Canonical Correlation Research Articles

Evaluation of Ground Motion Damage Potential with Consideration of Compound Intensity Measures Using Principal Component Analysis and Canonical Correlation Analysis

Evaluation of Ground Motion Damage Potential with Consideration of Compound Intensity Measures Using Principal Component Analysis and Canonical Correlation Analysis

Students' Barriers and Emotional Presence in Online Learning: A Canonical Correlation Analysis

Addressing the barriers students face in online learning is critical to creating an inclusive, supportive, and practical educational experience that benefits many students. The aim of this study is to examine students' online learning barriers and emotional presence, which affect many variables in online learning environments, and the relationship between these variables. For this purpose, data was collected from 402 university students enrolled in the same course at a university. According to the data collected from the students, it can be said that the student's barriers to online learning are at a medium level for eight sub-factors ((a) administrative issues, (b) social interaction, (c) academic skills, (d) technical skills, (e) learner motivation, (f) time and support for studies, (g) cost and access to the Internet, and (h) technical problems). Similarly, it was found that students' emotional presence in online learning is at an average level for Receiving Emotions (x̄=3,438) and Giving Emotions (x̄=2,994) factors. Canonical correlation analysis was used to show the relationship between student barriers and emotional presence. As a result of this analysis, it was concluded that the canonical correlation was low. It was observed that the factor that contributed the most to the significant model among student barriers was the factor of administrative/instructor issues, and the factor of receiving emotions factor was one of the emotional presence factors.

Comparing the stability and reproducibility of brain-behavior relationships found using canonical correlation analysis and partial least squares within the ABCD sample

Abstract Canonical correlation analysis (CCA) and partial least squares correlation (PLS) detect linear associations between two data matrices by computing latent variables (LVs) having maximal correlation (CCA) or covariance (PLS). This study compared the similarity and generalizability of CCA- and PLS-derived brain-behavior relationships. Data were accessed from the baseline Adolescent Brain Cognitive Development (ABCD) dataset (N > 9,000, 9–11 years). The brain matrix consisted of cortical thickness estimates from the Desikan-Killiany atlas. Two phenotypic scales were examined separately as the behavioral matrix; the Child Behavioral Checklist (CBCL) subscale scores and NIH Toolbox performance scores. Resampling methods were used to assess significance and generalizability of LVs. LV1 for the CBCL brain relationships was found to be significant, yet not consistently stable or reproducible, across CCA and PLS models (singular value: CCA = .13, PLS = .39, p < .001). LV1 for the NIH brain relationships showed similar relationships between CCA and PLS and was found to be stable and reproducible (singular value: CCA = .21, PLS = .43, p < .001). The current study suggests that stability and reproducibility of brain-behavior relationships identified by CCA and PLS are influenced by the statistical characteristics of the phenotypic measure used when applied to a large population-based pediatric sample.

The Effects of Education and Financial Development on Energy Poverty Reduction in Latin America

Using data from 14 Latin American countries over the period 2000–2019, this study examines the influence of education and financial development on energy poverty reduction. Evidence from different estimation approaches, such as dynamic ordinary least squares (DOLS), fully modified ordinary least squares (FMOLS), and canonical correlation regression (CCR), showed that education substantially decreases energy poverty reduction in Latin America. The results also indicated that the nexus between financial development and energy poverty reduction is positively and significantly associated. Conversely, the linkage between education, financial development, and energy poverty reduction considerably varies among Latin American countries. We suggest that policies stimulating financial development and education will be pivotal in addressing energy poverty reduction in Latin America.

Fasting hyperglycaemia and fatty liver drive colorectal cancer: a retrospective analysis in 1145 patients.

