Question Classification Research Articles

Trends in Accuracy and Appropriateness of Alopecia Areata Information Obtained from a Popular Online Large Language Model, ChatGPT

Background: Patients with alopecia areata (AA) may access a wide range of sources for information about AA, including the recently developed ChatGPT. Assessing the quality of health information provided by these sources is crucial, as patients are utilizing them in increasing numbers. Objectives: The aim of the study was to evaluate appropriateness and accuracy of responses to common patient questions about AA generated by ChatGPT. Methods: Responses generated by ChatGPT 3.5 and ChatGPT 4.0 to 25 questions addressing common patient concerns were assessed by multiple attending dermatologists in an academic center for appropriateness and accuracy. Appropriateness of responses by both models for use in two hypothetical contexts as follows: (1) for patient-facing general information websites, and (2) for electronic health record (EHR) message drafts. Results: The accuracy across all responses was 4.41 out of 5. Accuracy scores of responses ChatGPT 3.5 responses had a mean of 4.29, whereas those generated by ChatGPT 4.0 had mean accuracy score of 4.53. Assessments ranged from 100% of responses rated as appropriate for the general question category to 79% questions about management for an EHR message draft. Raters largely preferred responses generated by ChatGPT 4.0 versus ChatGPT 3.5. Reviewer agreement was found to be moderate across all questions, with a 53.7% agreement and Fleiss’ κ co-efficient of 0.522 (p value <0.001). Conclusions: The large language model ChatGPT outputted mostly appropriate information for common patient concerns. While not all responses were accurate, the trend toward improvement with newer iterations suggests potential future utility for patients and dermatologists.

중학교 수학 학습을 위한 질의응답 챗봇의 설계와 활용

Objectives The purpose of this study is to determine the most effective approach for constructing training data comprising questions and answers for the development of a mathematical question-answering chatbot. This involves a theoretical analysis of mathematical knowledge and an empirical analysis of questions posed by students using the chatbot.&#x0D; Methods The study targeted second-year middle school students and created a total of 1,936 pairs of math-related question-answer sets virtually. The chatbot was trained using the Doc2Vec technique, which is based on similarity. During the data construction process, various considerations related to mathematical knowledge and math learning were taken into account to design different types of questions. The data was then used for training the chatbot. Additionally, students utilized the trained chatbot during their classes for seven months, and their input questions were collected and analyzed.&#x0D; Results A comparison between the collected dataset of 1,636 questions from 92 middle school students and the types of question-answer pairs used for training the chatbot revealed that conceptual knowledge questions accounted for over 70% of the student-generated data. Students also posed questions to test the chatbot's capabilities and inquire about attitudes and emotional aspects related to mathematics. Within the category of conceptual knowledge questions, deeper inquiries involving comparisons, extensions, and connections of concepts were observed.&#x0D; Conclusions The study revealed that when students used the question-answering chatbot, inquiries related to c conceptual understanding of mathematics when students interact with the question-answering chatbot highlights the importance of focusing on conceptual questions. Furthermore, the activity of questioning concepts holds the potential to evolve into more advanced inquiries, fostering connections and extensions of mathematical concepts. Further research is needed to explore more suitable and effective ways of responding to students through the chatbot and utilizing it in educational contexts.

Mortuary palisades, single graves and cultural admixture: The establishment of Corded Ware culture on the Jutland Peninsula

