Applied System Innovation

Research

Jump to: Review

22 pages, 1609 KiB

Open AccessArticle

Evaluation of the Performance of Neural and Non-Neural Methods to Classify the Severity of Work Accidents Occurring in the Footwear Industry Complex

by Jonhatan Magno Norte da Silva, Maria Luiza da Silva Braz, Joel Gomes da Silva, Lucas Gomes Miranda Bispo, Wilza Karla dos Santos Leite and Elamara Marama de Araujo Vieira

Appl. Syst. Innov. 2024, 7(5), 85; https://doi.org/10.3390/asi7050085 - 15 Sep 2024

Viewed by 822

Abstract

In the footwear industry, occupational risks are significant, and work accidents are frequent. Professionals in the field prepare documents and reports about these accidents, but the need for more time and resources limits learning based on past incidents. Machine learning (ML) and deep [...] Read more.

In the footwear industry, occupational risks are significant, and work accidents are frequent. Professionals in the field prepare documents and reports about these accidents, but the need for more time and resources limits learning based on past incidents. Machine learning (ML) and deep learning (DL) methods have been applied to analyze data from these documents, identifying accident patterns and classifying the damage’s severity. However, evaluating the performance of these methods in different economic sectors is crucial. This study examined neural and non-neural methods for classifying the severity of workplace accidents in the footwear industry complex. The random forest (RF) and extreme gradient boosting (XGBoost) methods were the most effective non-neural methods. The neural methods 1D convolutional neural networks (1D-CNN) and bidirectional long short-term memory (Bi-LSTM) showed superior performance, with parameters above 98% and 99%, respectively, although with a longer training time. It is concluded that using these methods is viable for classifying accidents in the footwear industry. The methods can classify new accidents and simulate scenarios, demonstrating their adaptability and reliability in different economic sectors for accident prevention. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

17 pages, 4163 KiB

Open AccessArticle

An Image-Retrieval Method Based on Cross-Hardware Platform Features

by Jun Yin, Fei Wu and Hao Su

Appl. Syst. Innov. 2024, 7(4), 64; https://doi.org/10.3390/asi7040064 - 23 Jul 2024

Viewed by 979

Abstract

Artificial intelligence (AI) models have already achieved great success in fields such as computer vision and natural language processing. However, deploying AI models based on heterogeneous hardware is difficult to ensure accuracy consistency, especially for precision sensitive feature-based image retrieval. In this article, [...] Read more.

Artificial intelligence (AI) models have already achieved great success in fields such as computer vision and natural language processing. However, deploying AI models based on heterogeneous hardware is difficult to ensure accuracy consistency, especially for precision sensitive feature-based image retrieval. In this article, we realize an image-retrieval method based on cross-hardware platform features, aiming to prove that the features of heterogeneous hardware platforms can be mixed, in which the Huawei Atlas 300V and NVIDIA TeslaT4 are used for experiments. First, we compared the decoding differences of heterogeneous hardware, and used CPU software decoding to help hardware decoding improve the decoding success rate. Then, we compared the difference between the Atlas 300V and TeslaT4 chip architectures and tested the differences between the two platform features by calculating feature similarity. In addition, the scaling mode in the pre-processing process was also compared to further analyze the factors affecting feature consistency. Next, the consistency of capture and correlation based on video structure were verified. Finally, the experimental results reveal that the feature results from the TeslaT4 and Atlas 300V can be mixed for image retrieval based on cross-hardware platform features. Consequently, cross-platform image retrieval with low error is realized. Specifically, compared with the Atlas 300V hard and CPU soft decoding, the TeslaT4 hard decoded more than 99% of the image with a decoding pixel maximum difference of +1/−1. From the average of feature similarity, the feature similarity between the Atlas 300V and TeslaT4 exceeds 99%. The difference between the TeslaT4 and Atlas 300V in recall and mAP in feature retrieval is less than 0.1%. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

23 pages, 3706 KiB

Open AccessArticle

A Residual Deep Learning Method for Accurate and Efficient Recognition of Gym Exercise Activities Using Electromyography and IMU Sensors

by Sakorn Mekruksavanich and Anuchit Jitpattanakul

Appl. Syst. Innov. 2024, 7(4), 59; https://doi.org/10.3390/asi7040059 - 2 Jul 2024

Cited by 1 | Viewed by 1702

Abstract

The accurate and efficient recognition of gym workout activities using wearable sensors holds significant implications for assessing fitness levels, tailoring personalized training regimens, and overseeing rehabilitation progress. This study introduces CNN-ResBiGRU, a novel deep learning architecture that amalgamates residual and hybrid methodologies, aiming [...] Read more.

