Traditional Methods Research Articles

The ringed residual u-net with non-natural regions feature for image splicing forgery detection and localization

Recently, with the emergence of many image editing tools (photoshop, Topaz studio, etc.), the authenticity of images has been severely challenged. However, the performance of some existing traditional feature extraction methods and detection methods based on convolutional neural network (CNN) is poor, and the information provided by the features extracted from the network is limited and single. In this paper, an end-to-end ringed residual U-Net is proposed to detect image splicing forgery by blending features of non-natural regions. Some regions with significant differences from the image background are defined as non-natural regions(such as the irregular border at the splicing of images). In this paper, a feature enhancement module for non-natural regions is constructed, which the image through the pooling of four different scales, and these features are then combined with the original image and input to the backbone network for processing, aiming to highlight regions of the image that differ significantly from the background. Therefore, after adding the feature enhancement module for non-natural regions to the end-to-end ring residual U-Net, more attention will be paid to the tampering regions in the feature extraction stage, image manipulation detection and localization will also become more accurate. Compared with some mainstream methods, this method achieves better performance on the three standard datasets(CASIA2.0, NIST2016, COLUMBIA). In addition, it has excellent robustness under JPEG compression attack and noise corruption attack.

Constructing a MOEA approach for product form Kansei design based on text mining and BPNN

Kansei Engineering (KE) is a product design method that aims to develop products to meet users’ emotional preferences. However, traditional KE faces the problem that the acquisition of Kansei factors does not represent the real consumers demands based on manual and reports, and using traditional methods to calculate relationship between Kansei factors and specific design elements, which can lead to the omission of key information. To address these problems, this study adopts text mining and backward propagation neural networks (BPNN) to propose a product form design method from a multi-objective optimization perspective. Firstly, Term Frequency-Inverse Document Frequency (TF-IDF) and WordNet are used to extract key user Kansei requirements from online review texts to obtain more accurate Kansei knowledge. Secondly, the BPNN is used to establish the non-linear relationship between product Kansei factors and specific design elements, and a preference mapping prediction model is constructed. Finally, BPNN is transformed into an iterative prediction value of non-dominated sorting genetic algorithm-II (NSGA-II), and the model is solved through multi-objective evolutionary algorithm (MOEA) to obtain the Pareto optimal solution set that satisfies the user’s multiple emotional needs, and the fuzzy Delphi method is used to obtain the best product form design scheme that meets the user’s multiple emotional images. Using the example of electric bicycle form design could show that this proposed method can effectively complete multi-objective product solutions innovation design.

Homochiral π‐Rich Covalent Organic Frameworks Enabled Chirality Imprinting in Conjugated Polymers: Confined Polymerization and Chiral Memory from Scratch

AbstractOptically active π‐conjugated polymers (OACPs) have garnered increasing research interest for their resemblance to biological helices and intriguing chirality‐related functions. Traditional methods for synthesizing involve decorating achiral conjugated polymer architectures with enantiopure side substituents through complex organic synthesis. Here, we report a new approach: the templated synthesis of unsubstituted OACPs via supramolecularly confined polymerizations of achiral monomers within nanopores of 2D or 3D chiral covalent organic frameworks (CCOFs). We show that the chiral π‐rich nanospaces facilitate the in situ enantiospecific polymerization and self‐propagation, akin to nonenzymatic polymerase chain reaction (PCR) system, resulting in chiral imprinting. The stacked polymer chains are kinetically inert enough to memorize the chiral information after liberating from CCOFs, and even after treatment at temperature up to 200 °C. The isolated OACPs demonstrate robust enantiodiscrimination, achieving up to 85 % ee in separating racemic amino acids. This underscores the potential of utilizing CCOFs as templates for supramolecularly imprinting optical activity into CPs, paving the way for synthetic evolution and advanced functional exploration of OACPs.

