Cosine Similarity Research Articles

Enhanced content-based fashion recommendation system through deep ensemble classifier with transfer learning

With the rise of online shopping due to the COVID-19 pandemic, Recommender Systems have become increasingly important in providing personalized product recommendations. Recommender Systems face the challenge of efficiently extracting relevant items from vast data. Numerous methods using deep learning approaches have been developed to classify fashion images. However, those models are based on a single model that may or may not be reliable. We proposed a deep ensemble classifier that takes the probabilities obtained from five pre-trained models such as MobileNet, DenseNet, Xception, and the two varieties of VGG. The probabilities obtained from the five pre-trained models are then passed as inputs to a deep ensemble classifier for the prediction of the given item. Several similarity measures have been studied in this work and the cosine similarity metric is used to recommend the products for a classified product given by a deep ensemble classifier. The proposed method is trained and validated using benchmark datasets such as Fashion product images dataset and Shoe dataset, demonstrating superior accuracy compared to existing models. The results highlight the potential of leveraging transfer learning and deep ensemble techniques to enhance fashion recommendation systems. The proposed model achieves 96% accuracy compared to the existing models.

Keystrokes: A practical exploration of semantic drift in timed word association tasks.

This study investigates the phenomena of semantic drift through the lenses of language and situated simulation (LASS) and the word frequency effect (WFE) within a timed word association task. Our primary objectives were to determine whether semantic drift can be identified over the short time (25 seconds) of a free word association task (a predicted corollary of LASS), and whether more frequent terms are generated earlier in the process (as expected due to the WFE). Respondents were provided with five cue words (tree, dog, quality, plastic and love), and asked to write as many associations as they could. We hypothesized that terms generated later in the task (fourth time quartile, the last 19-25 seconds) would be semantically more distant (cosine similarity) from the cue word than those generated earlier (first quartile, the first 1-7 seconds), indicating semantic drift. Additionally, we explored the WFE by hypothesizing that earlier generated words would be more frequent and less diverse. Utilizing a dataset matched with GloVe 300B word embeddings, BERT and WordNet synsets, we analysed semantic distances among 1569 unique term pairs for all cue words across time. Our results supported the presence of semantic drift, with significant evidence of within-participant, semantic drift from the first to fourth time (LASS) and frequency (WFE) quartiles. In terms of the WFE, we observed a notable decrease in the diversity of terms generated earlier in the task, while more unique terms (greater diversity and relative uniqueness) were generated in the 4th time quartile, aligning with our hypothesis that more frequently used words dominate early stages of a word association task. We also found that the size of effects varied substantially across cues, suggesting that some cues might invoke stronger and more idiosyncratic situated simulations. Theoretically, our study contributes to the understanding of LASS and the WFE. It suggests that semantic drift might serve as a scalable indicator of the invocation of language versus simulation systems in LASS and might also be used to explore cognition within word association tasks more generally. The findings also add a temporal and relational dimension to the WFE. Practically, our research highlights the utility of word association tasks in understanding semantic drift and the diffusion of word usage over a sub-minute task, arguably the shortest practically feasible timeframe, offering a scalable method to explore group and individual changes in semantic relationships, whether via the targeted diffusion of influence in a marketing campaign, or seeking to understand differences in cognition more generally. Possible practical uses and opportunities for future research are discussed.

Clinical advantage staging and underlying mechanisms of Wangbi Tablets against knee osteoarthritis based on "disease-formula" interaction network