Abstract The emergence and nature of the supra-regional Corded Ware complex in the 3rd millennium BCE is a longstanding, classical question in European archaeology. While recent genomic and genome-wide aDNA analyses have shown that migration was part of the process by which this phenomenon spread across a vast part of Central and northern Europe, the archaeological record of the same area makes equally clear that the character of its more specific regional manifestations was not uniform. This combination calls for renewed regional studies aimed at understanding the pre-existing contexts, the situated processes and the variable outcomes of this new cultural formation, taking a contribution of migration as a starting premise rather than an explanatory end goal. This article presents such a study, focusing on a particular aspect of the so-called Single Grave burial custom on the Jutland Peninsula, which constitutes the most obvious element of Corded Ware culture in the region. Dating to the early part of this tradition (c. 2850–2600 BCE), circular arrangements of wooden posts or planks lodged vertically in a ditch to form a circular palisade are frequently found encircling a contemporaneous human grave. Here, we provide the first systematic study of these structures, which we argue are most meaningfully referred to as mortuary palisades. We present their distribution, chronology and basic morphology as well as their association with graves and relation to burial mounds, leading to a discussion of their probable ritual function in connection with a particular, processual understanding and handling of death. Based on this general presentation, we then analyse geographical variation in the manifestation of the mortuary palisade custom and discuss potentially underlying causes, emphasizing widely differing degrees of cultural admixture of Corded Ware culture and local, pre-existing traditions and preferences in different parts of the Jutland Peninsula.

Evaluation Strategies for Understanding Experiences With Virtual Care in Canada: Mixed Methods Study.

Virtual care was rapidly integrated into pediatric health services during the COVID-19 pandemic. While virtual care offers many benefits, it is necessary to better understand the experiences of those who receive, deliver, and coordinate virtual care in order to support sustainable, high-quality, and patient-centered health care. To date, methods implemented to evaluate users' experiences of virtual care have been highly variable, making comparison and data synthesis difficult. This study aims to describe evaluation strategies currently used to understand personal experiences with pediatric virtual care in Canada. In this mixed methods environmental scan, we first distributed a web-based questionnaire to clinical, research, and operational leaders delivering and evaluating pediatric virtual care in Canada. The questionnaire collected information about how experiences with virtual care have been or are currently being evaluated and whether these evaluations included the perspectives of children or youth, families, providers, or support staff. Second, respondents were asked to share the questions they used in their evaluations, and a content analysis was performed to identify common question categories. Third, we conducted semistructured interviews to further explore our respondents' evaluation experiences across 4 domains-evaluation approaches, distribution methods, response rates, and lessons learned-and interest in a core set of questions for future evaluations. There were 72 respondents to the web-based questionnaire; among those who had conducted an evaluation, we identified 15 unique evaluations, and 14 of those provided a copy of the tools used to evaluate virtual care. These evaluations measured the virtual care experiences of parents or caregivers (n=15, 100%), children or youth (n=11, 73%), health care providers (n=11, 73%), and support staff (n=4, 27%). The most common data collection method used was electronic questionnaires distributed by email. Two respondents used validated tools; the remainder modified existing tools or developed new tools. Content analysis of the 14 submitted questionnaires revealed that the most common questions were about overall participant satisfaction, the comparison of virtual care to in-person care, and whether participants would choose virtual care options in the future. Interview findings indicate respondents frequently relied on methods used by peers and that a standardized, core set of questions to evaluate experiences with virtual care would be helpful to improve evaluation practices and support pediatric health care delivery. At our institution and elsewhere in Canada, experiences with pediatric virtual care have been evaluated using a variety of methods. A more consistent evaluation approach using standardized tools may enable more regular comparisons of experiences with virtual care and the synthesis of findings across health care settings. In turn, this may better inform our approach to virtual care, improve its integration into health systems, and facilitate sustainable, high-quality, patient-centered care.