A machine-based corpus optimization method for extracting domain-oriented technical words: an example of COVID-19 corpus data

Although natural language processing (NLP) refers to a process involving the development of algorithms or computational models that empower machines to understand, interpret, and generate human language, machines are still unable to fully grasp the meanings behind words. Specifically, they cannot assist humans in categorizing words with general or technical purposes without predefined standards or baselines. Empirically, prior researches have relied on inefficient manual tasks to exclude these words when extracting technical words (i.e., terminology or terms used within a specific field or domain of expertise) for obtaining domain information from the target corpus. Therefore, to enhance the efficiency of extracting domain-oriented technical words in corpus analysis, this paper proposes a machine-based corpus optimization method that compiles an advanced general-purpose word list (AGWL) to serve as the exclusion baseline for the machine to extract domain-oriented technical words. To validate the proposed method, this paper utilizes 52 COVID-19 research articles as the target corpus and an empirical example. After compared to traditional methods, the proposed method offers significant contributions: (1) it can automatically eliminate the most common function words in corpus data; (2) through a machine-driven process, it removes general-purpose words with high frequency and dispersion rates –57% of word types belonging to general-purpose words, constituting 90% of the total words in the target corpus. This results in 43% of word types representing domain-oriented technical words that makes up 10% of the total words in the target corpus are able to be extracted. This allows future researchers to focus exclusively on the remaining 43% of word types in the optimized word list (OWL), enhancing the efficiency of corpus analysis for extracting domain knowledge. (3) The proposed method establishes a set of standard operation procedure (SOP) that can be duplicated and generally applied to optimize any corpus data.

HVAC energy consumption prediction based on RJITL deep neural network model

During the operation of HVAC (Heating, Ventilation, and Air-Conditioning) systems, precise energy consumption prediction plays an important role in achieving energy savings and optimizing system performance. However, the HVAC system is a complex and dynamic system characterized by a large number of variables that exhibit significant changes over time. Therefore, it is inadequate to rely on a fixed offline model to adapt to the dynamic changes in the system that consume tremendous computation time. To solve this problem, a deep neural network (DNN) model based on Just-in-Time learning with hyperparameter R (RJITL) is proposed in this paper to predict HVAC energy consumption. Firstly, relevant samples are selected using Euclidean distance weighted by Spearman coefficients. Subsequently, local models are constructed using deep neural networks supplemented with optimization techniques to enable real-time rolling energy consumption prediction. Then, the ensemble JITL model mitigates the influence of local features, and improves prediction accuracy. Finally, the local models can be adaptively updated to reduce the training time of the overall model by defining the update rule (hyperparameter R) for the JITL model. Experimental results on energy consumption prediction for the HVAC system show that the proposed DNN-RJITL method achieves an average improvement of 5.17% in accuracy and 41.72% in speed compared to traditional methods.

StegEraser: Defending cybersecurity against malicious covert communications

Traditionally, the mission of intercepting malicious traffic between the Internet and the internal network of entities like organizations and corporations, is largely fulfilled by techniques such as deep packet inspection (DPI). However, steganography, the methodology of hiding secret data in seemingly benign public mediums (e.g., images), has been leveraged by advanced persistent threat (APT) groups in recent years, and is almost impossible to be detected and intercepted by traditional techniques, posing a pervasive and realistic threat to cybersecurity. Additionally, internal networks’ vulnerability to steganography is further exacerbated by the connectivity and large attack surface of the Internet of Things (IoT), whose adoption and deployment are quickly expanding. To protect computer systems against malicious communications that apply steganographic methods potentially unknown to cybersecurity stakeholders, we propose StegEraser, an approach to removing the secret information embedded in public mediums by adversaries, that is fundamentally distinct from existing research which is primarily designed for known steganographic methods. Implemented for images, StegEraser injects an excessively huge amount of random binary data with a novel steganographic method into the images, by utilizing the information-merging capabilities of invertible neural networks (INNs), in order to “overload” adversaries’ steganographic hiding capacity of images transmitted through the firewall performing DPI. In the meantime, StegEraser preserves the perceptual quality of the images. In other words, StegEraser “defeats unknown steganography with steganography”. Extensive evaluation verifies that StegEraser significantly outperforms state-of-the-art (SOTA) methods in terms of removing secret information embedded with both traditional and neural network-based steganographic methods, while visually maintaining the image quality.

