Prerequisite-link Prediction (PLP) aims to discover the condition relations of a specific event or a concerned variable, which is a fundamental problem in a large number of fields, such as educational data mining. Current studies on PLP usually developed graph neural networks (GNNs) to learn the representations of pairs of nodes. However, these models fail to distinguish non-isomorphic graphs and integrate multiscale structures, leading to the insufficient expressive capability of GNNs. To this end, we in this paper proposedk-dimensional Weisferiler-Leman directed GNNs, dubbedk-WediGNNs, to recognize non-isomorphic graphs via the Weisferiler-Leman algorithm. Furthermore, we integrated the multiscale structures of a directed graph intok-WediGNNs, dubbed multiscalek-WediGNNs, from the bidirected views of in-degree and out-degree. With the Siamese network, the proposed models are extended to address the problem of PLP. Besides, the expressive power is then interpreted via theoretical proofs. The experiments were conducted on four publicly available datasets for concept prerequisite relation prediction (CPRP). The results show that the proposed models achieve better performance than the state-of-the-art approaches, where our multiscalek-WediGNN achieves a new benchmark in the task of CPRP.

Acquiring big-size datasets to raise the performance of deep models has become one of the most critical problems in representation learning (RL) techniques, which is the core potential of the emerging paradigm of federated learning (FL). However, most current FL models concentrate on seeking an identical model for isolated clients and thus fail to make full use of the data specificity between clients. To enhance the classification performance of each client, this study introduces the FDRL, a federated discriminative RL model, by partitioning the data features of each client into a global subspace and a local subspace. More specifically, FDRL learns the global representation for federated communication between those isolated clients, which is to capture common features from all protected datasets via model sharing, and local representations for personalization in each client, which is to preserve specific features of clients via model differentiating. Toward this goal, FDRL in each client trains a shared submodel for federated communication and, meanwhile, a not-shared submodel for locality preservation, in which the two models partition client-feature space by maximizing their differences, followed by a linear model fed with combined features for image classification. The proposed model is implemented with neural networks and optimized in an iterative manner between the server of computing the global model and the clients of learning the local classifiers. Thanks to the powerful capability of local feature preservation, FDRL leads to more discriminative data representations than the compared FL models. Experimental results on public datasets demonstrate that our FDRL benefits from the subspace partition and achieves better performance on federated image classification than the state-of-the-art FL models.

Understanding the semantic concept recognition in the learning brain benefits disorder treatment and high-quality education. This paper provides a study on identifying different semantic concepts from fMRIs to investigate their different brain activities. In this study, we extracted the functional connection matrixes from fMRIs after their pre-proceedings. To recognize the semantic concepts, we employed two deep learning models, i.e., LeNet and ResNet, compared to the classic support vector machine. Experimental evaluations were conducted on a publicly available dataset to recognize the word classes, including Nonu, verb, and adjective. The results manifest that two deep models perform much better than SVM in accuracy, precision, and recall, whereas ResNet is better than LeNet. That is to say, there exists a different pattern between different concepts. In addition, we also used the mentioned models to identify the different concepts for each individual, indicating their different patterns in individuals. This study contributes to understanding human cognition and language processing and puts forward prospects for disorder treatment and education design.

Functional magnetic resonance imaging (fMRI) can measure changes in blood oxygenation level-dependent (BOLD) in the human brain caused by some stimuli or tasks, which helps us understand human brain mechanisms and functions. Recent studies demonstrate that there are both shared and specific neural representations between nature and drawing images. However, there is a lack of more detailed studies on the differences at the various levels of image abstraction. In this paper, we proposed to recognize concept levels from the image abstractions, including photographs, drawings, and sketches. More specifically, this study first conducts preprocessing processes to mitigate fMRI noise, such as respiratory and head movement, and then constructs the functional connectivity matrix based on the fMRI segmentation corresponding to different stimuli, leading to our used datasets. On the resulting dataset, we trained several data classifiers to obtain the mapping from fMRIs to the three types of stimuli. In addition, we discussed experimental parameters to check their impacts on classification performance. The evaluated results show that different concept-level images could be recognized at an effective accuracy, where the deep learning model achieves the best performance. This study contributes to an understanding of the abstraction level of concept formulation in the brain. The results can help treat brain disorders and make learning plans.