The clinical advantage staging and underlying mechanisms of Wangbi Tablets against knee osteoarthritis(KOA) were studied based on the "disease-formula" interaction network. Firstly, the clinical symptoms and related genes corresponding to Wangbi Tablets and KOA in the acute, remission, and recovery phases were collected from clinical guidelines/consensus and SoFDA database, and the putative targets of Wangbi Tablets were obtained from ETCM 2.0. Then, Jaccard similarity and cosine similarity were employed to assess the similarities of clinical symptoms, genes, and enriched pathways between Wangbi Tablets and KOA in different phases. The "disease-formula" interaction network of the drug targets and disease genes was constructed, and the key targets were screened by topological feature calculation. KEGG and Reactome database were used for the functional enrichment of the key targets, on the basis of which the functional characteristics of Wangbi Tablets against KOA in the acute, remission, and recovery phases were predicted. Finally, the SW1353 cells exposed to lipopolysaccharide were used to decipher the mechanism of Wangbi Tablets against KOA. The results showed that 92/3 921, 138/3 708, 139/3 800, and 196/3 946 clinical symptoms and the related genes corresponded to KOA in the acute, remission, and recovery phases and Wangbi Tablets were collected from SoFDA, and 260 putative targets of Wangbi Tablets were obtained from ETCM 2.0. Wangbi Tablets had highest similarity of clinical symptoms, genes, and enriched pathways with KOA in the remission phase and the secondary highest similarity with KOA in the recovery phase. The key targets of Wangbi Tablets mainly participated in the regulation of immunity-inflammation imbalance and exerted pain-relieving and bone-protecting effects to alleviate symptoms such as knee joint pain, joint swelling, soreness, fatigue, and dysfunction. Intriguingly, the key targets of Wangbi Tablets possessed antioxidant effects during KOA in the acute and remission phases, while they maintained material and energy metabolism homeostasis and protected vessels during KOA in the recovery phase. The cell experiment indicated that Wangbi Tablets down-regulated the expression of interleukin(IL)-6, IL-1β, tumor necrosis factor-α(TNF-α), and Bcl-2-associated X protein(Bax)/B-cell lymphoma 2(Bcl-2) via regulating the phosphatidylinositol 3-kinase(PI3K)-protein kinase B(Akt) signaling pathway. The findings lay a theoretical foundation for further clarifying the clinical advantage stage and precise clinical application of Wangbi Tablets in treating KOA.

A multiscale 3D network for lung nodule detection using flexible nodule modeling.

Lung cancer is the most common type of cancer. Detection of lung cancer at an early stage can reduce mortality rates. Pulmonary nodules may represent early cancer and can be identified through computed tomography (CT) scans. Malignant risk can be estimated based on attributes like size, shape, location, and density. Deep learning algorithms have achieved remarkable advancements in this domain compared to traditional machine learning methods. Nevertheless, many existing anchor-based deep learning algorithms exhibit sensitivity to predefined anchor-box configurations, necessitating manual adjustments to obtain optimal outcomes. Conversely, current anchor-free deep learning-based nodule detection methods normally adopt fixed-size nodule models like cubes or spheres. To address these technical challenges, we propose a multiscale 3D anchor-free deep learning network (M3N) for pulmonary nodule detection, leveraging adjustable nodule modeling (ANM). Within this framework, ANM empowers the representation of target objects in an anisotropic manner, with a novel point selection strategy (PSS) devised to accelerate the learning process of anisotropic representation. We further incorporate a composite loss function that combines the conventional L2 loss and cosine similarity loss, facilitating M3N to learn nodules' intensity distribution in three dimensions. Experiment results show that the M3N achieves 90.6% competitive performance metrics (CPM) with seven predefined false positives per scan on the LUNA 16 dataset. This performance appears to exceed that of other state-of-the-art deep learning-based networks reported in their respective publications. Individual test results also demonstrate that M3N excels in providing more accurate, adaptive bounding boxes surrounding the contours of target nodules. The newly developed nodule detection system reduces reliance on prior knowledge, such as the general size of objects in the dataset, thus it should enhance overall robustness and versatility. Distinct from traditional nodule modeling techniques, the ANM approach aligns more closely with the morphological characteristics of nodules. Time consumption and detection results demonstrate promising efficiency and accuracy which should be validated in clinical settings.