Improved metric-based meta learning with attention mechanism for few-shot cross-domain train bearing fault diagnosis

Traction motor bearings, as a crucial component of subway trains, play a pivotal role in ensuring the safety of train operations. Therefore, intelligent diagnosis of train bearings holds significant importance. However, due to the complex and dynamic nature of bearing conditions coupled with limited fault data availability, traditional diagnostic methods fail to yield satisfactory results. To address this issue, we propose an improved metrics-based meta-learning approach for accurate few-shot cross-domain fault diagnosis of train bearings. Firstly, we introduce a 1D-signal channel attention mechanism that effectively extracts latent features and enhances recognition accuracy. Secondly, by incorporating the Adabound algorithm into our model framework, we further enhance its classification performance. Finally, through several case studies, we demonstrate the effectiveness of our proposed method in comparison to other approaches within similar settings.

Unaligned Hip Radiograph Assessment Utilizing Convolutional Neural Networks for the Assessment of Developmental Dysplasia of the Hip

Abstract Developmental dysplasia of the hip (DDH) is a condition in which the acetabular socket inadequately contains the femoral head (FH). If left untreated, DDH can result in degenerative changes in the hip joint. Several imaging techniques are used for DDH assessment. In radiographs, the acetabular index (ACIN), center-edge angle, Sharp's angle (SA), and migration percentage (MP) metrics are used to assess DDH. Determining these metrics is time-consuming and repetitive. This study uses a convolutional neural network (CNN) to identify radiographic measurements and improve traditional methods of identifying DDH. The dataset consisted of 60 subject radiographs rotated along the craniocaudal and mediolateral axes 25 times, generating 1500 images. A CNN detection algorithm was used to identify key radiographic metrics for the diagnosis of DDH. The algorithm was able to detect the metrics with reasonable accuracy in comparison to the manually computed metrics. The CNN performed well on images with high contrast margins between bone and soft tissues. In comparison, the CNN was not able to identify some critical points for metric calculation on a few images that had poor definition due to low contrast between bone and soft tissues. This study shows that CNNs can efficiently measure clinical parameters to assess DDH on radiographs with high contrast margins between bone and soft tissues with purposeful rotation away from an ideal image. Results from this study could help inform and broaden the existing bank of information on using CNNs for radiographic measurement and medical condition prediction.

Embedding higher order thinking skills in Islamic history (Sirah) education in Malaysia

<span lang="EN-US">The promotion of integrating higher order thinking skills (HOTS) in the Malaysian education system is pivotal to producing analytical thinkers to keep up with the rapid globalization of this era. Thus, the implementation of HOTS in learning and facilitation (PdPc) has been emphasized. However, little is known about the integration of HOTS in Islamic history and civilization (</span><em><span lang="EN-US">Sirah</span></em><span lang="EN-US">) lessons. Furthermore, teachers continue to use traditional methods and disregard incorporating HOTS in PdPc. Therefore, the current study conducted an in-depth investigation into the components of </span><em><span lang="EN-US">Sirah</span></em><span lang="EN-US"> lessons to support the implementation of the Malaysian education development plan (PPPM) 2013-2015 by strengthening teaching techniques through the use of HOTS for a better understanding of </span><em><span lang="EN-US">Sirah</span></em><span lang="EN-US"> lessons. Using an explanatory sequential mixed method design, the findings reveal that the teaching components of </span><em><span lang="EN-US">Sirah</span></em><span lang="EN-US"> lessons with the integration of HOTS are at a moderate level in the conclusion stage of a lesson. Therefore, the current study proposes a model that integrates HOTS, i.e., the element of attitude, preparation and planning, teaching aids with relevant aspects of PAK21, knowledge of HOTS, mastery of the subject content, teaching techniques and approach, assessment, and evaluation.</span>