Locating what comes to mind in empirically derived representational spaces

Real-world judgements and decisions often require choosing from an open-ended set of options which cannot be exhaustively considered before a choice is made. Recent work has found that the options people do consider tend to have particular features, such as high historical value. Here, we pursue the idea that option generation during decision making may reflect a more general mechanism for calling things to mind, by which relevant features in a context-appropriate representational space guide what comes to mind. In this paper, we evaluate this proposal primarily based on what comes to mind in different familiar categories. We first introduce an empirical approach for deriving the space of features that people use to represent items in a particular category and for locating the category members that come to mind within that space. We show that in both familiar and ad hoc categories, a category member’s location along certain dimensions of the derived feature space predicts its likelihood of coming to mind. Next, we show that category members from these feature space locations come to mind by default in a way that is somewhat impervious to conscious control. We then demonstrate that the extent to which a given dimension is a predictor of what comes to mind within a category is related to how relevant that feature is for representing the category in question, using a novel measure of general feature relevance. Finally, we illustrate the usefulness of this framework in the context of a decision making task. We close with the proposal that people call category members to mind according to their location in representational space, specifically based on the predicted usefulness of considering category members with particular features.

Предмет и порядок рассмотрения Конституционным Судом РФ дел о соответствии Конституции Российской Федерации инициативы проведения референдума по предложенному вопросу (предложенным вопросам) референдума

Introduction. The issue of the subject of review of cases of compliance of the initiative to hold a referendum on the proposed issue (proposed issues) of the referendum with the Constitution of the Russian Federation by the Constitutional Court of the Russian Federation is not regulated in the federal legislation. In addition, the federal legislation regulates the procedure for considering cases of this category only in general terms. Theoretical analysis. The systematization of the opinions presented in the scientific literature regarding the subject of verification in this category of cases allows us to distinguish two positions: the subject of consideration by the Constitutional Court should include both the issues proposed for a referendum and the entire procedure for the initiative of holding a referendum; only the issues proposed for a referendum should be the subject of consideration by the Constitutional Court. Empirical analysis. The analysis of the criteria (limits) of verification for the category of cases under consideration showed that, in any case, they should include compliance of the issues proposed for a referendum, the division of jurisdiction and powers between the state authorities of the Russian Federation and the state authorities of the constituent entities of the Russian Federation with the Constitution of the Russian Federation. The criteria for the admissibility of the motion, the procedural form and procedural periods of consideration of cases of the category under study need to be improved. Conclusions. The subject of verification in cases of the category under consideration should be not only the issues proposed for a referendum but also the entire procedure for the initiative of holding a referendum. The author proposes to amend the legislation providing for the exemption of the President of the Russian Federation from paying state duty when bringing the matter before the Constitutional Court of the Russian Federation in cases of the category in question, the possibility of participation of representatives of the electoral association or public authority proposing the initiative to hold a referendum, the Central Election Commission of the Russian Federation and the President of the Russian Federation in consideration of cases of this category. The author also proposes to provide a special chapter in the relevant law regulating the consideration of cases of compliance of the initiative to hold a referendum on the proposed issue (proposed issues) of the referendum with the Constitution of the Russian Federation.

Performance of ChatGPT and GPT-4 on Neurosurgery Written Board Examinations.

Interest surrounding generative large language models (LLMs) has rapidly grown. Although ChatGPT (GPT-3.5), a general LLM, has shown near-passing performance on medical student board examinations, the performance of ChatGPT or its successor GPT-4 on specialized examinations and the factors affecting accuracy remain unclear. This study aims to assess the performance of ChatGPT and GPT-4 on a 500-question mock neurosurgical written board examination. The Self-Assessment Neurosurgery Examinations (SANS) American Board of Neurological Surgery Self-Assessment Examination 1 was used to evaluate ChatGPT and GPT-4. Questions were in single best answer, multiple-choice format. χ 2 , Fisher exact, and univariable logistic regression tests were used to assess performance differences in relation to question characteristics. ChatGPT (GPT-3.5) and GPT-4 achieved scores of 73.4% (95% CI: 69.3%-77.2%) and 83.4% (95% CI: 79.8%-86.5%), respectively, relative to the user average of 72.8% (95% CI: 68.6%-76.6%). Both LLMs exceeded last year's passing threshold of 69%. Although scores between ChatGPT and question bank users were equivalent ( P = .963), GPT-4 outperformed both (both P < .001). GPT-4 answered every question answered correctly by ChatGPT and 37.6% (50/133) of remaining incorrect questions correctly. Among 12 question categories, GPT-4 significantly outperformed users in each but performed comparably with ChatGPT in 3 (functional, other general, and spine) and outperformed both users and ChatGPT for tumor questions. Increased word count (odds ratio = 0.89 of answering a question correctly per +10 words) and higher-order problem-solving (odds ratio = 0.40, P = .009) were associated with lower accuracy for ChatGPT, but not for GPT-4 (both P > .005). Multimodal input was not available at the time of this study; hence, on questions with image content, ChatGPT and GPT-4 answered 49.5% and 56.8% of questions correctly based on contextual context clues alone. LLMs achieved passing scores on a mock 500-question neurosurgical written board examination, with GPT-4 significantly outperforming ChatGPT.

