Data Science Mentoring Support using Neural Networks

Automatic question generation from text or paragraph is a great challenging task which attracts broad attention in natural language processing. Because of the verbose texts and fragile ranking methods, the quality of top generated questions is poor. In this paper, we present a novel framework Automatic Chinese Question Generation (ACQG) to generate questions from text or paragraph. In ACQG, we use an adopted TextRank to extract key sentences and a template-based method to construct questions from key sentences. Then a multi-feature neural network model is built for ranking to obtain the top questions. The automatic evaluation result reveals that the proposed framework outperforms the state-of-the-art systems in terms of perplexity. In human evaluation, questions generated by ACQG rate a higher score.

Automatic question generator (AQG) technology is a system developed to create questions automatically from input in the form of text, images, and videos. AQG has been developed using various approaches such as natural language processing (NLP), statistical approaches, and other machine approaches. AQG has a very important role in the world of education, especially in independent education, because it can be used as a good evaluation medium for students. Utilizing AQG in independent education gives students full control to determine their learning. AQG turns learning into a more interactive experience by generating questions that can trigger critical thinking and problem-solving skills. AQG technology developed in independent learning will encourage students to respond actively to the material and understand concepts more deeply. Approximately 60% of research related to AQG has been conducted for assessment, 18% for knowledge acquisition, and the remainder for validation and other purposes. This research was conducted by conducting a comprehensive review of 63 articles related to AQG in education.

The advent of the Internet resulted in availability of e-content in huge quantities; this e-content includes information in different domains. This information is being exploited for various reasons and applications. As the formal evaluation and assessment techniques are undergoing transformations, there is also a need for a voluminous Question Bank (QB). The manual creation of questions is a time-consuming and expensive task as it requires domain experts and there is a growing trend towards question bank based question paper generation for assessment. Hence the problem of automatic question generation has attracted attention of the Natural Language Processing (NLP) research community; this problem can be handled by NLP. Automatic generation of high-quality questions is the long-term goal of the Question Generation (QG) research community. The automatic question generation aims to generate all possible questions from the input text or document. This survey article presents the issues, challenges and approaches as well as reviews the reported research work in the context of automatic question generation for different languages at national and international level.

The progress in natural language processing (NLP) using large language models (LLMs) has greatly improved patient information extraction from clinical narratives. However, most methods based on the fine-tuning strategy have limited transfer learning ability for cross-domain applications. This study proposed a novel approach that employs a soft prompt-based learning architecture, which introduces trainable prompts to guide LLMs toward desired outputs. We examined two types of LLM architectures, including encoder-only GatorTron and decoder-only GatorTronGPT, and evaluated their performance for the extraction of social determinants of health (SDoH) using a cross-institution dataset from the 2022 n2c2 challenge and a cross-disease dataset from the University of Florida (UF) Health. The results show that decoder-only LLMs with prompt tuning achieved better performance in cross-domain applications. GatorTronGPT achieved the best F1 scores for both datasets, outperforming traditional fine-tuned GatorTron by 8.9% and 21.8% in a cross-institution setting, and 5.5% and 14.5% in a cross-disease setting.

Pre-trained language models (PLMs) have driven substantial progress in natural language processing but remain vulnerable to adversarial attacks, raising concerns about their robustness in real-world applications. Previous studies have sought to mitigate the impact of adversarial attacks by introducing adversarial perturbations into the training process, either implicitly or explicitly. While both strategies enhance robustness, they often incur high computational costs. In this work, we propose a simple yet effective add-on module that enhances the adversarial robustness of PLMs by removing instance-level principal components, without relying on conventional adversarial defenses or perturbing the original training data. Our approach transforms the embedding space to approximate Gaussian properties, thereby reducing its susceptibility to adversarial perturbations while preserving semantic relationships. This transformation aligns embedding distributions in a way that minimizes the impact of adversarial noise on decision boundaries, enhancing robustness without requiring adversarial examples or costly training-time augmentation. Evaluations on eight benchmark datasets show that our approach improves adversarial robustness while maintaining comparable before-attack accuracy to baselines, achieving a balanced trade-off between robustness and generalization.

