Academic misconduct poses a growing challenge for higher education institutions worldwide. While AI presents valuable opportunities for learning enhancement, Unauthorized Content Generation (UCG) poses a significant t...
详细信息
ISBN:
(数字)9783031643156
ISBN:
(纸本)9783031643149;9783031643156
Academic misconduct poses a growing challenge for higher education institutions worldwide. While AI presents valuable opportunities for learning enhancement, Unauthorized Content Generation (UCG) poses a significant threat to academic integrity. This paper addresses the challenges posed by UCG and explores innovative approaches to detection, focusing on the underutilised concept of authorship verification (AV). Despite the recognition of AV's potential, its application in education has been limited. This study investigates the feasibility of utilising students' academic writing profiles for AV to detect contract cheating and unacknowledged AI usage in academic contexts. Building upon previous research, this study enhances the existing Feature Vector Difference (FVD) AV method by introducing improvements to support better analysis, explainability, and interpretability of the classification process in an educational context. The refined classifier provides probability-based outputs, offering a transparent alternative to traditional "black box" binary outputs, and is able to identify stylometric features suitable for differentiating student's writing profiles. Through this research, we contribute to the advancement of AV technology in education towards explainability, providing educators with a valuable tool to uphold academic integrity and combat the proliferation of UCG in educational environments.
Previous studies have evidenced that musical training can change the brain functional and structural organizations, but it is still unclear how interactions within and between functional networks are affected by music...
详细信息
ISBN:
(纸本)9789819705757;9789819705764
Previous studies have evidenced that musical training can change the brain functional and structural organizations, but it is still unclear how interactions within and between functional networks are affected by musical training. Using the resting-state fMRI dataset with a relatively large sample, the present study examined the effects of musical training on inter- and intra-network functional connectivity (FC). The results revealed the decreased inter- and intra-network FC extensively which reflect greater movement efficiency and automaticity as well as five pairs of increased inter-network FC that possibly refer to cognitive function in participants with musical training compared to their counterparts without musical training. The current study provided a new perspective that musical training can induce the brain network changes.
Regional poetry is an important part of Chinese intangible cultural heritage;a characteristic example is the poetry of the Hexi Corridor, an important historical region with a rich poetry tradition located in the mode...
详细信息
ISBN:
(纸本)9783031659898;9783031659904
Regional poetry is an important part of Chinese intangible cultural heritage;a characteristic example is the poetry of the Hexi Corridor, an important historical region with a rich poetry tradition located in the modern western Gansu province of China. In this paper, we present our work on an ontology-based model for Hexi regional poetry that captures historical and cultural information including poetic imageries and allusions, based on CIDOC-CRM and other standard Semantic Web ontologies. Our work lays the foundation for transforming digitised poetry into a knowledge graph and facilitating the research of the history and culture of the Hexi region.
In the speech enhancement (SE) model, using auxiliary loss based on acoustic parameters can improve enhancement effects. However, currently used acoustic parameters focus on frequency domain information, neglecting th...
详细信息
ISBN:
(纸本)9789819706006;9789819706013
In the speech enhancement (SE) model, using auxiliary loss based on acoustic parameters can improve enhancement effects. However, currently used acoustic parameters focus on frequency domain information, neglecting the importance of time domain information. This study explores the effects of training the SE model utilizing temporal envelope parameters as auxiliary losses on perceptual quality and intelligibility, primarily focusing on accurately extracting temporal envelope parameters and optimizing losses. Initially, the effectiveness and robustness of the temporal envelope parameters in optimizing the SE model were verified. Subsequently, a temporal envelope loss based on sub-band weighted (ENVLoss) is proposed based on the perceptual characteristics of human hearing. Then, a method for constructing a joint loss function is proposed, integrating temporal and spectral acoustic features to promote the enhancement effect. Experimental results show that temporal envelope characteristics improve both the time domain and the time-frequency domain SE models. Compared to other acoustic parameter losses, the SE model using a sub-band temporal envelope auxiliary loss shows improvement in the PESQ, STOI, and MOS estimation metrics.
Streaming games is increasingly popular amongst people of childbearing age. So, what happens when streamers become parents? There are few accounts, scarce advice and a dearth of research. Streaming seems incompatible ...
详细信息
ISBN:
(数字)9783031514524
ISBN:
(纸本)9783031514517;9783031514524
Streaming games is increasingly popular amongst people of childbearing age. So, what happens when streamers become parents? There are few accounts, scarce advice and a dearth of research. Streaming seems incompatible with parenting due to its tensions with familial responsibilities. However, it is also conceivable that parent and gamer identities can be reconciled to yield successful outcomes. This exploratory study investigates the identity and experience of parent streamers. Six parent streamers, mothers and fathers from an online Portuguese community, were interviewed. Thematic analysis revealed several foci: (i) the difficulty of managing time effectively;(ii) the tension between work and play;(iii) the parent streamer identity (particularly, differences between mothers and fathers);(iv) the criticality of familial support;(v) the increasingly complex relationship between parents and games;(vi) children's interference in streaming practices;and (vii) the benefits of streaming including communal connection and improved mental health. These findings highlight how online media increasingly challenges the way in which modern parents navigate parenthood and their own personal lives. They also pave the way for evidence-led guidance that can support parent steamers.
When children start to learn how to read in primary school, they are given children's books, collections of tales, to read in class or at home, but not all children enjoy books to the fullest, and worse, many do n...