The 2022 PSOGI International Consensus on HIPEC Regimens for Peritoneal Malignancies: Epithelial Ovarian Cancer.

We report the results of an international consensus on hyperthermic intraperitoneal chemotherapy (HIPEC) regimens for epithelial ovarian cancer (EOC) performed with the following goals: To define the indications for HIPEC To identify the most suitableHIPEC regimens for each indication in EOC To identify areas of future research on HIPEC To provide recommendations for some aspects of perioperative care for HIPEC METHODS: The Delphi technique was used with two rounds of voting. There were three categories of questions: evidence-based recommendations [using theGrades of Recommendation, Assessment, Development, and Evaluation (GRADE) system with the patient, intervention, comparator, and outcome (PICO) method], an opinion survey, and research recommendations. Seventy-three (67.5%) of 108 invited experts responded in round I, and 68 (62.9%) in round II. Consensus was achieved for 34/38 (94.7%) questions. However, a strong positive consensus that would lead to inclusion in routine care was reached for only 6/38 (15.7%) questions. HIPEC in addition to interval cytoreductive surgery (CRS) received a strong positive recommendation that merits inclusion in routine care. Single-agent cisplatin was the only drug recommended for routine care, and OVHIPEC-1 was the most preferred regimen. The panel recommended performing HIPEC for a minimum of 60 min with a recommended minimum intraabdominal temperature of 41°C. Nephroprotection with sodium thiosulfate should be used for cisplatin HIPEC. The results of this consensus shouldguide clinicaldecisions on indications of HIPEC and the choice and various parameters of HIPEC regimens and could fill current knowledge gaps. These outcomes should be the basis for designing future clinical trials on HIPEC in EOC.

Instructor-assisted question classification system using machine learning algorithms with N-gram and weighting schemes

One aspect of natural language processing, text classification, has become necessary in the educational domain due to the increasing number of students and the COVID-19 outbreak. The advent of the devastating pandemic and the need to remain safe have surged the discussions around online learning and integrated modules in teaching and learning. In this study, we employed machine learning to develop an automatic instructor-assisted question classification module for learning management systems. In selecting the best classifier, the conventional and the ensemble machine learning algorithms were compared using the tenfold and the fivefold cross-validation techniques. In addition, the N-gram feature selection mechanism and three weighting schemes were evaluated for performance enhancement. The detailed analysis indicates that the ensemble algorithms outperform the conventional ones with decreasing accuracy as the N-gram size increases. For all compared algorithms, the AdaBoost (SVM) ensemble algorithm has the highest accuracy of 78.55% for Unigram (TP, TF, TF-IDF). In addition, the AdaBoost (SVM) emerged with the highest F1-score of 0.782, whiles the ensemble Bagging (RF) algorithm had the highest ROC value of 0.955 for Unigram (TP).