Predicting and understanding student learning performance has been a long-standing task in learning science, which can benefit personalized teaching and learning. This study shows that the progress towards this task can be accelerated by using learning record data to feed a deep learning model that considers the intrinsic course association and the structured features. We proposed a multi-source sparse attention convolutional neural network (MsaCNN) to predict the course grades in a general formulation. MsaCNN adopts multi-scale convolution kernels on student grade records to capture structured features, a global attention strategy to discover the relationship between courses, and multiple input-heads to integrate multi-source features. All achieved features are then poured into a softmax classifier towards an end-to-end supervised deep learning model. Conducting insights into higher education on real-world university datasets, the results show that MsaCNN achieves better performance than traditional methods and delivers an interpretation of student performance by virtue of the resulted course relationships. Inspired by this interpretation, we created an association map for all mentioned courses, followed by evaluating the map with a questionnaire survey. This study provides computer-aided system tools and discovers the course-space map from the educational data, potentially facilitating the personalized learning progress.

Breast cancer is a heterogeneous disease caused by various alterations in the genome or transcriptome. Molecular subtypes of breast cancer have been reported, but useful biomarkers remain to be identified to uncover underlying biological mechanisms and guide clinical decisions. Towards biomarker discovery, several studies focus on genomic alterations that provide differences, while few works concern transcriptomic characterizations that mediate tumor progression. Rather than using differential expression (DE) or weighted network analysis, we propose a feature selection method, dubbed GLassonet, to identify discriminative biomarkers from transcriptome-wide expression profiles by embedding the relationship graph of high-dimensional expressions into the Lassonet model. GLassonet comprises a nonlinear neural network for identifying cancer subtypes, a skipping fully connected layer for canceling the connections of hidden layers from input features to output categories, and a graph enhancement for preserving the discriminative graph into the selected subspace. First, an iterative optimization algorithm learns model parameters on the TCGA breast cancer dataset to investigate the classification performance. Then, we probe the distribution patterns of GLassonet-selected gene sets across the cancer subtypes and compare them to gene sets outputted from the state-of-the-art. More profoundly, we conduct the overall survival analysis on three GLassonet-selected new marker genes, i.e., SOX10, TPX2, and TUBA1C, to investigate their expression changes and assess their prognostic impacts. Finally, we perform the enrichment analysis to discover the functional associations of the GLassonet-selected genes with GO terms and KEGG pathways. Experimental results show that GLassonet has a powerful ability to select the discriminative genes, which improve cancer subtype classification performance and provide potential biomarkers for cancer personalized therapy.

This paper proposes a representation learning model to identify task-state fMRIs for knowledge-concept recognition, which has the potential to model the human cognitive expression system. The traditional CNN-LSTM is usually employed to learn deep features from fMRIs, where CNN aims at extracting the spatial structure and LSTM accounts for the temporal structure. However, the manifold smoothness of the latent features caused by the fMRI sequence is often ignored, leading to unsteady data representation. In this paper, we model latent features as a hidden Markov chain and introduce a Markov-guided Spatio-Temporal Network (MSTNet) for brain image representation. Concretely, MSTNet has three parts: CNN that aims to learn latent features from 3D fMRI frames where a Markov Regularization enforces the neighborhood frames to have similar features, LSTM integrates all frames of an fMRI sequence into a feature vector and fully connected network (FCN) that is to implement the brain image classification. Our model is trained towards minimizing the cross entropy (CE) loss. Our experiment is conducted on the brain fMRI datasets achieved by scanning college students when they were learning five concepts of computer science. The results show that the proposed MSTNet can benefit from the introduced Markov regularization and thus result in improved performance on the brain activity classification. This study not only shows an effective fMRI classification model with Markov regularization but also provides the potential to understand brain intelligence and help patients with language disabilities.

The joint problem of imbalance samples and noisy labels challenges the current text classifiers in real-world applications. Existing approaches are mostly devoted to handling either former or latter while fail to manage the fused issue. This paper introduces a novel weakly-supervised framework, dubbed WeSt-coin, to take into account the sensitivity cost on misclassifications between classes and seek seed words towards noisy-label corrections. After BERT that creates a contextualized corpus, WeStcoin learns a predicted label vector from the contextualized samples and meanwhile calculates a pseudo probability vector from seed words, and then projects the concatenated representation into an output space, followed by multiplying by a cost-sensitive matrix. WeStcoin is ultimately trained to decrease the residual between the model outputs and the noisy labels, where seed words are also updated in an iterative manner. Extensive experiments and ablation studies on two public text datasets demonstrate that the proposed model outperforms the state-of-the-art model in the text classification with imbalance samples and noisy labels. Codes are made available at https://github.com/ypzhaang.