The accurate and efficient recognition of gym workout activities using wearable sensors holds significant implications for assessing fitness levels, tailoring personalized training regimens, and overseeing rehabilitation progress. This study introduces CNN-ResBiGRU, a novel deep learning architecture that amalgamates residual and hybrid methodologies, aiming to precisely categorize gym exercises based on multimodal sensor data. The primary goal of this model is to effectively identify various gym workouts by integrating convolutional neural networks, residual connections, and bidirectional gated recurrent units. Raw electromyography and inertial measurement unit data collected from wearable sensors worn by individuals during strength training and gym sessions serve as inputs for the CNN-ResBiGRU model. Initially, convolutional neural network layers are employed to extract unique features in both temporal and spatial dimensions, capturing localized patterns within the sensor outputs. Subsequently, the extracted features are fed into the ResBiGRU component, leveraging residual connections and bidirectional processing to capture the exercise activities’ long-term temporal dependencies and contextual information. The performance of the proposed model is evaluated using the Myogym dataset, comprising data from 10 participants engaged in 30 distinct gym activities. The model achieves a classification accuracy of 97.29% and an F1-score of 92.68%. Ablation studies confirm the effectiveness of the convolutional neural network and ResBiGRU components. The proposed hybrid model uses wearable multimodal sensor data to accurately and efficiently recognize gym exercise activity. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

13 pages, 5379 KiB

Open AccessArticle

A Comparative Analysis of Oak Wood Defect Detection Using Two Deep Learning (DL)-Based Software

by Branimir Jambreković, Filip Veselčić, Iva Ištok, Tomislav Sinković, Vjekoslav Živković and Tomislav Sedlar

Appl. Syst. Innov. 2024, 7(2), 30; https://doi.org/10.3390/asi7020030 - 15 Apr 2024

Viewed by 1692

Abstract

The world’s expanding population presents a challenge through its rising demand for wood products. This requirement contributes to increased production and, ultimately, the high-quality and efficient utilization of basic materials. Detecting defects in wood elements, which are inevitable when working with a natural [...] Read more.

The world’s expanding population presents a challenge through its rising demand for wood products. This requirement contributes to increased production and, ultimately, the high-quality and efficient utilization of basic materials. Detecting defects in wood elements, which are inevitable when working with a natural material such as wood, is one of the difficulties associated with the issue above. Even in modern times, people still identify wood defects by visually scrutinizing the sawn surface and marking the defects. Industrial scanners equipped with software based on convolutional neural networks (CNNs) allow for the rapid detection of defects and have the potential to accelerate production and eradicate human subjectivity. This paper evaluates the suitability of defect recognition software in industrial scanners against software specifically designed for this task within a research project conducted using Adaptive Vision Studio, focusing on feature detection techniques. The research revealed that the software installed as part of the industrial scanner is more effective for analyzing knots (77.78% vs. 70.37%), sapwood (100% vs. 80%), and ambrosia wood (60% vs. 20%), while the software derived from the project is more effective for analyzing cracks (70% vs. 65%), ingrown bark (42.86% vs. 28.57%), and wood rays (81.82% vs. 27.27%). Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

19 pages, 8245 KiB

Open AccessArticle

Deep Learning Method to Detect Missing Welds for Joist Assembly Line

by Hamed Raoofi, Asa Sabahnia, Daniel Barbeau and Ali Motamedi

Appl. Syst. Innov. 2024, 7(1), 16; https://doi.org/10.3390/asi7010016 - 13 Feb 2024

Cited by 1 | Viewed by 2484

Abstract

Traditional methods of supervision in the construction industry are time-consuming and costly, requiring significant investments in skilled labor. However, with advancements in artificial intelligence, computer vision, and deep learning, these methods can now be automated, resulting in time and cost savings, as well [...] Read more.

Traditional methods of supervision in the construction industry are time-consuming and costly, requiring significant investments in skilled labor. However, with advancements in artificial intelligence, computer vision, and deep learning, these methods can now be automated, resulting in time and cost savings, as well as improvements in product quality. This research focuses on the application of computer vision approaches to monitor the quality of welding in prefabricated steel elements. A high-performance network was designed, consisting of a video capturing station, a customized classifier based on a YOLOv4 detector and an IoU tracker, and a user interface software for any interaction with quality control workers. The network demonstrated over 98% accuracy in identifying steel connection types and detecting missed welds on the assembly line in real-time. Extensive validation was conducted using a large dataset from a real production environment. The proposed framework aims to reduce rework, minimize hazards, and enhance product quality. This research contributes to the automation of quality control processes in the construction industry. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

20 pages, 1720 KiB

Open AccessArticle

Research on Chinese Nested Entity Recognition Based on IDCNNLR and GlobalPointer

by Weijun Li, Jintong Liu, Yuxiao Gao, Xinyong Zhang and Jianlai Gu

Appl. Syst. Innov. 2024, 7(1), 8; https://doi.org/10.3390/asi7010008 - 8 Jan 2024

Viewed by 2452

Abstract

The task of named entity recognition (NER) is to identify entities in the text and predict their categories. In real-life scenarios, the context of the text is often complex, and there may exist nested entities within an entity. This kind of entity is [...] Read more.

