Application of supervised machine learning and Taylor diagrams for prognostic analysis of performance and emission characteristics of biogas-powered dual-fuel diesel engine DOI Creative Commons
Komarova Le, Minh Thai Duong, Dao Nam Cao

et al.

International Journal of Renewable Energy Development, Journal Year: 2024, Volume and Issue: 13(6), P. 1175 - 1190

Published: Oct. 27, 2024

In the ongoing search for an alternative fuel diesel engines, biogas is attractive option. Biogas can be used in dual-fuel mode with as pilot fuel. This work investigates modeling of injecting strategies a waste-derived biogas-powered engine. Engine performance and emissions were projected using supervised machine learning methods including random forest, lasso regression, support vector machines (SVM). Mean Squared Error (MSE), R-squared (R²), Absolute Percentage (MAPE) among criteria evaluations models. Random Forest has shown better Brake Thermal Efficiency (BTE) test R² 0.9938 low MAPE 3.0741%. once more exceeded other models 0.9715 4.2242% estimating Specific Energy Consumption (BSEC). With 0.9821 2.5801% emerged most accurate model according to carbon dioxide (CO₂) emission modeling. Analogous results monoxide (CO) prediction based on obtained 0.8339 3.6099%. outperformed Linear Regression 0.9756% 7.2056% case nitrogen oxide (NOx) emissions. showed constant overall criteria. paper emphasizes how well especially prognosticate engines.

Language: Английский

Fuzzy logic-supported building design for low-energy consumption in urban environments DOI Creative Commons

M. Arun,

Cristina Efremov, Van Nhanh Nguyen

et al.

Case Studies in Thermal Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 105384 - 105384

Published: Oct. 1, 2024

Language: Английский

Citations

12
Predicting Early Employability of Vietnamese Graduates: Insights from Data-Driven Analysis Through Machine Learning Methods DOI Creative Commons
Long‐Sheng Chen, Thao-Trang Huynh-Cam, Van-Canh Nguyen

et al.

Big Data and Cognitive Computing, Journal Year: 2025, Volume and Issue: 9(5), P. 134 - 134

Published: May 19, 2025

Graduate employability remains a crucial challenge for higher education institutions, especially in developing economies. This study investigates the key academic and vocational factors influencing early employment outcomes among recent graduates at public university Vietnam’s Mekong Delta region. By leveraging predictive analytics, research explores how data-driven approaches can enhance career readiness strategies. The analysis employed AI-driven models, particularly classification regression trees (CARTs), using dataset of 610 from to predict employability. input included gender, field study, entrance scores, grade point average (GPA) scores four years. output factor was graduates’ (un)employment within six months after graduation. Among all factors, third-year GPA, final-year performance are most significant predictors employment. tested CARTs achieved highest accuracy (93.6%), offering interpretable decision rules that inform curriculum design support services. contributes intersection artificial intelligence by providing actionable insights universities, policymakers, employers, supporting alignment with labor market demands improving graduate outcomes.

Language: Английский

Citations

0
Leveraging LLMs for Optimised Feature Selection and Embedding in Structured Data: A Case Study on Graduate Employment Classification DOI Creative Commons
Radiah Haque, Hui-Ngo Goh, Choo‐Yee Ting

et al.

Computers and Education Artificial Intelligence, Journal Year: 2024, Volume and Issue: 8, P. 100356 - 100356

Published: Dec. 22, 2024

Language: Английский

Citations

1
Emotion Recognition and Multi-class Classification in Music with MFCC and Machine Learning DOI Creative Commons

Gilsang Yoo,

Sungdae Hong,

Hyeocheol Kim

et al.

International Journal on Advanced Science Engineering and Information Technology, Journal Year: 2024, Volume and Issue: 14(3), P. 818 - 825

Published: June 3, 2024

Background music in OTT services significantly enhances narratives and conveys emotions, yet users with hearing impairments might not fully experience this emotional context. This paper illuminates the pivotal role of background user engagement on platforms. It introduces a novel system designed to mitigate challenges hearing-impaired face appreciating nuances music. adeptly identifies mood translates it into textual subtitles, making content accessible all users. The proposed method extracts key audio features, including Mel Frequency Cepstral Coefficients (MFCC), Root Mean Square (RMS), MEL Spectrograms. then harnesses power leading machine learning algorithms Logistic Regression, Random Forest, AdaBoost, Support Vector Classification (SVC) analyze traits embedded accurately identify its sentiment. Among these, Forest algorithm, applied MFCC demonstrated exceptional accuracy, reaching 94.8% our tests. significance technology extends beyond mere feature identification; promises revolutionize accessibility multimedia content. By automatically generating emotionally resonant can enrich viewing for all, particularly those impairments. advancement only underscores critical storytelling but also highlights vast potential enhancing inclusivity enjoyment digital entertainment across diverse audiences.

Language: Английский

Citations

0
Research on SVM Analysis Model of Influencing Factors of Employability of Graduates from Higher Vocational Colleges and Universities in Jiangxi Province DOI Open Access

K.T. Chen,

Jacquline Tham, Ali Khatibi

et al.

Applied Mathematics and Nonlinear Sciences, Journal Year: 2024, Volume and Issue: 9(1)

Published: Jan. 1, 2024

Abstract With the expansion of higher vocational colleges and universities, difficulty employment graduates has become an increasingly serious problem in development society. The reason for this phenomenon, addition to total pressure, lies difference between knowledge, ability, quality college students needs employers. This paper crawls information data universities from relevant websites, combines text classification method based on SVM analysis model mine system graduates’ takes it as a survey scale. Then Jiangxi Province’s were selected research object. After testing reliability validity scale, combined with independent samples, T-test regression analysis, other mathematical statistical methods explore factors affecting ability Province. Among them, there is no significant overall employability terms gender specialty category (P>0.05), while having or not work experience (P<0.05). training objectives strategies significantly contribute improvement professional competence graduates, which key factor influence employability. Accordingly, actively improve themselves at same time. Colleges should develop enterprises give necessary career guidance timely manner countermeasure suggestions.

Language: Английский

Citations

0
Application of supervised machine learning and Taylor diagrams for prognostic analysis of performance and emission characteristics of biogas-powered dual-fuel diesel engine DOI Creative Commons
Komarova Le, Minh Thai Duong, Dao Nam Cao

et al.

International Journal of Renewable Energy Development, Journal Year: 2024, Volume and Issue: 13(6), P. 1175 - 1190

Published: Oct. 27, 2024

In the ongoing search for an alternative fuel diesel engines, biogas is attractive option. Biogas can be used in dual-fuel mode with as pilot fuel. This work investigates modeling of injecting strategies a waste-derived biogas-powered engine. Engine performance and emissions were projected using supervised machine learning methods including random forest, lasso regression, support vector machines (SVM). Mean Squared Error (MSE), R-squared (R²), Absolute Percentage (MAPE) among criteria evaluations models. Random Forest has shown better Brake Thermal Efficiency (BTE) test R² 0.9938 low MAPE 3.0741%. once more exceeded other models 0.9715 4.2242% estimating Specific Energy Consumption (BSEC). With 0.9821 2.5801% emerged most accurate model according to carbon dioxide (CO₂) emission modeling. Analogous results monoxide (CO) prediction based on obtained 0.8339 3.6099%. outperformed Linear Regression 0.9756% 7.2056% case nitrogen oxide (NOx) emissions. showed constant overall criteria. paper emphasizes how well especially prognosticate engines.

Language: Английский

Citations

0