Evaluation of anticancer therapy-related dermatologic adverse events: Insights from Food and Drug Administration's Adverse Event Reporting System dataset

New anticancer therapies have improved patient outcomes but associated dermatologic adverse events (AEs) may cause morbidity and treatment discontinuation. A comprehensive estimation of associations between cancer drugs and skin AEs is lacking. This study utilized the Food and Drug Administartion (FDA)'s Adverse Event Reporting System database (January 2013-September 2022), with 3,399,830 reports involving 3084 drugs and 16,348 AEs. A nearest neighbor matching model was employed to select 10 controls for each case report, utilizing the cosine similarity of demographic and AE severity factors to minimize false positives/negatives. There were 10,698 unique anticancer drugs (n=212) to skin AE (n=873) pairs, of which 676 had significant reporting odds ratios (ROR)>1, comprising 113 drugs and 144 AEs. The minimum ROR was 1.25, and 50% of associations displayed a ROR >10. The most common were rash (51 agents) and dry skin (28 drugs). Methotrexate induced the most distinct AEs (34), then mechlorethamine (33), and vemurafenib (24). Targeted therapies accounted for 49% of pairs, cytotoxic chemotherapies for 35.9%, and immunotherapies for 11%. A total of 113 anticancer drugs were identified as significantly associated with skin AEs, most frequently rash and dry skin. Data are likely under-reported but enable quick postmarketing identification of skin toxicity signals.

Scientific paper recommender system using deep learning and link prediction in citation network

Today, the number of published scientific articles is increasing day by day, and this has made the process of searching for articles more difficult. The need to provide specific recommender systems (RSs) for suggesting scientific articles is strongly felt in this situation. Because searching for articles based only on matching the titles or content of other articles is not an efficient process. In this research, the combination of two content analysis and citation network is used to design an RS for scientific articles (RECSA). In RECSA, natural language processing and deep learning techniques are used to process the titles and extract the content attributes of the articles. For this purpose, first, the titles of the articles are pre-processed, and by using the Term Frequency Inverse Document Frequency (TF-IDF) criterion, the importance of each word in the title is estimated. Then the dimensions of the obtained attributes are reduced by using a convolutional neural network (CNN). Then, by using the cosine similarity criterion, the content similarity matrix of the articles is calculated based on the attribute vectors. Also, the link prediction approach is used to analyze the connections of scientific articles' citation network. Finally, in the third step of RECSA, the two similarity matrices calculated in the previous steps are combined using an influence coefficient parameter to obtain the final similarity matrix, and the recommendation operation is based on the highest similarity value. The efficiency of RECSA has been evaluated from different aspects and the results have been compared with previous works. According to the results, utilizing the combination of TF-IDF and CNN for analyzing content-based features, leads to at least 0.32 % improvement in terms of precision compared to previous works. Also, by integrating citation and content-based data, the precision of first suggestion in RECSA would be 99.01 % which indicates the minimum improvement of 0.9 % compared to compared methods. The results show that by using RECSA, the recommendation can be done with higher accuracy and efficiency.

Enhancing speaker identification in criminal investigations through clusterization and rank-based scoring

This paper introduces an approach that supports speaker identification in criminal investigations, specifically addressing challenges associated with large volumes of audio recordings featuring unknown speaker identities. Our approach clusters related recordings – potentially from the same person – based on representative voice embeddings extracted using the ECAPA-TDNN speaker recognition model. Grouping audio recordings from the same person enhances variability and richness in voice patterns, thereby improving confidence in automatic speaker recognition. We propose a combination of cosine similarity and a rank-based adjustment function to determine matches of audio clusters with individuals in an enrollment database. Our approach was validated through experiments on a Common Voice-based synthesized dataset and a real-life application involving cell phones seized in prisons, which contained thousands of conversational audio recordings. Results demonstrated satisfactory performance and stability, consistently reducing the pool of candidate speakers for subsequent analysis by a human investigator.

