251 research outputs found
A Hybrid Approach to Music Recommendations for Improving ADHD Productivity
No full textThis study explores the development of a web application designed to enhance focus, motivation, and productivity in individuals with ADHD by inte-grating music recommendation algorithms, task management tools, and note-taking features. Music recommendations were generated using a hybrid approach combining Collaborative Filtering and Content-based Filtering to create tailored playlists based on tracks users “liked” during the procedure. The application was evaluated through a systematic framework using the Pomodoro Technique, where twenty ADHD-diagnosed students (aged 18+) completed two 20-minute sessions, each followed by a 5-minute break in between. In the first session, participants were asked to listen to one of three pre-selected playlists (Lo-Fi, Classical, Binaural Beats) and indicated their preferences by liking tracks while performing focused tasks. Using participants’ selections from the first session, the recommendation model generated a personalized "For You" playlist during the break, which they engaged with under identical condi-tions in the second session. A mixed-methods analysis was then used to combine quantitative data from Likert scale ratings and qualitative feedback from open-ended responses and structured questionnaires. The results of this study revealed significant improvements in all key areas, supporting the effectiveness of personalized music recommendations in academic and professional settings. Future work will focus on refining the application and expanding the recommendation system to accommodate a broader range of musical preferences
Diagnosis of Polycystic Ovarian Syndrome (PCOS) Using Deep Learning
No full textPolycystic Ovarian Syndrome (PCOS) is a silent disorder that causes women to have weight gain, infertility, hair loss, and irregular menstrual cycles. It is a complex health issue, and one of the methods to diagnose patients with PCOS is to count the number of follicles in the ovaries. The issue with the traditional method is that it is time-consuming and prone to human errors as it can be challenging for medical professionals to distinguish between healthy ovaries and polycystic ovaries. Using Deep Learning, the concept was to create and use various Deep Learning Models such as a CNN, Custom VGG-16, ResNet- 50, and Custom ResNet-50, to obtain a high-accuracy result that will detect between healthy and polycystic ovaries. From the results and evaluation obtained, the CNN model achieved 99% accuracy, VGG 16 model: 58%, ResNet-50 Model: 58%, and Custom ResNet-50 Model: 96.7%.</p
Exploring the Effects of Gamification in Assisting Students Maintain a Better Work-Life Balance
No full textThe concept of work-life balance has become difficult for students to grasp in a world where the distinction between work and other activities in life is a blur. The aim of the study was to investigate the effects of gamification in apps on student’s perceived work-life balance. Participants were divided into a gamified app group and a non-gamified app group and were given online questionnaires before and after the experiment, which lasted two weeks. Several data analysis methods were adopted such as descriptive statistics, inferential statistics, and thematic analysis. While the non-gamified app group experienced a greater positive change in perceived work-life balance, the gamified app group showed higher satisfaction in work-life balance. Gamification did not significantly affect autonomy and competence satisfaction but increased relatedness satisfaction and engagement levels. Rewards and achievements were identified as the most effective game elements. Further research is needed to explore the effects of gamification on work-life balance in-depth as research on gamification continues to expand.</p
Interaction Design Strategies for ADHD Learning Attention - A Review
No full textInteraction design and development strategies contain the proficiency to mimic user attention and engagement. These strategies are proven to be beneficial for learning attention while employing various types of digital learning platforms. However, special education need (SEN) learners possess special needs due to their neurodevelopmental disorder and deficiency of proper brain functioning and there-fore giving rise to malfunctioning of inhibitory control, sustained attention, and working memory. Hence, there is a need to develop user-centric interaction design and development strategies and implications to increase the attention span of ADHD (attention deficit hyperactivity disrorder) while learning. In this literature review, we aim to highlight the attention problem of ADHD due to malfunctioning of executive functioning and working memory. We have summarized the existing IxD (interaction design)based solutions for ADHD learning attention. The related limitations, chal-lenges and findings of the literature review are presented along with the future possi-bilities. This paper highlights the need to develop user-centric solutions for ADHD attention improvement during learning and the incorporation of the machine learning and artificial intellegnce based interfaces for the advance user-centric solutions
OntoCOVID:Ontology for Semantic Modeling of COVID19 Statistical Data