Sustainable Energy Siting, Affect, and Climate Mitigation: Questions for a Future Research Agenda

The affective sciences are essential to research regarding sustainable transition towards renewable energy. We focus on important questions that should be addressed by affective science in relation to the siting of large-scale renewable energy projects like wind and solar. Considering the recent acceleration of the transition, a more holistic understanding of negative and positive emotional responses to energy development will be essential. This is particularly important as the least controversial sites begin to dwindle in number. We break this commentary down into four primary question categories for future research. These are the causal relationships between emotional elicitations and their sources, dual processing through cognitive and affective response, affect and its relationship with place, and how all of these elements change over time.

Extending the reach of the point-to-set principle

The point-to-set principle of J. Lutz and N. Lutz (2018) has recently enabled the theory of computing to be used to answer open questions about fractal geometry in Euclidean spaces Rn. These are classical questions, meaning that their statements do not involve computation or related aspects of logic.In this paper we extend the reach of the point-to-set principle from Euclidean spaces to arbitrary separable metric spaces X. We first extend two algorithmic dimensions—computability-theoretic versions of classical Hausdorff and packing dimensions that assign dimensions dim⁡(x) and Dim(x) to individual pointsx∈X—to arbitrary separable metric spaces and to arbitrary gauge families. Our first two main results then extend the point-to-set principle to arbitrary separable metric spaces and to a large class of gauge families.We demonstrate the power of our extended point-to-set principle by using it to prove new theorems about classical fractal dimensions in hyperspaces.

Success of ChatGPT, an AI language model, in taking the French language version of the European Board of Ophthalmology examination: A novel approach to medical knowledge assessment

The purpose of this study was to evaluate the performance of ChatGPT, a cutting-edge artificial intelligence (AI) language model developed by OpenAI, in successfully completing the French language version of the European Board of Ophthalmology (EBO) examination and to assess its potential role in medical education and knowledge assessment. ChatGPT, based on the GPT-4 architecture, was exposed to a series of EBO examination questions in French, covering various aspects of ophthalmology. The AI's performance was evaluated by comparing its responses with the correct answers provided by ophthalmology experts. Additionally, the study assessed the time taken by ChatGPT to answer each question as a measure of efficiency. ChatGPT achieved a 91% success rate on the EBO examination, demonstrating a high level of competency in ophthalmology knowledge and application. The AI provided correct answers across all question categories, indicating a strong understanding of basic sciences, clinical knowledge, and clinical management. The AI model also answered the questions rapidly, taking only a fraction of the time needed by human test-takers. ChatGPT's performance on the French language version of the EBO examination demonstrates its potential to be a valuable tool in medical education and knowledge assessment. Further research is needed to explore optimal ways to implement AI language models in medical education and to address the associated ethical and practical concerns.

An Improved Model for Medical Forum Question Classification Based on CNN and BiLSTM

Question Classification (QC) is the fundamental task for Question Answering Systems (QASs) implementation, and is a vital task, as it helps in identifying the question category. It plays a big role in predicting the answer to a question while building a QAS. However, classifying medical questions is still a challenging task due to the complexity of medical terms. Many researchers have proposed different techniques to solve these problems, but some of these problems remain partially solved or unsolved. With the help of deep learning technology, various text-processing problems have become much easier to solve. In this paper, an improved deep learning-based model for Medical Forum Question Classification (MFQC) is proposed to classify medical questions. In the proposed model, feature representation is performed using Word2Vec, which is a word embedding model. Additionally, the features are extracted from the word embedding layer based on Convolutional Neural Networks (CNNs). Finally, a Bidirectional Long Short Term Memory (BiLSTM) network is used to classify the extracted features. The BiLSTM model analyzes the target information of the representation and then outputs the question category via a SoftMax layer. Our model achieves state-of-the-art performance by effectively capturing semantic and syntactic features from the input questions. We evaluate the proposed CNN-BiLSTM model on two benchmark datasets and compare its performance with existing methods, demonstrating its superiority in accurately categorizing medical forum questions.