Triple decomposition and sparse representation for noisy pressure-sensitive paint data

Triple decomposition is a powerful analytical method for a deep understanding of the flow structure by extracting the mean value, organized coherent motion, and stochastic part from a fluctuating quantity. Here, we perform the triple decomposition of the spatial two-dimensional data, especially pressure-sensitive paint (PSP) data, since the PSP method is widely used to measure the pressure distribution on a surface in wind tunnel testing. However, the PSP data measuring near atmospheric pressure contain significant noise, and this makes it difficult to conduct the decomposition. To construct phase-averaged data representing an organized coherent motion, we propose a relatively simple method based on a multi-dimensional scaling plot of the cosine similarity between each PSP datum. Then, the stochastic part is extracted by selecting phase-averaged data with an appropriate phase angle based on the similarity between the measurement and phase-averaged data, and the PSP data are successfully decomposed. Moreover, we consider sparse optimal sensor positions, in which the data are effectively represented, based on the stochastic part as a data-driven approach. The optimal sensor positions are determined as a combinatorial optimization problem and estimated using Fujitsu computing as a service digital annealer. We reconstruct the pressure distribution from the pressure data at the optimal sensor positions using the mean value, organized coherent motion, and stochastic part obtained from the triple decomposition. The root mean square error between the pressure measured by a pressure transducer and the reconstructed pressure obtained by the proposed method is small, even when the number of modes and sensor points is small. The application of PSP measurement is expected to expand further, and the framework for calculating triple decomposition and sparse representation based on the decomposition will be useful for detailed flow analysis.

A data-driven intravoxel mean diffusivities distribution approach for molecular classifications and MIB-1 prediction of gliomas.

Measuring non-parametric intravoxel mean diffusivity distributions (MDDs) using magnetic resonance imaging (MRI) is a sensitive method for detecting intracellular diffusivity changes during physiological alterations. Histological and molecular glioma classifications are essential for prognosis and treatment, with distinct water diffusion dynamics among subtypes. We developed a data-driven approach using a fully connected network (FCN) to enhance the speed and stability of calculating MDDs across varying SNRs, enable tumor microstructural mapping, and test its reliability in identifying MIB-1 labeling index (LI) levels and molecular status of gliomas. An FCN was trained to learn the mapping between the simulated diffusion decay curves and the ground truth MDDs. We performed 5 000 000 simulation curves with various diffusivity components and random SNR . Eighty percent of simulation curves were used for the FCN training, 10% for validation, and the others were external tests for the FCN performance evaluation. In vivo data were collected to evaluate its clinical reliability. One hundred one patients (44 years 14, 67 men) with gliomas and six healthy controls underwent a 3.0 T MRI examination with a spin echo-echo planar imaging (SE-EPI) diffusion-weighted imaging (DWI) sequence. The trained FCN was employed to calculate MDDs of each brain voxel by voxel. We used the Fuzzy C-means algorithm to cluster the MDDs of tumor voxels, facilitating the characterization of distinct glioma tissues. Quantitative assessments were conducted through sectional integrals of the MDDs, demarcated by six bands to derive signal fractions ( ) and diffusivities of the maximum peaks ( ). Cosine similarity scores (CSS) were used for MDD similarity. ANOVA and Mann-Whitney U test were used for difference analysis. Logistic regression and area under the receiver operator characteristic curve (AUC) were used for classification evaluation. The simulation results showed that the FCN-based MDD approach (FCN-MDD) achieved higher CSS than non-negative least squares-based MDD (NNLS-MDD). For in vivo data, the spectra of ET and NET obtained by FCN-MDD are more distinguishable than NNLS-MDD. Fraction maps delineate the characteristics of different tumor tissues (enhancing and non-enhancing tumor, edema, and necrosis). showed a positive and negative correlation with MIB-1 respectively ( , all ). The AUC of for predicting MIB-1 LI levels was 0.900 (95% CI, 0.826-0.974), versus 0.781 (0.677-0.886) of ADC. The highest AUC of isocitrate dehydrogenase (IDH) mutation status, assessed by a logistic regression model ( ) was 0.873 (95% CI, 0.802-0.944). The proposed FCN-MDD method was more robust to variations in SNR and less reliant on empirically set regularization values than the NNLS-MDD method. FCN-MDD also enabled qualitative and quantitative evaluation of the composition of gliomas.