Metabolic dysfunction-associated steatotic liver disease (MASLD) represents the hepatic manifestation of increased adiposopathy, whose pathogenetic features have been proposed as tumourigenic triggers for colorectal cancer (CRC). We aim to identify specific metabolic signatures involved in CRC development that may be used as non-invasive biomarkers, paving the way for specific and personalized strategies of CRC prevention and early detection. We retrospectively assessed CRC onset during a time frame of 8 years in a cohort of 1145 out-patients individuals who had previously been evaluated for Metabolic Syndrome. 28 patients developed CRC. No association between CRC development and visceral and general obesity was detected, while baseline fasting plasma glucose (FPG) and non-invasive liver fibrosis scores were significantly higher in patients with CRC, compared to those who did not develop cancer. Liver steatosis and MASLD were more frequently diagnosed in patients who developed CRC compared to no cancer developers. Canonical correlations among metabolic biomarkers were not present in CRC developers, differently from no cancer group. In ROC analysis, FPG and non-invasive scores also showed good sensitivity and specificity in predicting colon cancer. We then calculated ORs for metabolic biomarkers, finding that higher FPG and non-invasive scores were associated with an increased risk of developing CRC. MASLD and increased FPG may play a role in the clinical background of CRC, bringing to light the fascinating possibility of a reversed gut-liver axis communication in the pathogenesis of CRC. Thus, the use of non-invasive scores of fatty liver may be helpful to predict the risk of CRC and serve as novel prognostic factors for prevention and therapeutic strategies.

Exploring combinations of dimensionality reduction, transfer learning, and regularization methods for predicting binary phenotypes with transcriptomic data

BackgroundNumerous transcriptomic-based models have been developed to predict or understand the fundamental mechanisms driving biological phenotypes. However, few models have successfully transitioned into clinical practice due to challenges associated with generalizability and interpretability. To address these issues, researchers have turned to dimensionality reduction methods and have begun implementing transfer learning approaches.MethodsIn this study, we aimed to determine the optimal combination of dimensionality reduction and regularization methods for predictive modeling. We applied seven dimensionality reduction methods to various datasets, including two supervised methods (linear optimal low-rank projection and low-rank canonical correlation analysis), two unsupervised methods [principal component analysis and consensus independent component analysis (c-ICA)], and three methods [autoencoder (AE), adversarial variational autoencoder, and c-ICA] within a transfer learning framework, trained on > 140,000 transcriptomic profiles. To assess the performance of the different combinations, we used a cross-validation setup encapsulated within a permutation testing framework, analyzing 30 different transcriptomic datasets with binary phenotypes. Furthermore, we included datasets with small sample sizes and phenotypes of varying degrees of predictability, and we employed independent datasets for validation.ResultsOur findings revealed that regularized models without dimensionality reduction achieved the highest predictive performance, challenging the necessity of dimensionality reduction when the primary goal is to achieve optimal predictive performance. However, models using AE and c-ICA with transfer learning for dimensionality reduction showed comparable performance, with enhanced interpretability and robustness of predictors, compared to models using non-dimensionality-reduced data.ConclusionThese findings offer valuable insights into the optimal combination of strategies for enhancing the predictive performance, interpretability, and generalizability of transcriptomic-based models.

A novel applied climate classification method for assessing atmospheric influence on anomalous coastal water levels

AbstractClimate classification is a commonly used tool to simplify, visualize and make sense of an otherwise unwieldy amount of climate data in applied climate science research. Typically, these classifications have stemmed from one of two perspectives, either a circulation‐to‐environment (C2E) approach, or an environment‐to‐circulation approach (E2C), each with advantages and drawbacks. This research discusses a novel environment‐to‐circulation‐to‐environment (ECE) perspective to applied climate classification, and develops a specific ECE methodology that utilizes canonical correlation and discriminant analysis—the CANDECE method. The benefits of the ECE approach generally, and the CANDECE methodology specifically, are demonstrated in applying climate classification to aid in modelling anomalous water levels (AWLs) along portions of the East and West coasts of the United States. Results show that the CANDECE method performs better than two traditional classification methods (k‐means and self‐organizing maps [SOMs]) at relating AWLs to their broad‐scale atmospheric setups, especially with regard to both high and low extreme AWLs. It is further demonstrated that, compared with the West coast, the CANDECE method is particularly advantageous along the southeastern US coast, where the primary modes of atmospheric variability (which drive the classifications produced by SOMs and k‐means) do not align with the relevant circulation‐based factors driving AWL variability. While AWLs were utilized for demonstrating the ECE proof‐of‐concept herein, ECE and CANDECE are designed to be useful for any climate application.