While large language models (LLM) have made impressive progress in natural language processing, it remains unclear how to utilize them in improving automatic speech recognition (ASR). In this work, we propose to train a single multilingual language model (LM) for shallow fusion in multiple languages. We push the limits of the multilingual LM to cover up to 84 languages by scaling up using a mixture-of-experts LLM, i.e., generalist language model (GLaM). When the number of experts increases, GLaM dynamically selects only two at each decoding step to keep the inference computation roughly constant. We then apply GLaM to a multilingual shallow fusion task based on a state-of-the-art end-to-end model. Compared to a dense LM of similar computation during inference, GLaM reduces the WER of an English long-tail test set by 4.4% relative. In a multilingual shallow fusion task, GLaM improves 41 out of 50 languages with an average relative WER reduction of 3.85%, and a maximum reduction of 10%. Compared to the baseline model, GLaM achieves an average WER reduction of 5.53% over 43 languages.

This study investigates the role of artificial intelligence (AI) in Arabic language processing, focusing on both its developmental potential and the challenges that threaten the language’s integrity and identity. Despite progress in natural language processing (NLP), the Arabic language still faces substantial barriers due to its intricate grammatical rules and semantic depth, which make it more complex to integrate into AI systems compared to other languages. In addition to technical obstacles, the application of AI raises ethical concerns—particularly around preserving the authenticity of Arabic and protecting intellectual property. Issues such as uncredited AI-generated content and reduced human involvement in language formulation may compromise the credibility and originality of information. The study aims to offer future-oriented solutions that address these ethical challenges while promoting a balanced approach between leveraging AI technologies and safeguarding the linguistic heritage of Arabic. Adopting a qualitative, literature-based methodology with a focus on ethical, linguistic, and technological analysis, the research underscores the urgent need for ethical and regulatory frameworks to guide AI implementation in this domain. The findings highlight the importance of continuous investment in research and development to support Arabic’s sustainable integration into AI-driven platforms. As a contribution, the study provides a conceptual foundation for responsible innovation, advocating for policies that protect the identity of the Arabic language while enabling its evolution within the digital age

Answering questions is one method to increase or measure understanding. However, creating relevant and answerable questions from the given context is not easy. Automatic Question Generation (AQG) is a part of Natural Language Processing (NLP) which can generate questions automatically from text input. Many studies related to AQG have been carried out but are still very limited in Indonesian texts, especially those that use the latest Transformer variations. This study proposes an AQG system that utilizes the latest power Transformer, the multilingual Text-to-Text Transfer Transformer (mT5). We fine-tune the mT5 model to extract answers from context and generate questions based on those answers. We use the Indonesian dataset extracted from the TyDiQA dataset and evaluate this model against the TyDiQA validation set using BLEU (BiLingual Evaluation Understudy) and ROUGE (Recall-Oriented Understudy for Gisting Evaluation) metrics. This model achieved BLEU-1, BLEU-2, BLEU-3, BLEU-4, and ROUGE-L scores of 36.54, 28.24, 22.61, 18.44, and 39.57, respectively. Our model performs well and generates questions in understandable Indonesian with good word choice and grammar based on manual validation.

A survey on textual entailment based question answering

In the last few years, researchers in the field of e-learning and Natural Language Processing (NLP) have shown an increased interest in automatic question generation. However, little research has discussed the automatic evaluation of listening comprehension in multimedia learning. In this work, we present an automatic quiz generation for TED Talks video clips, called TED Quiz. TED Quiz generates multiple-choice questions in two question types, gist-content questions and detail questions. We use a graph-based algorithm, Lex Rank, to identify the most important part of a talk, as the main concept of a gist-content question. We also proposed an approach to distractor selection for detail question generation that generates grammatically correct but semantically wrong sentences as distractors. The experimental results demonstrated that the measured results from automatically generated questions are comparable with that from manually generated questions because their scores were significantly correlated. Moreover, most subjects agreed that the generated listening comprehension questions were of quality and usefulness.