HK-SPSA based performance optimization method for steam generator liquid level control

Steam generator level control systems play a crucial role in ensuring safe, economical, and stable operation of nuclear power plants. However, traditional control systems exhibit limited efficiency in commissioning and are susceptible to human error. Additionally, traditional parameter tuning methods are typically experienced-based, cumbersome, and time-consuming, making it challenging to obtain optimal parameters. To address these issues, we propose a hybrid knowledge-guided improved simultaneous perturbation stochastic approximation algorithm (HK-SPSA) to optimize the control parameters of steam generator level control systems. Firstly, the algorithm utilizes iteration point adjacency information to approximate the current optimization process status and guide the adaptive adjustment of the iteration step size. Secondly, it utilizes composite gradient information generated by the estimated historical and current gradients to guide the optimization direction and the tuning of iteration step size. Simulation experiments have shown that HK-SPSA improves optimization performance and reduces the costs of optimization and adjustment compared to traditional methods.

Critical Review on Water Quality Analysis using IoT and Machine Learning Models

Water quality and its management are the most precise concerns confronting humanity globally. This article evaluates the various sensors used for water quality monitoring and focuses on the water quality index considering the multiple physical, chemical, and biological parameters. A Review of Internet of Things (IoT) research for water quality monitoring and analysis, sensors used for water quality can help remote monitoring of the water quality parameters using various IoT-based sensors that convey the assembled estimations utilizing Low-Power Wide Area Network innovations. Overall, the IoT system was 95% accurate in measuring pH, Turbidity, TDS, and Temperature, while the traditional method was only 85% accurate. Also, this study reviewed the different A.I. techniques used to assess water quality, including conventional machine learning techniques, Support Vector Machines, Deep Neural Networks, and K-nearest neighbors. Compared to traditional methods, machine learning and deep learning can significantly increase the accuracy of measurements of groundwater quality. However, various variables, such as the caliber of the training data, the water quality metrics' complexity, and the monitoring frequency, will affect the accuracy. The geographical information system (GIS.) is used for spatial data analysis and managing water resources. The quality of its data is also reviewed in the paper. Based on these analyses, the study has forecasted the future sensors, Geospatial Technology, and machine learning techniques for water quality monitoring and analysis.

Nowhere to Hide: Online Rumor Detection Based on Retweeting Graph Neural Networks.

Online rumor detection is crucial for a healthier online environment. Traditional methods mainly rely on content understanding. However, these contents can be easily adjusted to avoid such supervision and are insufficient to improve the detection result. Compared with the content, information propagation patterns are more informative to support further performance promotion. Unfortunately, learning the propagation patterns is difficult, since the retweeting tree is more topologically complicated than linear sequences or binary trees. In light of this, we propose a novel rumor detection framework based on structure-aware retweeting graph neural networks. To capture the propagation patterns, we first design a novel conversion method to transform the complex retweeting tree as more tractable binary tree without losing the reconstruction information. Then, we serialize the retweeting tree as a corpus of meta-tree paths, where each meta-tree can preserve a basic substructure. A deep neural network is then designed to integrate all meta-trees and to generate the global structural embeddings. Furthermore, we propose to integrate content, users, and propagation patterns to enhance more reliable performance. To this end, we propose a novel self-attention-based retweeting neural network to learn individual features from both content and users. We then fuse the node-level features with our global structural embeddings via a mutual attention unit. In this way, we can generate more comprehensive representations for rumor detection. Extensive evaluations on two real-world datasets show remarkable superiorities of our model compared with existing methods.