Long non-coding RNAs (lncRNAs) can disrupt the biological functions of protein-coding genes (PCGs) to cause cancer. However, the relationship between lncRNAs and PCGs remains unclear and difficult to predict. Machine learning has achieved a satisfactory performance in association prediction, but to our knowledge, it is currently less used in lncRNA-PCG association prediction. Therefore, we introduce GAE-LGA, a powerful deep learning model with graph autoencoders as components, to recognize potential lncRNA-PCG associations. GAE-LGA jointly explored lncRNA-PCG learning and cross-omics correlation learning for effective lncRNA-PCG association identification. The functional similarity and multi-omics similarity of lncRNAs and PCGs were accumulated and encoded by graph autoencoders to extract feature representations of lncRNAs and PCGs, which were subsequently used for decoding to obtain candidate lncRNA-PCG pairs. Comprehensive evaluation demonstrated that GAE-LGA can successfully capture lncRNA-PCG associations with strong robustness and outperformed other machine learning-based identification methods. Furthermore, multi-omics features were shown to improve the performance of lncRNA-PCG association identification. In conclusion, GAE-LGA can act as an efficient application for lncRNA-PCG association prediction with the following advantages: It fuses multi-omics information into the similarity network, making the feature representation more accurate; it can predict lncRNA-PCG associations for new lncRNAs and identify potential lncRNA-PCG associations with high accuracy.

Sepsis is a systemic inflammatory response caused by pathogens such as bacteria. Because its pathogenesis is not clear, the clinical manifestations of patients vary greatly, and the alarming incidence and mortality pose a great threat to patients and medical systems, especially in the ICU (Intensive Care Unit). The traditional judgment criteria have the problem of low specificity. Artificial intelligence models could greatly improve the accuracy of sepsis prediction and judgment. Based on the XGBoost machine learning framework taking demographic, vital signs, laboratory tests and medical intervention data as input, this paper proposes a novel model for dynamically predicting sepsis and assessing risk. To realize the model, two methods for feature construction are introduced. For the observed time-series data of vital signs and laboratory tests, the time-dependent method performs to construct the time-dependent characteristics after the statistical screening. For the clinical intervention data, the statistical counting method is applied to construct count-dependent characteristics. Moreover, a new objective function is proposed for the XGBoost framework, and the first-order and second-order gradients of the objective function are also given for model training. Compared with the state-of-the-art methods at present, the proposed model has the best performance, with AUROC improved by 5.4% on the MIMIC-III dataset and 2.1% on PhysioNet Challenge 2019 dataset. The data processing and training methods of this model can be conveniently applied in different electronic health record systems and has a wide application prospect.

Tumor-liltrating lymphocytes (TILs) are predictive for response to neoadjuvant treatment in tumors. Still, the abundance of tumor-liltrating cell types has not yet been produced in large quantities, hampering researchers exploring their characteristics. As the levels of genomics or transcriptomics could reflect changes in cell-type proportions, several computational tools have been developed to estimate cell-type abundances based on the reference gene expression proliles. Differential expression analysis is the most widely used to recognize marker genes. However, it ignores the correlation between genes. To this end, we propose a feature selection method, dubbed Discriminative Concrete Autoencoder (DCAE), to identify informative genes on single-cell RNA-seq data, which are then used to quantity cell-type proportions. To evaluate the performance of DCAE on selecting discriminative genes, we conduct experiments on our collected and processed single-cell RNA-seq dataset. First, we compare DCAE to the original Concrete Autoencoder by the cell-type classification accuracies resulting from their selected genes. Then we infer cell-type abundance by using deconvolution function with the chosen small cohort of genes. Next, we evaluate the deconvolution accuracy by the Pearson correlation coefficient between the estimated cell-type proportions and the true proportions, and the corresponding P-value. Finally, we compare the effects of the selected genes and the differential expression genes on the deconvolution accuracy. The results show that our selected genes by DCAE have higher discriminant power to distinguish cell types and effectively infer cell-type abundance. Thus, DCAE provides insights into acquiring candidate biomarkers for cell-type quantification.

Different subtypes of the same cancer often show distinct genomic signatures and require targeted treatments. The differences at the cellular and molecular levels of tumor microenvironment in different cancer subtypes have significant effects on tumor pathogenesis and prognostic outcomes. Although there have been significant researches on the prognostic association of tumor infiltrating lymphocytes in selected histological subtypes, few investigations have systemically reported the prognostic impacts of immune cells in molecular subtypes, as quantified by machine learning approaches on multi-omics datasets. This paper describes a new computational framework, ProTICS, to quantify the differences in the proportion of immune cells in tumor microenvironment and estimate their prognostic effects in different subtypes. First, we stratified patients into molecular subtypes based on gene expression and methylation profiles by applying nonnegative tensor factorization technique. Then we quantified the proportion of cell types in each specimen using an mRNA-based deconvolution method. For tumors in each subtype, we estimated the prognostic effects of immune cell types by applying Cox proportional hazard regression. At the molecular level, we also predicted the prognosis of signature genes for each subtype. Finally, we benchmarked the performance of ProTICS on three TCGA datasets and another independent METABRIC dataset. ProTICS successfully stratified tumors into different molecular subtypes manifested by distinct overall survival. Furthermore, the different immune cell types showed distinct prognostic patterns with respect to molecular subtypes. This study provides new insights into the prognostic association between immune cells and molecular subtypes, showing the utility of immune cells as potential prognostic markers. Availability: R code is available at https://github.com/liu-shuhui/ProTICS.