The task of named entity recognition (NER) is to identify entities in the text and predict their categories. In real-life scenarios, the context of the text is often complex, and there may exist nested entities within an entity. This kind of entity is called a nested entity, and the task of recognizing entities with nested structures is referred to as nested named entity recognition. Most existing NER models can only handle flat entities, and there has been limited research progress in Chinese nested named entity recognition, resulting in relatively few models in this direction. General NER models have limited semantic extraction capabilities and cannot capture deep semantic information between nested entities in the text. To address these issues, this paper proposes a model that uses the GlobalPointer module to identify nested entities in the text and constructs the IDCNNLR semantic extraction module to extract deep semantic information. Furthermore, multiple-head self-attention mechanisms are incorporated into the model at multiple positions to achieve data denoising, enhancing the quality of semantic features. The proposed model considers each possible entity boundary through the GlobalPointer module, and the IDCNNLR semantic extraction module and multi-position attention mechanism are introduced to enhance the model’s semantic extraction capability. Experimental results demonstrate that the proposed model achieves F1 scores of 69.617% and 79.285% on the CMeEE Chinese nested entity recognition dataset and CLUENER2020 Chinese fine-grained entity recognition dataset, respectively. The model exhibits improvement compared to baseline models, and each innovation point shows effective performance enhancement in ablative experiments. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

21 pages, 4029 KiB

Open AccessArticle

Stock Market Prediction Using Deep Reinforcement Learning

by Alamir Labib Awad, Saleh Mesbah Elkaffas and Mohammed Waleed Fakhr

Appl. Syst. Innov. 2023, 6(6), 106; https://doi.org/10.3390/asi6060106 - 10 Nov 2023

Cited by 9 | Viewed by 15702

Abstract

Stock value prediction and trading, a captivating and complex research domain, continues to draw heightened attention. Ensuring profitable returns in stock market investments demands precise and timely decision-making. The evolution of technology has introduced advanced predictive algorithms, reshaping investment strategies. Essential to this [...] Read more.

Stock value prediction and trading, a captivating and complex research domain, continues to draw heightened attention. Ensuring profitable returns in stock market investments demands precise and timely decision-making. The evolution of technology has introduced advanced predictive algorithms, reshaping investment strategies. Essential to this transformation is the profound reliance on historical data analysis, driving the automation of decisions, particularly in individual stock contexts. Recent strides in deep reinforcement learning algorithms have emerged as a focal point for researchers, offering promising avenues in stock market predictions. In contrast to prevailing models rooted in artificial neural network (ANN) and long short-term memory (LSTM) algorithms, this study introduces a pioneering approach. By integrating ANN, LSTM, and natural language processing (NLP) techniques with the deep Q network (DQN), this research crafts a novel architecture tailored specifically for stock market prediction. At its core, this innovative framework harnesses the wealth of historical stock data, with a keen focus on gold stocks. Augmented by the insightful analysis of social media data, including platforms such as S&P, Yahoo, NASDAQ, and various gold market-related channels, this study gains depth and comprehensiveness. The predictive prowess of the developed model is exemplified in its ability to forecast the opening stock value for the subsequent day, a feat validated across exhaustive datasets. Through rigorous comparative analysis against benchmark algorithms, the research spotlights the unparalleled accuracy and efficacy of the proposed combined algorithmic architecture. This study not only presents a compelling demonstration of predictive analytics but also engages in critical analysis, illuminating the intricate dynamics of the stock market. Ultimately, this research contributes valuable insights and sets new horizons in the realm of stock market predictions. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

20 pages, 2302 KiB

Open AccessArticle

Personalized E-Learning Recommender System Based on Autoencoders

by Lamyae El Youbi El Idrissi, Ismail Akharraz and Abdelaziz Ahaitouf

Appl. Syst. Innov. 2023, 6(6), 102; https://doi.org/10.3390/asi6060102 - 27 Oct 2023

Cited by 8 | Viewed by 4055

Abstract

Through the Internet, learners can access available information on e-learning platforms to facilitate their studies or to acquire new skills. However, finding the right information for their specific needs among the numerous available choices is a tedious task due to information overload. Recommender [...] Read more.