Clustering statistical method of high dimensional sparse data based on fuzzy data

Abstract Developing effective clustering and statistical methods for high-dimensional sparse data presents unique challenges compared to traditional low-dimensional data. To address this, a novel approach is proposed, leveraging fuzzy data principles to enhance the clustering and statistical performance of high-dimensional sparse datasets. The method builds upon the fuzzy C-means clustering algorithm, introducing key modifications for better suitability to high-dimensional sparse data. One crucial enhancement involves tackling the local optimization problem by optimizing the initial clustering center, significantly reducing clustering statistical time. Replacing the original Euclidean distance with cosine distance improves the clustering and statistical performance of high-dimensional sparse data. Experimental results have shown that this method has superior clustering statistical performance when the data dimensions are different. When the data dimension is low, and the blocking ratio is 10%, the clustering statistical effect is optimal. When the data dimension is high, and the blocking ratio is 40%, the clustering statistical effect is optimal. This method has higher hit rates and clustering statistical efficiency at different sparsity levels.

El derecho informático y la influencia en los sistemas de información: un análisis bibliográfico bajo la perspectiva jurídica y tecnológica en Ecuador

This research focused on computer law and information systems in Ecuador. The aim was to analyze the legal and technological perspectives of computer laws and their influence on information systems in the Ecuadorian context. A PRISMA review protocol was established to obtain the bibliography, and advanced techniques such as Word2Vec and cosine similarity analyses were applied to two different linguistic corpora (Spanish and English). Hierarchical dendrograms provided a visual representation, highlighting groups of articles and the complexity of the relationship between computer law and information systems. The comparison between the corpora showed similarities and differences, such as the more marked connection between two of the articles in the Spanish corpus. The study revealed a significant connection between computer law and information systems, underscoring common themes such as cybercrime, data protection, and intellectual property. Links supported by Ecuadorian legislation were identified in areas such as software protection and e-commerce. However, the need to effectively implement the laws and broader international cooperation to address challenges such as cybercrime is emphasized. The study demonstrated disparities and connections between Ecuador's computer law and information systems. The need to adopt an interdisciplinary approach for future research is underlined.

Banks' Portfolio Similarity During the COVID-19 Pandemic

Purpose: This study tries to identify the impact of the COVID-19 pandemic on the similarity of banks' loan portfolios. Design/methodology/approach: Using the entire firm-level bank loan data from 2009 to 2020 in the Korean market, this study analyzes the cosine similarity in terms of credit rating, industry, and region between banks. Findings: The study discovered that interbank portfolio similarity has escalated over time, with the similarity notably intensifying during the COVID-19 pandemic. Additionally, it was found that larger banks exhibit greater similarities, a trend that has become increasingly pronounced during the pandemic period. Research limitations/implications: This study provides empirical evidence of the synchronization in lending behaviors among major South Korean banks during the COVID-19 pandemic. It posits that while the synchronicity of banks' lending patterns during typical financial crises is generally accepted, further research in financial markets of other countries is necessary to substantiate this claim universally. Originality/value: Therefore, the role of financial authorities to alleviate deepening interbank portfolio synchronization triggered by the COVID-19 pandemic and the resulting escalation of financial market instability is emphasized.

PANet: Pluralistic Attention Network for Few-Shot Image Classification

Traditional deep learning methods require a large amount of labeled data for model training, which is laborious and costly in real word. Few-shot learning (FSL) aims to recognize novel classes with only a small number of labeled samples to address these challenges. We focus on metric-based few-shot learning with improvements in both feature extraction and metric method. In our work, we propose the Pluralistic Attention Network (PANet), a novel attention-oriented framework, involving both a local encoded intra-attention(LEIA) module and a global encoded reciprocal attention(GERA) module. The LEIA is designed to capture comprehensive local feature dependencies within every single sample. The GERA concentrates on the correlation between two samples and learns the discriminability of representations obtained from the LEIA. The two modules are complementary to each other and ensure the feature information within and between images can be fully utilized. Furthermore, we also design a dual-centralization (DC) cosine similarity to eliminate the disparity of data distribution in different dimensions and enhance the metric accuracy between support and query samples. Our method is thoroughly evaluated with extensive experiments, and the results demonstrate that with the contribution of each component, our model can achieve high-performance on four widely used few-shot classification benchmarks of miniImageNet, tieredImageNet, CUB-200-2011 and CIFAR-FS.