DOMAS: Data Oriented Medical Visual Question Answering Using Swin Transformer

"The Medical Visual Question Answering problem is a joined Computer Vision and Natural Language Processing task that aims to obtain answers in natural language to a question, posed in natural language as well, regarding an image. Both the image and question are of a medical nature. In this paper we introduce DOMAS, a deep learning model that solves this task on the Med-VQA 2019 dataset. The method is based on dividing the task into smaller classification problems by using a BERT-based question classification and a unique approach that makes use of dataset information for selecting the suited model. For the image classification problems, transfer learning using a pre-trained Swin Transform based architecture is used. DOMAS uses a question classifier and seven image classifiers along with the image classifier selection strategy and achieves 0.616 strict accuracy and 0.654 BLUE score. The results are competitive with other state-of-the-art models, proving that our approach is effective in solving the presented task. 2010 Mathematics Subject Classification. 68T45, 68T50. 1998 CR Categories and Descriptors. I.2.7 [Artificial Intelligence]: Natural Language Processing – Language parsing and understanding; I.2.7 [Artificial Intelligence]: Applications and Expert Systems – Medicine and science. Key words and phrases. Medical Visual Question Answering, Swin Transformer."

Semantic Similarity Analysis for Examination Questions Classification Using WordNet

Question classification based on Bloom’s Taxonomy (BT) has been widely accepted and used as a guideline in designing examination questions in many institutions of higher learning. The misclassification of questions may happen when the classification task is conducted manually due to a discrepancy in the understanding of BT by academics. Hence, several automated examination question classification systems have been proposed by researchers to perform question classification accurately. Most of this research has focused on specific subject areas only or single-sentence type questions. There has been a lack of research on question classification for multi-sentence type and multi-subject questions. This paper proposes a question classification system (QCS) to perform the examination of question classification using a semantic and synthetic approach. The questions were taken from various subjects of an engineering diploma course, and the questions were either single- or multiple-sentence types. The QCS was developed using a natural language toolkit (NLTK), Stanford POS tagger (SPOS), Stanford parser’s universal dependencies (UD), and WordNet similarity approaches. The QCS used the NLTK to process the questions into sentences and then word tokens, such as SPOS, to tag the word tokens and UD to identify the important word tokens, which were the verbs of the examination questions. The identified verbs were then compared with the BT’s verbs list in terms of word sense using the WordNet similarity approach before finally classifying the questions according to BT. The developed QCS achieved an overall 83% accuracy in the classification of a set of 200 examination questions, according to BT.

Evaluation of occupational factors influencing the experience of an autopsy demonstration

Abstract Background Autopsy demonstrations have been an integral part of medical school, medical and nonmedical education for many decades; however, a direct comparison of the experience of an autopsy demonstration of different groups of participants is missing so far. In this study, the experience of such a situation by the different professional groups was analyzed and discussed in order to improve future dissections for the participant groups. Methods Evaluations of 54 autopsy demonstrations were performed at the Institute of Legal Medicine of the University Hospital Halle (Saale) and the branch office in Magdeburg in the period 2017–2020. A total of 794 questionnaires were answered and statistically analyzed. Participants rated the autopsies in various question categories and, if applicable, provided reasons for discontinuation of the demonstration on their part (e.g., circulatory dysregulation). To analyze the results the Kruskall Wallis test, the Mann-Whitney U-test and the χ2-test were applied. Results Findings showed that the professional background has a significant influence on the experience and the dropout rates from an autopsy. Furthermore, the professional groups have different areas of interest and expectations of an autopsy. Conclusion The results require customized preparation of an autopsy demonstration as well as differentiation of knowledge transfer to different groups of participants to optimize their benefit.