Resting state EEG assisted imagined vowel phonemes recognition by native and non-native speakers using brain connectivity measures.

Communication is challenging for disabled individuals, but with advancement of brain-computer interface (BCI) systems, alternative communication systems can be developed. Current BCI spellers, such as P300, SSVEP, and MI, have drawbacks like reliance on external stimuli or conversation irrelevant mental tasks. In contrast to these systems, Imagined speech based BCI systems rely on directly decoding the vowels/words user is thinking, making them more intuitive, user friendly and highly popular among Brain-Computer-Interface (BCI) researchers. However, more research needs to be conducted on how subject-specific characteristics such as mental state, age, handedness, nativeness and resting state activity affects the brain's output during imagined speech. In an overt speech, it is evident that native and non-native speakers' brains function differently. Therefore, this paper explores how nativeness to language affects EEG signals while imagining vowel phonemes, using brain-map analysis and scalogram and also investigates the inclusion of features extracted from resting state EEG with imagined state EEG. The Fourteen-channel EEG for Imagined Speech (FEIS) dataset was used to analyse the EEG signals recorded while imagining vowel phonemes for 16 subjects (nine native English and seven non-native Chinese). For the classification of vowel phonemes, different connectivity measures such as covariance, coherence, and Phase Synchronous Index-PSI were extracted and analysed using statistics based Multivariate Analysis of Variance (MANOVA) approach. Different fusion strategies (difference, concatenation, Common Spatial Pattern-CSP and Canonical Correlation Analysis-CCA) were carried out to incorporate resting state EEG connectivity measures with imagined state connectivity measures for enhancing the accuracy of imagined vowel phoneme recognition. Simulation results revealed that concatenating imagined state and rest state covariance and PSI features provided the maximum accuracy of 92.78% for native speakers and 94.07% for non-native speakers.

Glucose Color Index: Development and Validation of a Novel Measure of the Shape of Glycemic Variability.

Standard continuous glucose monitoring (CGM) metrics: mean glucose, standard deviation, coefficient of variation, and time in range, fail to capture the shape of variability in the CGM time series. This information could facilitate improved diabetes management. We analyzed CGM data from 141 adults with type 2 diabetes in the Hyperglycemic Profiles in Obstructive Sleep Apnea (HYPNOS) trial. Participants in HYPNOS wore CGM sensors for up to two weeks at two time points, three months apart. We calculated the log-periodogram for each time period, summarizing using disjoint linear models. These summaries were combined into a single value, termed the Glucose Color Index (GCI), using canonical correlation analysis. We compared the between-wear correlation of GCI with those of standard CGM metrics and assessed associations between GCI and diabetes comorbidities in 398 older adults with type 2 diabetes from the Atherosclerosis Risk in Communities (ARIC) study. The GCI achieved a test-retest correlation of R = .75. Adjusting for standard CGM metrics, the GCI test-retest correlation was R = .55. Glucose Color Index was significantly associated (p < .05) with impaired physical functioning, frailty/pre-frailty, cardiovascular disease, chronic kidney disease, and dementia/mild cognitive impairment after adjustment for confounders. We developed and validated the GCI, a novel CGM metric that captures the shape of glucose variability using the periodogram signal decomposition. Glucose Color Index was reliable within participants over a three-month period and associated with diabetes comorbidities. The GCI suggests a promising avenue toward the development of CGM metrics which more fully incorporate time series information.