Stochastic Economic Lot Scheduling via Self-Attention Based Deep Reinforcement Learning

The Stochastic Economic Lot Scheduling Problem (SELSP) is a difficult dynamic optimization problem with wide industrial applications. Traditional methods such as hyper-heuristics are manually designed based on substantial expert knowledge, which may limit their optimization performance. Recently, Deep Reinforcement Learning (DRL) is shown to be promising in automatically learning scheduling policies for SELSP. However, its performance is still quite far from that of hyper-heuristics, due to the lack of suitable deep models. In this paper, we propose a novel DRL method to learn dynamic scheduling policies for SELSP in an end-to-end fashion. Based on self-attention, our method can effectively extract useful features from raw state information, and is flexible in handling different numbers of products, which is not viable for previous methods. Experiments on a complex biopharmaceutical manufacturing process show that our method outperforms a recent DRL method and state-of-the-art hyper-heuristics. Moreover, the trained policy performs better in environments different from training with demand forecast errors and varying number of products, showing its strong robustness and generalization ability. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —The Stochastic Economic Lot Scheduling Problem (SELSP) is an important problem for manufacturing enterprises, which is to optimally balance the production and inventory so as to minimize the total cost. However, SELSP is very challenging to solve due to the involvement of uncertain factors such as customer demands and machine failures. Traditional methods for solving SELSP, such as heuristic policies and hyper-heuristics, heavily rely on human experiences to design and hence the performance could be limited. This paper proposes a Deep Reinforcement Learning (DRL) based method to automatically learn scheduling policy for solving SELSP, which could alleviate the above limitation through a self-attention based feature extraction mechanism and reward based training. Experimental results on a realistic manufacturing process show that our method can deliver higher revenue than conventional manual policy and an existing DRL based method.

Emphasis on Nanostructured Lipid Carriers in the Ocular Delivery of Antibiotics.

Drug distribution to the eye is still tricky because of the eye's intricate structure. Systemic delivery, as opposed to more traditional methods like eye drops and ointments, is more effective but higher doses can be harmful. The use of solid lipid nanoparticles (SLNPs) as a method of drug delivery has been the subject of research since the 1990s. Since SLNPs are derived from naturally occurring lipids, they pose no health risks to the user. To raise the eye's absorption of hydrophilic and lipophilic drugs, SLNs can promote corneal absorption and improve the ocular bioavailability of SLNPs. To address problems related to ocular drug delivery, many forms of nano formulation were developed. Some of the methods developed are, emulsification and ultra-sonication, high-speed stirring and ultra-sonication, thin layer hydration, adapted melt-emulsification, and ultrasonication techniques, hot o/w micro-emulsion techniques, etc. Results: Nanostructured lipid carriers are described in this review in terms of their ocular penetration mechanism, structural characteristic, manufacturing process, characterization, and advantages over other nanocarriers. Recent developments in ocular formulations with nanostructured bases, such as surfacemodified attempts have been made to increase ocular bioavailability in both the anterior and posterior chambers by incorporating cationic chemicals into a wide variety of polymeric systems.

Why Motivational Interviewing Is a Person-Centered Approach

Abstract: Background: The original developers of motivational interviewing (MI) describe the foundation of their therapeutic method in the person-centered tradition of Carl Rogers. Although MI is sometimes interpreted as a cognitive-behavior therapy (CBT), it did not derive from learning theory and has from the beginning been understood and expounded as a person-centered approach (PCA). Method: Narrative and conceptual review. Results: As with Rogers’s original method, the development of MI has been guided by empirical science based on the close observation of clinical practice. It is a behavioral approach in the broad sense of behavioral science. The underlying attitude or spirit with which MI is practiced is unambiguously humanistic in emphasizing partnership, acceptance, compassion, and empowerment. MI focuses on specific observable elements of therapeutic relationship that predict better client outcomes. Various behavioral scientists have distinguished MI from CBT both conceptually and with distinct clinical procedures. The psychological content of MI overlaps substantially with PCA practices such as accurate empathy and also with therapist skills often described as common or nonspecific factors in psychotherapy. Its component processes are engaging, focusing, evoking, and planning. The evoking process of MI distinguishes it from both CBT and nondirective PCA, involving MI-specific technical skills focused on specific client language known as change talk and sustain talk. Conclusions: Like PCA more generally, MI has found broad applications in helping professions well beyond counseling and psychotherapy such as health care, education, pastoral care, conflict resolution, and leadership. Possible historic reasons for confusing MI with CBT are explored.