Through the Internet, learners can access available information on e-learning platforms to facilitate their studies or to acquire new skills. However, finding the right information for their specific needs among the numerous available choices is a tedious task due to information overload. Recommender systems are a good solution to personalize e-learning by proposing useful and relevant information adapted to each learner using a set of techniques and algorithms. Collaborative filtering (CF) is one of the techniques widely used in such systems. However, the high dimensions and sparsity of the data are major problems. Since the concept of deep learning has grown in popularity, various studies have emerged to improve this form of filtering. In this work, we used an autoencoder, which is a powerful model in data dimension reduction, feature extraction and data reconstruction, to learn and predict student preferences in an e-learning recommendation system based on collaborative filtering. Experimental results obtained using the database created by Kulkarni et al. show that this model is more accurate and outperforms models based on K-nearest neighbor (KNN), singular value decomposition (SVD), singular value decomposition plus plus (SVD++) and non-negative matrix factorization (NMF) in terms of the root-mean-square error (RMSE) and mean absolute error (MAE). Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

19 pages, 1307 KiB

Open AccessArticle

Application of Deep Learning in the Early Detection of Emergency Situations and Security Monitoring in Public Spaces

by William Villegas-Ch and Jaime Govea

Appl. Syst. Innov. 2023, 6(5), 90; https://doi.org/10.3390/asi6050090 - 8 Oct 2023

Cited by 2 | Viewed by 2358

Abstract

This article addresses the need for early emergency detection and safety monitoring in public spaces using deep learning techniques. The problem of discerning relevant sound events in urban environments is identified, which is essential to respond quickly to possible incidents. To solve this, [...] Read more.

This article addresses the need for early emergency detection and safety monitoring in public spaces using deep learning techniques. The problem of discerning relevant sound events in urban environments is identified, which is essential to respond quickly to possible incidents. To solve this, a method is proposed based on extracting acoustic features from captured audio signals and using a deep learning model trained with data collected both from the environment and from specialized libraries. The results show performance metrics such as precision, completeness, F1-score, and ROC-AUC curve and discuss detailed confusion matrices and false positive and negative analysis. Comparing this approach with related works highlights its effectiveness and potential in detecting sound events. The article identifies areas for future research, including incorporating real-world data and exploring more advanced neural architectures, and reaffirms the importance of deep learning in public safety. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

13 pages, 541 KiB

Open AccessArticle

Robust Sales forecasting Using Deep Learning with Static and Dynamic Covariates

by Patrícia Ramos and José Manuel Oliveira

Appl. Syst. Innov. 2023, 6(5), 85; https://doi.org/10.3390/asi6050085 - 28 Sep 2023

Cited by 3 | Viewed by 2361

Abstract

Retailers must have accurate sales forecasts to efficiently and effectively operate their businesses and remain competitive in the marketplace. Global forecasting models like RNNs can be a powerful tool for forecasting in retail settings, where multiple time series are often interrelated and influenced [...] Read more.

Retailers must have accurate sales forecasts to efficiently and effectively operate their businesses and remain competitive in the marketplace. Global forecasting models like RNNs can be a powerful tool for forecasting in retail settings, where multiple time series are often interrelated and influenced by a variety of external factors. By including covariates in a forecasting model, we can often better capture the various factors that can influence sales in a retail setting. This can help improve the accuracy of our forecasts and enable better decision making for inventory management, purchasing, and other operational decisions. In this study, we investigate how the accuracy of global forecasting models is affected by the inclusion of different potential demand covariates. To ensure the significance of the study’s findings, we used the M5 forecasting competition’s openly accessible and well-established dataset. The results obtained from DeepAR models trained on different combinations of features indicate that the inclusion of time-, event-, and ID-related features consistently enhances the forecast accuracy. The optimal performance is attained when all these covariates are employed together, leading to a 1.8% improvement in RMSSE and a 6.5% improvement in MASE compared to the baseline model without features. It is noteworthy that all DeepAR models, both with and without covariates, exhibit a significantly superior forecasting performance in comparison to the seasonal naïve benchmark. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

Review

Jump to: Research

24 pages, 3726 KiB

Open AccessReview

Machine Learning and Deep Learning Models for Demand Forecasting in Supply Chain Management: A Critical Review

by Kaoutar Douaioui, Rachid Oucheikh, Othmane Benmoussa and Charif Mabrouki

Appl. Syst. Innov. 2024, 7(5), 93; https://doi.org/10.3390/asi7050093 - 26 Sep 2024

Viewed by 3861

Abstract

This paper presents a comprehensive review of machine learning (ML) and deep learning (DL) models used for demand forecasting in supply chain management. By analyzing 119 papers from the Scopus database covering the period from 2015 to 2024, this study provides both macro- [...] Read more.