Big data analytics capabilities, direct and mediating relationships with innovative and business performance

This study examines the effect of key big data analytics capabilities (data, skills, technology, and culture) on innovation performance and business performance. Data were gathered from 91 companies and analyzed to determine correlations in the proposed model. The results find that big data analytics capabilities (BDAC) partially mediate the effect between innovative performance and business performance. Further, as a company’s performance is a multidimensional element, was necessary to analyze more than one attribute to evaluate the relationship with BDAC through a canonical correlation analysis. The results in this sense reveal that the four big data capabilities increase the growth of sales, revenue, the number of workers, the net profit margin, innovation management, the development of new products and services, and the adoption of new information technologies.

The covariant structural and functional neuro-correlates of cognitive impairments in patients with end-stage renal diseases.

Cognitive impairment (CI) is a common complication of end-stage renal disease (ESRD) that is associated with structural and functional changes in the brain. However, whether a joint structural and functional alteration pattern exists that is related to CI in ESRD is unclear. In this study, instead of looking at brain structure and function separately, we aim to investigate the covariant characteristics of both functional and structural aspects. Specifically, we took the fusion analysis approach, namely, multimodal canonical correlation analysis and joint independent component analysis (mCCA+jICA), to jointly study the discriminative features in gray matter volume (GMV) measured by T1-weighted (T1w) MRI, fractional anisotropy (FA) in white matter measured by diffusion MRI, and the amplitude of low-frequency fluctuation (ALFF) measured by blood oxygenation-level-dependent (BOLD) MRI in 78 ESRD patients versus 64 healthy controls (HCs), followed by a mediation effect analysis to explore the relationship between neuroimaging findings, cognitive impairments and uremic toxins. Two joint group-discriminative independent components (ICs) were found to show covariant abnormalities across FA, GMV, and ALFF (all p < 0.05). The most dominant joint IC revealed associative patterns of alterations of GMV (in the precentral gyrus, occipital lobe, temporal lobe, parahippocampal gyrus, and hippocampus), alterations of ALFF (in the precuneus, superior parietal gyrus, and superior occipital gyrus), and of white matter FA (in the corticospinal tract and inferior frontal occipital fasciculus). Another significant IC revealed associative alterations of GMV (in the dorsolateral prefrontal and orbitofrontal cortex) and FA (in the forceps minor). Moreover, the brain changes identified by FA and GMV in the above-mentioned brain regions were found to mediate the negative correlation between serum phosphate and mini-mental state examination (MMSE) scores (all p < 0.05). The mCCA+jICA method was demonstrated to be capable of revealing covariant abnormalities across neuronal features of different types in ESRD patients as contrasted to HCs, and joint brain changes may play an important role in mediating the relationship between serum toxins and CIs in ESRD. Our results show the mCCA+jICA fusion analysis approach may provide new insights into similar neurobiological studies.

Predicting conditional probability distributions of redshifts of Active Galactic Nuclei using Hierarchical Correlation Reconstruction

Abstract The Large Area Telescope (LAT) on board the Fermi Gamma-ray observatory continuously scans the sky in an energy range from 50 MeV to 1 TeV. The telescope has identified over 6000 gamma-ray emitting sources, approximately half of which are classified as active galactic nuclei (AGN). However, not all of these gamma-ray sources have known redshift values for the reason that red-shift estimation following traditional methods can be an expensive, challenging task. Alternatively, as an effort to robustly predict the AGN redshift values, many researchers have recently turned to machine learning methods. However, while the focus has primarily been on predicting specific values, real-world data often allows us only to predict conditional probability distributions, constrained by conditional entropy (H(Y|X)). In our study, we employ the Hierarchical Correlation Reconstruction (HCR) approach to inexpensively predict complex conditional probability distributions, including multimodal ones. This is achieved through independent Mean Squared Error (MSE) estimation of multiple moment-like parameters, combined into reconstruction of the conditional distribution. By employing linear regression for this purpose, we can develop interpretable models where coefficients describe the contributions of features to conditional moments. This article extends the original approach by incorporating Canonical Correlation Analysis (CCA) for feature optimization and l1 ‘lasso’ regularization. Our primary focus is on the practical problem of predicting the redshift of Active Galactic Nuclei (AGN) using data from the Fourth Fermi-LAT Data Release 3 (4LAC-DR3) dataset.