Unperforated film-covered planting contributes to improved film recovery rates and foxtail millet grain yields in sandy soils

Plastic film mulching improves the soil water content and temperature and, thus, contributes to enhanced grain yield in arid and semi-arid areas, such as Northwest China. However, in traditional film-covering methods, plastic films are difficult to recycle, causing environmental issues. Therefore, developing cultivation methods with enhanced film recovery rates is important for sustainable agricultural development. Here, we evaluated how the planting method affected the soil temperature, water content, film recovery rate, and grain yield by applying the following three planting patterns to the traditional crop, foxtail millet, in 2021–2022: no plastic film mulching (NP), plastic film mulching (MP), and planting on the film edge (PFE); in the latter, holes were not punched in the film and crops were planted approximately 5 cm from the film edge. The results for 2021 showed that the biomass ranged from 27.69 g to 33.76 g per plant, and the biomass was increased by 21.9% and 3.2% in PFE compared with NP and MP, respectively. However, in 2022, the biomass increased in PFE by 52.5% and 17.7% compared with NP and MP, respectively. In 2021, the highest grain yield was observed in PFE (2480 kg/hm2), followed by MP (2470 kg/hm2) and NP (2343 kg/hm2); in 2022, a significantly higher grain yield was observed in PFE than that in MP and NP. Although the soil temperature increased in PFE, it was not as high as in MP. The relatively higher soil temperature increased the root growth process, stimulated the production of root exudates (acetic acid and citric acid), and increased the soil enzyme (sucrase and urease) activity. Increases in organic acids and soil enzymes assisted in the decomposition of soil organic matter and induced relatively higher levels of available nutrients, such as nitrogen and phosphorus, and the higher root reuse rates contributed to higher grain yields. In addition, fewer holes were punched in the film of the PFE model, which improved film recovery, and the recovery rate of PFE was higher than that of MP. These results indicate that the proposed unperforated film-covered planting pattern is suitable for millet production in sandy soil.

Plasma-fixed Nitrogen Improves Lettuce Field Holding Potential

This study shows that plasma-fixed nitrogen applied as an inorganic biostimulant can improve marketable lettuce (Lactuca sativa var. longifolia) yield following delayed harvest. Using just one-tenth of the conventional nitrogen, plasma-fixed nitrogen—which is generated by a dielectric barrier discharge over water—was field-tested against traditional fertilization methods. Although no statistically significant differences were observed in total weight of heads among treatments, plasma-fixed nitrogen–treated plants had significantly increased marketable yields of 250% compared to those grown conventionally, despite reducing applied nitrogen fertilizer.

Effectiveness of Using a Self-Directed Learning Program to Teach Physical Examination and Health Assessment Skills: A Quasi-Experimental Study.

Conducting health assessments and physical examinations are essential skills for independent practice. Exploring how to teach these skills effectively is essential. This study was designed to validate the effectiveness of a self-directed learning program in improving nursing student outcomes in terms of their learning health assessment and physical examination skills. A quasi-experimental design was used, and second-year nursing students enrolled in a health assessment and physical examination course at a university of science and technology in central Taiwan were recruited using purposive sampling. Both the control and experimental groups were taught using a traditional teaching method. The experimental group additionally participated in a self-directed learning program. Structured questionnaires, including the Chinese version of the Self-Directed Learning Readiness Scale, Scales of Motivation and Learning Strategies, Interpersonal Skills Assessment Tool, the Course Satisfaction Evaluation Scale, and the Perception of Health Assessment and Physical Examination Competence Scale, were used to collect data at three time points. After controlling for the effects of pretest scores, after the intervention, the experimental group achieved significantly higher scores than the control group for self-directed learning (creative learning and love of learning), learning motivation (goal orientation, work value, expected success, and test anxiety), and cognitive strategies (total score, elaboration strategy, recitation strategy, and monitoring strategy). In addition, the health assessment exercise improved interpersonal and communication skills, and learning satisfaction was significantly higher in the experimental group than the control group. A physical examination and health assessment course designed using the self-directed learning framework can effectively improve student competence in related skills to enhance their ability to assess patient health problems in clinical settings. This study presents an alternative approach to teaching health assessment and physical examination courses and validates the positive effect of this approach on student learning outcomes.