Number of people having expertise in a certain domain is less than people who need information in that domain. In this situation, an automatic question answering (QA) system is necessary. Observing available manual QA sites on internet, the real world question that people usually ask have different expected answer type (EAT) compared to a common automatic QA. Addressing a case study of a religion domain which makes it a closed domain QA, we proposed the EAT into 6 types: LAW, DEFINITION, COMPARISON, METHOD, TIME and PERSON. Different with common QA approach, we built the QA system using case based approach which consists of two main components: Question Analyzer and Case Retriever. Related with the case based reasoning (CBR) framework, these two main components act as the Retrieve and Reuse process while the Revise and Retain process is handle by Case Retainer component. The QA system was built using available Indonesian Natural Language Processing (NLP) tools and FreeCBR as the CBR library. The experiments were done to calculate the accuracy and testing the system with unknown case. By using 77 cases collected from internet with assumption that all answers are available, the experiments achieved 97% accuracy. And by using 10 test cases for the unknown case, the similarity score calculated by the system showed that the test questions have no answer in the available case base.

Yioop is an open source search engine developed and managed by Dr. Christopher Pollett. Currently, Yioop returns the search results of the query in the form of list of URLs, just like other search engines (Google, Bing, DuckDuckGo, etc.) This paper created a new module for Yioop. This new module, known as the Question-Answering (QA) System, takes the search queries in the form of natural language questions and returns results in the form of a short answer that is appropriate to the question asked. This feature is achieved by implementing various functionalities of Natural Language Processing (NLP). By using NLP, the new Question-Answering (QA) System attempts to extract the necessary information from the query provided by the user and provides an appropriate answer from the available data.

Question-Answering (QA) system is improved version of search engines or can be stated as specific Search engines because search engines deliver data in the form of URLs whereas Question-Answering system produces direct answer to the query asked. That's why more focus has been put on developing better Question Answering systems that could respond to large variety of user query. Semantic Web Technology has aided to the development of such systems from past many years because it could add meaning to the query being asked. Such systems are more efficient than the systems developed in other languages. In this paper, we have proposed a QA system of JAVA Programming Language, where knowledge will be represented in the form of Ontology and based on relations derived from domain knowledge to obtained final answer. Information present in Ontology is represented in the form of tables to feed in knowledge base. It combines the techniques of Natural Language Processing and Semantic Web.

Open-domain question answering is an emerging field of natural language processing. Application areas of question answering (QA) systems are increasing with the availability of new datasets, more powerful machines and easy access to voice-enabled devices that interact with humans in their daily lives. Users can get into interactions with QA systems without the need to be limited by any domain. However, there is still need for QA systems that can answer questions in a specific field such as international criminal law, violence of human rights or bill of rights. The recent approaches on open-domain question answering made it possible to extend open-domain QA systems towards specific fields. This paper presents Quill-QA, the new feature of the Quill Platform. Quill Platform is a web platform providing set of tools that digitize and visualize negotiated texts, any text in the form of formal negotiations. Quill-QA is based on Facebook's DrQA system and as a result it adopts a flexible Ranker-Reader architecture. Quill-QA answers questions addressed on the samples of negotiated texts such as U.S. Constitutional Convention of 1787, Utah State Constitutional Convention of 1895 and European Union Act 2017. Without limiting the scope of the input questions to a finite set of templates, Quill-QA is the first QA system that focuses on negotiated texts.

IBM Research has over 200 people working on Unstructured Information Management (UIM) technologies with a strong focus on Natural Language Processing (NLP). These researchers are engaged in activities ranging from natural language dialog, information retrieval, topic-tracking, named-entity detection, document classification and machine translation to bioinformatics and open-domain question answering. An analysis of these activities strongly suggested that improving the organization's ability to quickly discover each other's results and rapidly combine different technologies and approaches would accelerate scientific advance. Furthermore, the ability to reuse and combine results through a common architecture and a robust software framework would accelerate the transfer of research results in NLP into IBM's product platforms. Market analyses indicating a growing need to process unstructured information, specifically multilingual, natural language text, coupled with IBM Research's investment in NLP, led to the development of middleware architecture for processing unstructured information dubbed UIMA. At the heart of UIMA are powerful search capabilities and a data-driven framework for the development, composition and distributed deployment of analysis engines. In this paper we give a general introduction to UIMA focusing on the design points of its analysis engine architecture and we discuss how UIMA is helping to accelerate research and technology transfer.