Parallel Hierarchical Multi-View Feature Fusion based on Canonical Correlation Analysis for Mammogram Retrieval

Parallel Hierarchical Multi-View Feature Fusion based on Canonical Correlation Analysis for Mammogram Retrieval

Bacterial and fungal community structures in Hulun Lake are regulated by both stochastic processes and environmental factors.

Microorganisms are a crucial component of lake ecosystems and significant contributors to biogeochemical cycles. However, the understanding of how primary microorganism groups (e.g., bacteria and fungi) are distributed and constructed within different lake habitats is lacking. We investigated the bacterial and fungal communities of Hulun Lake using high-throughput sequencing techniques targeting 16S rRNA and Internal Transcribed Spacer 2 genes, including a range of ecological and statistical methodologies. Our findings reveal that environmental factors have high spatial and temporal variability. The composition and community structures vary significantly depending on differences in habitats. Variance partitioning analysis showed that environmental and geographical factors accounted for <20% of the community variation. Canonical correlation analysis showed that among the environmental factors, temperature, pH, and dissolved oxygen had strong control over microbial communities. However, the microbial communities (bacterial and fungal) were primarily controlled by the dispersal limitations of stochastic processes. This study offers fresh perspectives regarding the maintenance mechanism of bacterial and fungal biodiversity in lake ecosystems, especially regarding the responses of microbial communities under identical environmental stress.IMPORTANCELake ecosystems are an important part of the freshwater ecosystem. Lake microorganisms play an important role in material circulation and energy flow owing to their unique enzymatic and metabolic capacity. In this study, we observed that bacterial and fungal communities varied widely in the water and sediments of Hulun Lake. The primary factor affecting their formation was identified as dispersal limitation during stochastic processes. Environmental and geographical factors accounted for <20% of the variation in bacterial and fungal communities, with pH, temperature, and dissolved oxygen being important environmental factors. Our findings provide new insights into the responses of bacteria and fungi to the environment, shed light on the ecological processes of community building, and deepen our understanding of lake ecosystems. The results of this study provide a reference for lake management and conservation, particularly with respect to monitoring and understanding microbial communities in response to environmental changes.

A comparative study of stereo-dependent SSVEP targets and their impact on VR-BCI performance.

Steady-state visual evoked potential brain-computer interfaces (SSVEP-BCI) have attracted significant attention due to their ease of deployment and high performance in terms of information transfer rate (ITR) and accuracy, making them a promising candidate for integration with consumer electronics devices. However, as SSVEP characteristics are directly associated with visual stimulus attributes, the influence of stereoscopic vision on SSVEP as a critical visual attribute has yet to be fully explored. Meanwhile, the promising combination of virtual reality (VR) devices and BCI applications is hampered by the significant disparity between VR environments and traditional 2D displays. This is not only due to the fact that screen-based SSVEP generally operates under static, stable conditions with simple and unvaried visual stimuli but also because conventional luminance-modulated stimuli can quickly induce visual fatigue. This study attempts to address these research gaps by designing SSVEP paradigms with stereo-related attributes and conducting a comparative analysis with the traditional 2D planar paradigm under the same VR environment. This study proposed two new paradigms: the 3D paradigm and the 3D-Blink paradigm. The 3D paradigm induces SSVEP by modulating the luminance of spherical targets, while the 3D-Blink paradigm employs modulation of the spheres' opacity instead. The results of offline 4-object selection experiments showed that the accuracy of 3D and 2D paradigm was 85.67 and 86.17% with canonical correlation analysis (CCA) and 86.17 and 91.73% with filter bank canonical correlation analysis (FBCCA), which is consistent with the reduction in the signal-to-noise ratio (SNR) of SSVEP harmonics for the 3D paradigm observed in the frequency-domain analysis. The 3D-Blink paradigm achieved 75.00% of detection accuracy and 27.02 bits/min of ITR with 0.8 seconds of stimulus time and task-related component analysis (TRCA) algorithm, demonstrating its effectiveness. These findings demonstrate that the 3D and 3D-Blink paradigms supported by VR can achieve improved user comfort and satisfactory performance, while further algorithmic optimization and feature analysis are required for the stereo-related paradigms. In conclusion, this study contributes to a deeper understanding of the impact of binocular stereoscopic vision mechanisms on SSVEP paradigms and promotes the application of SSVEP-BCI in diverse VR environments.