Approach for Tattoo Detection and Identification Based on YOLOv5 and Similarity Distance

The large number of images in the different areas and the possibilities of technologies lead to various solutions in automatization using image data. In this paper, tattoo detection and identification were analyzed. The combination of YOLOv5 object detection methods and similarity measures was investigated. During the experimental research, various parameters have been investigated to determine the best combination of parameters for tattoo detection. In this case, the influence of data augmentation parameters, the size of the YOLOv5 models (n, s, m, l, x), and the three main hyperparameters of YOLOv5 were analyzed. Also, the efficiency of the most popular similarity distances cosine and Euclidean was analyzed in the tattoo identification process with the purpose of matching the detected tattoo with the person’s tattoo in the database. Experiments have been performed using the deMSI dataset, where images were manually labeled to be suitable for use by the YOLOv5 algorithm. To validate the results obtained, the newly collected tattoo dataset was used. The results have shown that the highest average accuracy of all tattoo detection experiments has been obtained using the YOLOv5l model, where mAP@0.5:0.95 is equal to 0.60, and mAP@0.5 is equal to 0.79. The accuracy for tattoo identification reaches 0.98, and the F-score is up to 0.52 when the highest cosine similarity tattoo is associated. Meanwhile, to ensure that no suspects will be missed, the cosine similarity threshold value of 0.15 should be applied. Then, photos with higher similarity scores should be analyzed only. This would lead to a 1.0 recall and would reduce the manual tattoo comparison by 20%.

Intelligent tumor tissue classification for Hybrid Health Care Units.

In the evolving healthcare landscape, we aim to integrate hyperspectral imaging into Hybrid Health Care Units to advance the diagnosis of medical diseases through the effective fusion of cutting-edge technology. The scarcity of medical hyperspectral data limits the use of hyperspectral imaging in disease classification. Our study innovatively integrates hyperspectral imaging to characterize tumor tissues across diverse body locations, employing the Sharpened Cosine Similarity framework for tumor classification and subsequent healthcare recommendation. The efficiency of the proposed model is evaluated using Cohen's kappa, overall accuracy, and f1-score metrics. The proposed model demonstrates remarkable efficiency, with kappa of 91.76%, an overall accuracy of 95.60%, and an f1-score of 96%. These metrics indicate superior performance of our proposed model over existing state-of-the-art methods, even in limited training data. This study marks a milestone in hybrid healthcare informatics, improving personalized care and advancing disease classification and recommendations.

CDTracker: Coarse-to-Fine Feature Matching and Point Densification for 3D Single-Object Tracking

Three-dimensional (3D) single-object tracking (3D SOT) is a fundamental yet not well-solved problem in 3D vision, where the complexity of feature matching and the sparsity of point clouds pose significant challenges. To handle abrupt changes in appearance features and sparse point clouds, we propose a novel 3D SOT network, dubbed CDTracker. It leverages both cosine similarity and an attention mechanism to enhance the robustness of feature matching. By combining similarity embedding and attention assignment, CDTracker performs template and search area feature matching in a coarse-to-fine manner. Additionally, CDTracker addresses the problem of sparse point clouds, which commonly leads to inaccurate tracking. It incorporates relatively dense sampling based on the concept of point cloud segmentation to retain more target points, leading to improved localization accuracy. Extensive experiments on both the KITTI and Waymo datasets demonstrate clear improvements in CDTracker over its competitors.

Stability and safety study of pumped storage units based on time-shifted multi-scale cosine similarity entropy

The diagnosis of vibration signals of pumped storage units is crucial to the safe and stable operation of the units. In this paper, a fault diagnosis method with high recognition rate and strong generalizability is proposed. Firstly, the original vibration signals are decomposed by Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) decomposition, the decomposed IMF1 components are proposed to be decomposed by Variational Mode Decomposition (VMD) for secondary decomposition, and then based on the fractal theory, a tool to measure the signal complexity is proposed - Time-Shifted Multiscale Cosine Similarity Entropy (TSMCSE). Finally, the entropy value of each component is input into the Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model as the feature value for diagnosis and identification. Through the measured data of unit operation, four states of 20 MW, 100 MW, 280 MW and 340 MW in the upper guide +X and −Y directions are used for validation, and the results show that the diagnostic accuracy of CEEMDAN-VMD-TSMCSE-CNN-LSTM reaches 100 %. At the same time, in order to experience the generalizability of the model, the open data set of rolling bearings from Jiangnan University is introduced, and four conditions in 600/rpm and 800 rpm are selected for re-validation, and the experimental results show that the accuracy of the model designed in this paper is 100 %. The method provides a new tool for the study of stability and safety of pumped storage units.

