to be processed in sequence, with the answer to an “inner” subquestion plugged into an “outer” subquestion. In this paper, we present a novel question decomposition framework capable of handling both decomposition types, built on top of the base IBM Watson™ QA system for Jeopardy!™. The framework contains a suite of decomposition rules that use predominantly lexico-syntactic features to identify facts within complex questions. It also contains a question-rewriting component and a candidate re-ranker, which uses machine learning and heuristic selection strategies to generate a final ranked answer list, taking into account answer confidences from the base QA system. We apply our decomposition framework to the particularly challenging domain of Final Jeopardy! questions, which are found to be difficult even for qualified Jeopardy! players, and we show a statistically significant improvement in the performance of our baseline QA system.
Read moreChallenges and opportunities for Arabic question-answering systems: current techniques and future directions
Artificial intelligence-based question-answering (QA) systems can expedite the performance of various tasks. These systems either read passages and answer questions given in natural languages or if a question is given, they extract the most accurate answer from documents retrieved from the internet. Arabic is spoken by Arabs and Muslims and is located in the middle of the Arab world, which encompasses the Middle East and North Africa. It is difficult to use natural language processing techniques to process modern Arabic owing to the language’s complex morphology, orthographic ambiguity, regional variations in spoken Arabic, and limited linguistic and technological resources. Only a few Arabic QA experiments and systems have been designed on small datasets, some of which are yet to be made available. Although several reviews of Arabic QA studies have been conducted, the number of studies covered has been limited and recent trends have not been included. To the best of our knowledge, only two systematic reviews focused on Arabic QA have been published to date. One covered only 26 primary studies without considering recent techniques, while the other covered only nine studies conducted for Holy Qur’an QA systems. Here, the included studies were analyzed in terms of the datasets used, domains covered, types of Arabic questions asked, information retrieved, the mechanism used to extract answers, and the techniques used. Based on the results of the analysis, several limitations, concerns, and recommendations for future research were identified. Additionally, a novel taxonomy was developed to categorize the techniques used based on the domains and approaches of the QA system.
Read moreMulti-Label Question Classification for Factoid and List Type Questions in Biomedical Question Answering
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., <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Coenzyme Q(10)</i> is classified under two categories in Unified Medical Language System (UMLS): <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">organic chemical</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">biologically active substance</i> . 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 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F_{1}$ </tex-math></inline-formula> and Mean Reciprocal Rank measures, respectively.
Read moreA Depth Evidence Score Fusion Algorithm for Chinese Medical Intelligence Question Answering System
Question answering (QA) system is becoming the focus of the research in medical health in terms of providing fleetly accurate answers to users. Numerous traditional QA systems are faced to simple factual questions and do not obtain accurate answers for complex questions. In order to realize the intelligent QA system for disease diagnosis and treatment in medical informationization, in this paper, we propose a depth evidence score fusion algorithm for Chinese Medical Intelligent Question Answering System, which can measure the text information in many algorithmic ways and ensure that the QA system outputs accurately the optimal candidate answer. At the semantic level, a new text semantic evidence score based on Word2vec is proposed, which can calculate the semantic similarity between texts. Experimental results on the medical text corpus show that the depth evidence score fusion algorithm has better performance in the evidence-scoring module of the intelligent QA system.
Read moreSemantic analysis and classification of medical questions for a logic-based medical question-answering system
Due to the continuous growth and crucial real-world impact of information produced in the biomedical and health domain, there is an acutely felt need for question-answering (QA) systems that can aid medical researchers and health care professionals in their information search. In order to provide users with accurate answers, such systems need to utilize semantic analysis and processing of medical texts and knowledge resources. Moreover, QA systems equipped with reasoning capabilities can derive more adequate answers by using inference mechanisms. We propose a logic-based medical QA system which uses Description Logic (DL) for knowledge representation and reasoning. As a first step in constructing the proposed QA system, we present a study on the semantic analysis and categorization of medical questions.
Read more