Short-term effects of a speech feminization program for transgender women: listener perceptions, self-perception and satisfaction of the voice

Purpose: This study measured and compared the short-term impact of pitch elevation training (PET) and articulation-resonance training (ART) in transgender women, on self-perception, satisfaction and masculinity-femininity perceptions of listeners. Methods: A randomized controlled study with cross-over design was used. Thirty transgender women were included and received fourteen weeks of speech training. All participants started with sham training (four weeks), after which they were randomly assigned to one of two groups: one group continued with PET (five weeks), followed by ART (five weeks), the second group received both trainings in opposite order. Participants were recorded four times, in between the training blocks: pre, post 1 (after sham), post 2 (after training 1) and post 3 (after training 2). Participants did a self-evaluation through the Trans Woman Voice Questionnaire (TWVQ) and visual analogues scales (VAS) concerning their self-perception and satisfaction. Two listening experiments (n = 75) were conducted researching the continuous masculinity-femininity rating (through a VAS) and categorical masculinity-femininity attribution. Results and conclusions: Transgender women perceive their voices more feminine after the training and experience a positive impact on the vocal functioning and the voice-related impact on their daily life. However, a lot of the participants acknowledge that they need more speech training after ten weeks. Listeners rate the participants’ voices more feminine after training, both during the continuous and categorical questions. Higher femininity scores were detected during self-perception and listener perceptions after the combination of both ART and PET, compared to the separate trainings. No order effects were detected between ART and PET, both for self-perception and listener perceptions. Defining outcome predictors is crucial in future research.

Assessing the efficacy of ChatGPT in addressing Chinese financial conundrums: An in-depth comparative analysis of human and AI-generated responses

ChatGPT, the latest iteration of OpenAI's natural language generation model, has found applications in a wide range of tasks such as question answering, text summarization, machine translation, classification, code generation, and dialogue A.I. Its potential in the financial industry has garnered significant attention. This paper aims to bridge the gap between chatGPT and human services in the financial domain, while also exploring the opportunities and challenges it presents in this industry. To comprehensively evaluate the processing capabilities of chatGPT in the financial field, we collected a dataset of n = 7165 financial questions and analyzed the perplexity value, emotion value, accuracy, professionalism, and real-time performance of both human-generated and chatGPT-generated content using machine learning algorithms and evaluation tests. The experimental results indicate that chatGPT exhibits higher levels of professionalism and accuracy compared to manual services, leading to improved efficiency, cost reduction, and enhanced customer satisfaction, thereby boosting the competitiveness and profitability of financial institutions. However, challenges such as a lack of emotional value in its responses, potential bias from one-sided training data, information errors, and the risk of job displacement need to be addressed. These findings provide theoretical and data-driven support for the future implementation of chatGPT in financial innovation and development.

Otwarty świat i narracja rozproszona w grze Elden Ring studia FromSoftware

&#x0D; &#x0D; &#x0D; The article Open World and Dispersed Narrative in FromSoftware’s “Elden Ring” tackles a number of design issues associated with open world design in video games. Building upon Tim Ingold’s dis- crepancy of two distinct modes of exploring the reality (wayfaring and trail-following), the author proceeds with analysing how the categories in question might have influenced the design of Elden Ring’s ‘The Lands Between’, which is torn between two different ambitions. One is to create a high fantasy narrative in allotopia, whilst the other is to sustain the principles of procedural rhetoric typical for soulsborne games, well-known for leaning towards a dark fantasy design style both in terms of their aesthetics and narrative content. Consequently, the main goal of this analysis is to propose a potential way of avoiding such inconsequence in open world game design, which is termed here a “ludotopian dissonance”.&#x0D; &#x0D; &#x0D;

Category-Guided Visual Question Generation (Student Abstract)

Visual question generation aims to generate high-quality questions related to images. Generating questions based only on images can better reduce labor costs and thus be easily applied. However, their methods tend to generate similar general questions that fail to ask questions about the specific content of each image scene. In this paper, we propose a category-guided visual question generation model that can generate questions with multiple categories that focus on different objects in an image. Specifically, our model first selects the appropriate question category based on the objects in the image and the relationships among objects. Then, we generate corresponding questions based on the selected question categories. Experiments conducted on the TDIUC dataset show that our proposed model outperforms existing models in terms of diversity and quality.