No full textSeveral COVID19 statistical datasets are provided to support stakeholders for better planning and decision making in healthcare. However, the datasets are in heterogeneous proprietary formats that create data silos and compatibility issues and make data discovery and reuse difficult. Further, the data integration for analysis is difficult and is performed by the domain experts manually which is time consuming and error prone. Therefore, an explicit, flexible, and widely acceptable methodology is needed to represent, store, query, and visualize COVID19 statistical data in the datasets. In this paper, we have presented the design and development of OntoCOVID ontology for representing, organizing, sharing, and reusing COVID19 statistical data in the datasets. The OntoCOVID is a lightweight ontology providing definitions of classes, properties, and axioms to semantically represent and relate information in the COVID19 statistical datasets. The OntoCOVID is evaluated to demonstrate its completeness and information retrieval for different use-case scenarios. The results obtained are promising and advocate for the improved ontological design and applications of the OntoCOVID.</p
Adoption of a Performance Evaluation Technique for the Development of a Framework for the Climatic Responsive Urban Design (CRUD)
No full textThis study investigates the effects of urban design elements i.e. street canyon geometry (Canyon
length, width, height, orientation, and SVF) and greenery on the urban microclimate using remote
sensing and computational fluid dynamics-based techniques. The important conclusion derived
from the current research can be used by designers for climate-responsive urban design in case
of the future extension of these cities
Malware Prediction Using LSTM Networks
No full textWith a recent increase in the use of the Internet, there has been a rise in malware attacks. Malware attacks can lead to stealing confidential data or make the target a source of further attacks. The detection of malware has been posing a unique challenge. Malware analysis is the study of malicious code to prevent cyber-attacks and vulnerability assessment. This article aims for classification of malware using a deep learning model to obtain an accurate and efficient performance. The system proposed in this study extracts a number of features and trains the long short-term memory (LSTM) model. The study utilises hyper-parameter tuning to improve the accuracy and efficiency of the LSTM model. The findings revealed 99.65% accuracy using sigmoid function that outperforms other activation function. This work can be helpful in malware detection to improve security posture.</p
Study of Abrar-ul- Haq's Punjabi Bhangra Songs in Pragmatics
No full textThe study of language and culture is known as linguistic anthropology. Etymological human sciences have become an interdisciplinary subject of study by utilizing the theoretical underpinnings of numerous disciplines. The current study investigated Punjabi songs using a qualitative analytical approach. The Relevance Theory of Pragmatics, proposed by Deirdre Wilson and Dan Sperber (2004), explained how Abrar-ul-Haq appeared on the horizon to promote Punjabi culture through his energetic Punjabi Bhangra songs. For this study, only two songs were chosen: Billo Day Ghar and Beh Ja Sakal Tay. The study’s findings revealed that Abrar ul Haq used singing as a medium of language to transport cultural norms in general and Punjabi culture in particular. Because he was a famous and world-famous singer, his message spread worldwide, and he became known not only as the Bhangra singer but also as the anthropologist who invented and promoted Punjabi culture through his singing. It is concluded that Punjabi Culture is rich in norms and values. MPhil scholars will expand the scope of the study to include a full-length version of these to promote Punjabi culture. It is suggested that other linguists investigate Punjabi culture to revive and preserve it
Deep Spatiotemporal Network-Based Spontaneous Macro- and Micro-facial Expression Recognition
No full textFacial expressions are the primary means of communicating human emotions, and their interpretation has earned significant interest from researchers due to their wide range of practical applications. However, there is a lack of research on the simultaneous recognition of spontaneous macro-expressions and micro-expressions. This paper aims to develop an end-to-end framework for the effective recognition of both spontaneous macro- and micro-expressions. The proposed framework utilizes Volume Local Directional Number (VLDN) for spatiotemporal feature extraction and ResNet101 for extracting deep spatial features from each frame. Additionally, we designed a Gated Recurrent Unit (GRU) to effectively learn the spatiotemporal features. Finally, we conduct comprehensive experiments on the CAS(ME)2 dataset to demonstrate the performance of our proposed method.</p
Predicting Commodity Prices in Futures Market Using Machine Learning
No full textThis study enhances predictive modelling of gold prices by employing advanced machine learning techniques, specifically Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs). Introducing an innovative Event Impact Score into the CNN-LSTM model, an improved forecasting accuracy was achieved compared with the models without this approach. This novel methodology that calculates an Event Impact Score based on economic events derived from an economic calendar dataset, quantifies the impact of significant economic indicators on commodity prices by assessing the associated price changes. Incorporating these scores, the CNN-LSTM model is adapted to accommodate the nuanced influence of external economic factors, offering a more refined analysis than conventional models
- …