This paper presents a comprehensive review of machine learning (ML) and deep learning (DL) models used for demand forecasting in supply chain management. By analyzing 119 papers from the Scopus database covering the period from 2015 to 2024, this study provides both macro- and micro-level insights into the effectiveness of AI-based methodologies. The macro-level analysis illustrates the overall trajectory and trends in ML and DL applications, while the micro-level analysis explores the specific distinctions and advantages of these models. This review aims to serve as a valuable resource for improving demand forecasting in supply chain management using ML and DL techniques. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

49 pages, 5929 KiB

Open AccessReview

Innovating Patent Retrieval: A Comprehensive Review of Techniques, Trends, and Challenges in Prior Art Searches

by Amna Ali, Ali Tufail, Liyanage Chandratilak De Silva and Pg Emeroylariffion Abas

Appl. Syst. Innov. 2024, 7(5), 91; https://doi.org/10.3390/asi7050091 - 26 Sep 2024

Viewed by 1179

Abstract

As the patent landscape continues to grow, so does the complexity of retrieving relevant “prior art”, “background art”, or “state of the art” from an expanding pool of publicly available patent data, a critical step in establishing novelty. However, retrieving this information presents [...] Read more.

As the patent landscape continues to grow, so does the complexity of retrieving relevant “prior art”, “background art”, or “state of the art” from an expanding pool of publicly available patent data, a critical step in establishing novelty. However, retrieving this information presents significant challenges due to its volume and complexity. This systematic literature review surveys patent retrieval techniques over the past decade, focusing on ‘prior art’ and ‘novelty’ searches. Adhering to the PRISMA 2020 guidelines, our research includes 78 pertinent articles selected from a corpus of 1441, providing an in-depth overview of recent advancements, emerging trends, challenges, and future directions in the field of patent prior art retrieval. The review addresses six research questions: defining the current state of the art, evaluating the efficacy of various approaches, examining commonly used patent data collections, exploring the impact of semantic search and natural language processing (NLP) technologies, identifying frequently used components of patent documents, and discussing ongoing challenges in the domain of patent prior art search and retrieval. Our findings highlight the growing use of NLP to enhance the precision and comprehensiveness of patent searches, particularly on the Cross-Language Evaluation Forum for Intellectual Property (CLEF-IP) and the United States Patent and Trademark Office (USPTO) databases. Despite advancements, the specialized and technical nature of patent language continues to pose significant challenges in achieving high accuracy in patent retrieval. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

28 pages, 3995 KiB

Open AccessReview

Buzzing through Data: Advancing Bee Species Identification with Machine Learning

by Ashan Milinda Bandara Ratnayake, Hartini Mohd Yasin, Abdul Ghani Naim and Pg Emeroylariffion Abas

Appl. Syst. Innov. 2024, 7(4), 62; https://doi.org/10.3390/asi7040062 - 22 Jul 2024

Viewed by 1474

Abstract

Given the vast diversity of bee species and the limited availability of taxonomy experts, bee species identification has become increasingly important, especially with the rise of apiculture practice. This review systematically explores the application of machine learning (ML) techniques in bee species determination, [...] Read more.

Given the vast diversity of bee species and the limited availability of taxonomy experts, bee species identification has become increasingly important, especially with the rise of apiculture practice. This review systematically explores the application of machine learning (ML) techniques in bee species determination, shedding light on the transformative potential of ML in entomology. Conducting a keyword-based search in the Scopus and Web of Science databases with manual screening resulted in 26 relevant publications. Focusing on shallow and deep learning studies, our analysis reveals a significant inclination towards deep learning, particularly post-2020, underscoring its ability to handle complex, high-dimensional data for accurate species identification. Most studies have utilized images of stationary bees for the determination task, despite the high computational demands from image processing, with fewer studies utilizing the sound and movement of the bees. This emerging field faces challenges in terms of dataset scarcity with limited geographical coverage. Additionally, research predominantly focuses on honeybees, with stingless bees receiving less attention, despite their economic potential. This review encapsulates the state of ML applications in bee species determination. It also emphasizes the growing research interest and technological advancements, aiming to inspire future explorations that bridge the gap between computational science and biodiversity conservation. Full article

(This article belongs to the Special Issue Advancements in Deep Learning and Its Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Advancements in Deep Learning and Its Applications

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (13 papers)

Research

Review

Further Information

Guidelines

MDPI Initiatives

Follow MDPI