A Review of Statistical-Based Fault Detection and Diagnosis with Probabilistic Models

As industrial processes grow increasingly complex, fault identification becomes challenging, and even minor errors can significantly impact both productivity and system safety. Fault detection and diagnosis (FDD) has emerged as a crucial strategy for maintaining system reliability and safety through condition monitoring and abnormality recovery to manage this challenge. Statistical-based FDD methods that rely on large-scale process data and their features have been developed for detecting faults. This paper overviews recent investigations and developments in statistical-based FDD methods, focusing on probabilistic models. The theoretical background of these models is presented, including Bayesian learning and maximum likelihood. We then discuss various techniques and methodologies, e.g., probabilistic principal component analysis (PPCA), probabilistic partial least squares (PPLS), probabilistic independent component analysis (PICA), probabilistic canonical correlation analysis (PCCA), and probabilistic Fisher discriminant analysis (PFDA). Several test statistics are analyzed to evaluate the discussed methods. In industrial processes, these methods require complex matrix operation and cost computational load. Finally, we discuss the current challenges and future trends in FDD.

Enhancing cross-subject transfer performance for SSVEP identification using small data-based transferability evaluation

Cross-subject steady-state visual evoked potential-based brain-computer interfaces (SSVEP-BCIs) have attracted increasing attention in recent years due to their potential to reduce calibration efforts and improve usability. This study proposed an efficient cross-subject transfer method for SSVEP-BCI, which significantly reduced the calibration effort and enhanced system performance. By employing a minimal dataset obtained from the target subject, specifically, a single trial for each frequency, we calculated the similarity between the SSVEPs of the target subject and the source subjects. Based on the similarities, a transferability evaluation model was constructed, which was then used to assign weights to the source subjects. The features calculated from the source subjects were then weighted and summed to obtain the frequency recognition results. The proposed cross-subject transfer method was evaluated in two pipelines based on the ensemble task-related component analysis (eTRCA) and task-discriminant component analysis (TDCA), respectively. The proposed method in eTRCA-pipeline initially achieving accuracies of 96.98 ± 3.66 % and 88.97 ± 14.66 % in our dataset and the Benchmark dataset, respectively, at a data length of 1 s. After fusing the features of eTRCA with those of the extended canonical correlation analysis (eCCA), the method exhibited enhanced performance, elevating accuracies to 98.23 ± 2.68 % and 92.83 ± 10.75 %. Similarly, in TDCA-pipeline, accuracies initially reached 97.47 ± 3.88 % and 93.53 ± 10.22 % in our dataset and the Benchmark dataset, respectively, and after fusing TDCA features with eCCA features, they remarkably increased to 98.13 ± 2.89 % and 95.00 ± 8.43 %, respectively. This study offers a promising solution for the fast and effective calibration of SSVEP-BCI, paving the way for more practical and accessible BCI systems.