Automated Answersheet Evaluation using BERT

The research introduces an innovative approach to automating answersheet evaluation. The methodology combines the power of BERT (Bidirectional Encoder Representations from Transformers) and cosine similarity within the realm of natural language inference, allowing for a sophisticated analysis of student responses. BERT’s contextual embeddings provide a deep understanding of the semantic nuances present in answers, while cosine similarity serves as a robust metric for measuring the semantic alignment between student responses and ideal answers. Natural language inference enriches the model’s capacity to decipher inferential relationships, enabling a more nuanced eval- uation that goes beyond surface-level language comprehension. Augmented by natural language inference, our model adeptly captures inferential relationships, enhancing its comprehension of nuanced contextual nuances. The results underscore its potential to redefine automated grading paradigms, offering educators a powerful tool for efficient and unbiased assessments of student answers, thereby contributing to the advancement of educational technology and assessment methodologies.

Navigating academia: Designing and evaluating a multidimensional recommendation system for university and major selection

AbstractSelecting the right academic major significantly shapes an individual's future career path, making it a longstanding focus of research. The shift to online platforms, accelerated by the challenges posed by the coronavirus pandemic, has transformed counseling and guidance systems. Consequently, developing robust online support systems has become imperative for extending guidance to all students. This article introduces the design, development, and evaluation of “My Future Career,” a multidimensional recommendation system (RS) crafted to aid students in navigating university and academic major selection decisions. The system relies on three key student‐driven parameters: central university entrance exam scores, rankings, and occupational personality types, utilizing cosine similarity and normalized distance to align user and item profiles. Following the system's completion, an assessment was conducted using data from real users, revealing an impressive accuracy (hit rate 100%, precision 88%) in recommendations following the inclusion of contextual post‐filtering features. The findings not only highlight the system's effectiveness but also underscore the positive user experience, as students express contentment with its ease of use and practical utility. The results emphasize the endorsement of expert's regarding the system's consistency (52%), relevance (96%), and acceptance (96%) in providing recommendations.

An improved manifold-projection trajectory based method for chemical kinetic mechanism reduction

An improved manifold-projection trajectory based dimension reduction algorithm (MTDR) to automatically generate skeletal chemical kinetic mechanisms was proposed and validated in this study. This algorithm is implemented by constructing an approximate manifold-based trajectory of the combustion system using Euclidean distances between two adjacent points and cosine similarities between two adjacent vectors in the manifold space, and then evaluating the importance of species by calculating the error of manifold-based trajectories before and after the projection of the manifold. In this method, the nature of the combustion system is well reflected since the combustion process information can be captured by the location of the state point in the manifold space. The calculation during the reduction process is simplified by using the manifold-based trajectory instead of the complex differential equations. Compared to our previous work, this study improves the prediction accuracy of the reduced mechanisms through a more comprehensive error evaluation in the mechanism reduction process. The detailed mechanisms of fuels including iso-cetane, iso-octane and gasoline surrogate mixtures of toluene, iso-octane and n-heptane were reduced to portray the prowess of the algorithm. It’s demonstrated that the number of species of the fuels is reduced from 492, 254 and 1389 to 95, 87 and 427 within an acceptable error, respectively. Meanwhile, the case of iso-octane mechanism reduction was chosen to compare the performance of MTDR with traditional methods including path flux analysis (PFA) and directed relation graph with error propagation and sensitivity analysis (DRGEPSA), and it is shown that the MDTR can generate more compact mechanism under a broad range of error limits.