详细信息
ISBN:
(数字)9783031514524
ISBN:
(纸本)9783031514517;9783031514524
When children start to learn how to read in primary school, they are given children's books, collections of tales, to read in class or at home, but not all children enjoy books to the fullest, and worse, many do not like reading at all. The aim of this paper is to propose a method to adapt such books into video games, hence giving the possibility to read the same stories in a different, interactive medium. Children being familiar with video games, and being avid players, they may more easily enjoy a story through this medium rather than another. But as we work on the matter at hand, two main issues arise, which we will discuss further below. How to adapt from text, often illustrated, to video game? Considering the target audience of this effort, how are we going to make it understandable, legible, and accessible to young children? Said questions aremeant to help developers and researchers understand better how to work with children's literature and children's games having always their audience in mind.
Large Language Models (LLMs), such as Generative Pre-trained Transformers (GPTs), have demonstrated remarkable capabilities in natural language processing (NLP). However, these models often encounter challenges such a...
详细信息
ISBN:
(数字)9783031643156
ISBN:
(纸本)9783031643149;9783031643156
Large Language Models (LLMs), such as Generative Pre-trained Transformers (GPTs), have demonstrated remarkable capabilities in natural language processing (NLP). However, these models often encounter challenges such as inaccuracies and hallucinations, which can undermine their utility. Retrieval-Augmented Generation (RAG) has emerged as a promising approach to enhance model accuracy and reliability by integrating external databases. This study investigates the use of RAG to improve the accuracy of GPT models in educational settings, particularly within the realm of Massive Open Online Courses (MOOCs). Through a comparative analysis of various GPT model iterations, we observed a significant improvement in accuracy, increasing from 60% with GPT-3.5 to 80% using the RAG-augmented GPT-4. This enhancement highlights the considerable potential of RAG-augmented GPT models in improving the accuracy of content generation. Such enhanced accuracy suggests revolutionizing assessment methodologies and learning experiences, fostering an educational environment that is more interactive and tailored to individual needs.
The educational landscape is evolving with the integration of AI, large language models (LLMs), and generative AI, requiring educators to adopt state-of-the-art technologies and strategies in their pedagogical practic...
详细信息
ISBN:
(纸本)9783031643118;9783031643125
The educational landscape is evolving with the integration of AI, large language models (LLMs), and generative AI, requiring educators to adopt state-of-the-art technologies and strategies in their pedagogical practices. Pedagogical Design Patterns (PDPs) have garnered attention for disseminating best practices and bridging the gap between research and practice. However, their widespread adoption is hindered by limited publicly available resources and fragmented publishing platforms. To address this, we propose leveraging LLMs to recommend pedagogical practices, drawing from existing PDPs. Our model utilizes a local knowledge base and the Retrieval Augmented Generation (RAG) framework to create query contexts for LLM prompts. Initial findings show promise, with an accuracy score of 0.83 and high relevance of recommendations to input queries. This study presents early results of our ongoing project, supporting further development of the model. The proposed system aims to empower novice educators by providing expert wisdom to enrich their teaching methodologies.
In this paper, a goal-setting oriented adaptive learning paths recommendation system is designed and implemented, aiming to provide learners with targeted and effective preparation for their career entry/reorientation...
详细信息
ISBN:
(数字)9783031616785
ISBN:
(纸本)9783031616778;9783031616785
In this paper, a goal-setting oriented adaptive learning paths recommendation system is designed and implemented, aiming to provide learners with targeted and effective preparation for their career entry/reorientation based on competences taxonomies. This system combines the principles of the classic recommendation system by utilizing a hybrid method that integrates content-based and knowledge-based recommendations. Complementary to the hybrid method, the SMART goal-setting approach is applied, assisting learners in setting specific, measurable, achievable, relevant, and time-bound learning objectives. The primary hypothesis explored in this work is to what extent the combination of competence-based course recommendations and goal-setting offers learners a goal-setting oriented learning path optimizing their educational journey. A mixed-methods approach was utilized, grounding system development in key theories and taxonomies. Ultimately, both quantitative and qualitative user testing were conducted to evaluate the effectiveness and usability of the prototype and to discuss the hypothesis of the research.
In the context of smart cities, where the deployment of surveillance systems and security cameras is becoming increasingly ubiquitous, the efficient management of digital images and their confidentiality has become a ...
详细信息
ISBN:
(数字)9783031525179
ISBN:
(纸本)9783031525162;9783031525179
In the context of smart cities, where the deployment of surveillance systems and security cameras is becoming increasingly ubiquitous, the efficient management of digital images and their confidentiality has become a critical challenge. In this work, we present an innovative scheme which considers two components: compressive sensing and S-Boxes for image compression and confusion property in the Shannon's information theory context, respectively. The integration of these two building blocks provides a comprehensive solution for the efficient and secure transmission of image data in urban environments. Our scheme expands the compressed image into a 24-bit RGB image and uses three S-Boxes to replace the information of each color channel. One of the new features is that the S-Boxes evolve based on a key. In this sense, the scheme offers a solution for smart cities aiming to optimize the management of digital image data and simultaneously achieving the security of transmitted information. The processed images have been analyzed, and obtained to show that our scheme brings perceptual and cryptographic security to digital images, without compromising the recovered image. Its implementation can significantly contribute to efficiency and security, in the use of surveillance cameras in modern urban environments of smart cities.
暂无评论