A Physics-Guided MFL Deformed Defect Recovery Method

Magnetic flux leakage (MFL) testing and analysis is an effective non-destructive testing (NDT) method of pipeline health. The health status of pipelines is evaluated by analyzing MFL signals, especially the defect analysis. However, complex operating conditions and equipment deviations may cause deformation on the defects. Availability and integrity of the MFL defects are the keys to accurate signal processing. Due to the limited interpretability of abstract networks and insufficient feature extraction, it is hard to accurately recover the deformed defects. Considering the problems above, an MFL deformed defect recovery method with the guidance of physical features is proposed. The proposed method integrates a deep feature extraction model into a sparse autoencoder, whose parameters are guided by the combination of MSE loss and physics loss. First, the deep feature extraction model fully extracts the abstract features, which contains more information on deformed defects. Second, the sparse penalty factor in the sparse autoencoder reduces the feature redundancy of deformed defects. Third, the physical features are extracted to guide the MFL deformed defect recovery relying on the MFL mechanism for the first time, which enhances the robustness and accuracy of the network. Finally, several comparison experiments have been conducted on measured and simulated MFL defects. The results show that the proposed method is effective in deformed defect recovery. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —The motivation of this paper is a practical problem of MFL deformed defect recovery. The deformation on MFL defects affects defect inversion and evaluation seriously, which causes great potential harm to pipeline safety transmission. In this paper, compared with traditional neural network-based methods, physical features based on MFL mechanism are extracted to guide the parameter update. In addition, a pre-trained model replaces regular convolutional layers to extract features sufficiently, while a sparse factor is applied to reduce redundant features. The proposed method not only utilizes deep features from deep networks, but also increases the guidance of physical features, so that the algorithm is more robust and accurate. The experiment results show that our proposed method is more effective than the comparative methods on MFL simulated and measured defects. In a word, the proposed method has strong theoretical research and practical value.

Training Demand Prediction Models by Decision Error for Two-Stage Lot-Sizing Problems

Demand prediction to support appropriate production decisions is being actively studied. Many prediction models are designed to minimize the prediction error, which is measured by determining the difference between the predicted and ground-truth demand. However, these models ignore the effect of the prediction error on downstream production decisions. This prompted our study, which focuses on demand prediction models for two-stage uncapacitated lot-sizing problems. In this paper, we present a prediction model that minimizes the decision error, which is measured by the optimization objective of lot-sizing problems. Our model mitigates the impact of prediction errors by leveraging the structure of the lot-sizing problems. To enhance the ability to accommodate imperfect data, such as data based on inaccurate information, we subsequently extend the prediction models to distributionally robust versions. We consider the worst-case formulation in the feature space to enhance the robustness of the model to data imperfection. Numerical experiments demonstrate that the proposed prediction models are significantly superior to the traditional prediction methods when the model being trained is misspecified. In addition, the robust extension enables the models to train well on imperfect datasets while requiring less training data. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This study is motivated by the two-stage lot-sizing problem in manufacturing systems which involves the determination of the time at which (the setup decisions) to produce before the demand is revealed, and the production quantity decisions are adjusted during production processing. Predicting the demand properly while making the set-up decisions helps to decrease the production cost. This paper proposes a novel method that aims to minimize the total production cost to train demand prediction models. Practitioners can deploy the proposed method to train popular prediction models including linear prediction, decision tree, and neural network models. It is shown that the production cost associated with the decisions based on our models is lower than that associated with the traditional prediction models. Furthermore, our method enhances the ability of the model to handle model misspecification and imperfect data. Practitioners can apply our method to overcome the problems caused by imperfect data such as those containing contaminated and inaccurate measurements.