Genomic Prediction of Root Traits via Aerial Traits in Soybean Using Canonical Variables

The phenotypic evaluation of root traits in soybeans presents challenges in breeding due to its high cost and the requirement for experimental plot destruction. Establishing relationships between aerial and root traits is crucial, given the relative ease of phenotypic evaluations for aerial traits. Therefore, this study aims to utilize the canonical correlation technique to estimate latent variables, subsequently employing GBLUP for the genomic prediction of the root traits (length, volume, surface area, and dry mass) using phenotypic information from aerial part traits (hypocotyl diameter and dry mass). Our results demonstrate the effectiveness of the technique in predicting the root part, even when not directly evaluated. The agreement observed between the top 10% of individuals selected based on the canonical variable and each root trait individually was considered moderate or substantial. This enables the simultaneous selection of genotypes based on both trait groups, providing a valuable approach for soybean breeding programs.

Deep Canonical Correlation Fusion Algorithm Based on Denoising Autoencoder for ASD Diagnosis and Pathogenic Brain Region Identification.

Autism Spectrum Disorder (ASD) is defined as a neurodevelopmental condition distinguished by unconventional neural activities. Early intervention is key to managing the progress of ASD, and current research primarily focuses on the use of structural magnetic resonance imaging (sMRI) or resting-state functional magnetic resonance imaging (rs-fMRI) for diagnosis. Moreover, the use of autoencoders for disease classification has not been sufficiently explored. In this study, we introduce a new framework based on autoencoder, the Deep Canonical Correlation Fusion algorithm based on Denoising Autoencoder (DCCF-DAE), which proves to be effective in handling high-dimensional data. This framework involves efficient feature extraction from different types of data with an advanced autoencoder, followed by the fusion of these features through the DCCF model. Then we utilize the fused features for disease classification. DCCF integrates functional and structural data to help accurately diagnose ASD and identify critical Regions of Interest (ROIs) in disease mechanisms. We compare the proposed framework with other methods by the Autism Brain Imaging Data Exchange (ABIDE) database and the results demonstrate its outstanding performance in ASD diagnosis. The superiority of DCCF-DAE highlights its potential as a crucial tool for early ASD diagnosis and monitoring.

Applying fluorescence spectroscopy coupled with Gaussian band fitting to reveal dynamic variation process of humus fractions from riparian soils along an urbanized river

Humus, an important fraction of soil organic matter, play an environmental role on nutrients, organic and inorganic pollutants in riparian zones of urbanized rivers. In this study, dynamic variation process of humus fractions from riparian soils was revealed along Puhe River. Composite soil samples of four depths were collected from four land-uses, i.e., eco-conservation area (ECA), industrial area (INA), urban/town area (UTA), rural/agricultural area (RAA). Based on synchronous fluorescence spectra coupled with Gaussian band fitting, fulvic/humic acid predominantly contained tyrosine-like (TYLF), tryptophan-like (TRLF), microbial-like (MLF), fulvic-like (FLF) and humic-like (HLF) substances within each soil profile. TRLF, MLF and FLF (89.43 %–90.30 %) are the representative components in fulvic-acid, while MLF and HLF (52.81–59.97 %) in humic-acid. Phenolic, carboxylic and humified materials were present in both humus. According to 2-dimensitonal correlation spectroscopy and canonical correlation analysis, fulvic/humic acid within the ECA soil profile could be mainly derived from the degradations of terrestrial plant metabolites and residuals. Within the INA, fulvic-acid could be associated with treated/untreated wastewater, which entered the river and flew into the riparian during high flow period; whereas humic-acid could be relative to the terrestrials. Fulvic-acid had the same source as humic-acid in the UTA, which might be concerned with scattered domestic sewage and livestock wastewater, rather than the fluvial water. Furthermore, the source of fulvic/humic acid in the RAA was the crop metabolites and residuals, apart from the livestock wastewater. Noticeably, the variations of humus fractions in the ECA and RAA roughly occurred in 0–60 cm, while approximately in 20–80 cm in the INA and UTA. This proved that humus fractions in the former were referred to the plant/crop residuals, whereas humus fractions in the latter were those the terrestrials and fluvial water. This study could provide a key support for the construction and restoration of the urbanized riparian zone.