BACKGROUND: The classification of cancer subtypes is of great importance to cancer disease diagnosis and therapy. Many supervised learning approaches have been applied to cancer subtype classification in the past few years, especially of deep learning based approaches. Recently, the deep forest model has been proposed as an alternative of deep neural networks to learn hyper-representations by using cascade ensemble decision trees. It has been proved that the deep forest model has competitive or even better performance than deep neural networks in some extent. However, the standard deep forest model may face overfitting and ensemble diversity challenges when dealing with small sample size and high-dimensional biology data. RESULTS: In this paper, we propose a deep learning model, so-called BCDForest, to address cancer subtype classification on small-scale biology datasets, which can be viewed as a modification of the standard deep forest model. The BCDForest distinguishes from the standard deep forest model with the following two main contributions: First, a named multi-class-grained scanning method is proposed to train multiple binary classifiers to encourage diversity of ensemble. Meanwhile, the fitting quality of each classifier is considered in representation learning. Second, we propose a boosting strategy to emphasize more important features in cascade forests, thus to propagate the benefits of discriminative features among cascade layers to improve the classification performance. Systematic comparison experiments on both microarray and RNA-Seq gene expression datasets demonstrate that our method consistently outperforms the state-of-the-art methods in application of cancer subtype classification. CONCLUSIONS: The multi-class-grained scanning and boosting strategy in our model provide an effective solution to ease the overfitting challenge and improve the robustness of deep forest model working on small-scale data. Our model provides a useful approach to the classification of cancer subtypes by using deep learning on high-dimensional and small-scale biology data.

The classification of cancer subtypes is of great importance in cancer disease diagnosis and therapy. Many supervised learning methods have been applied to classification of cancer subtypes in the past few years, especially of deep learning based methods. Recently, a deep forest model has been proposed as an alternative of deep neural networks to learn hyper-representations by using cascade ensemble decision trees, and it has been proved that deep forest model has competitive or even better performance than deep neural networks. However, the original deep forest may face under-fitting and ensemble diversity problems when dealing with small sample size, and high-dimension biology data. It is important to improve the deep forest model to work better on small-scale biology data. In this paper, we propose a deep learning model to follow the mission of cancer subtype classification on small-scale biology data sets, which can be viewed as modification of original deep forest model. Our model distinguishes from the original deep forest model with two main contributions: First, a named multi-class-scanning method is proposed to train multiple simple binary classifiers to encourage diversity of ensemble. Meanwhile, the fitting quality of each classifier is considered in representations learning. Second, we propose a boosting strategy to emphasize more important features in cascade forests of representations learning, thus to propagate the benefits of discriminative features among layers to improve the overall classification performance. Systematical experiments on both microarray and RNA-seq data sets demonstrate that our method consistently outperforms the most state-of-the-art classification methods in application of cancer subtype classifications.

In this paper, we propose a distributed resource allocation mechanism with inter-cell interference (ICI) coordination for practical orthogonal frequency division multiple access (OFDMA) uplink. The designed scheme combines soft frequency reuse (SFR) and interference limited power control of neighboring cells. It avoids severe interference to neighboring cells through adjusting the interference limited power constraints according to the load variations across cells. The computational complexity is greatly lowered by decomposing a multi-cell optimization problem into distributed single-cell sub-problems. The simulation results demonstrate that the proposed scheme not only guarantees the cell-edge UEs' quality of service (QoS) but also improves the overall system performance significantly. Moreover, this scheme can adapt to the load variations in a certain cell or across cells.

The problem of distributed coordination in multi-cell wireless network has been concerned in the research community recently due to its ability to enhance intended signal and mitigate interference in the system. In the previous research, the design of beamforming weights at transmitter has been addressed from the perspective of egoism and altruism concepts considering the tradeoff between egoistic and altruistic solutions. The main contribution of this paper is to extend the existing strategy and exploit a much simpler hybrid coordination scheme with semi-distributed mode in multi-cell wireless network using the analytic hierarchy process. The significant advantage of our approach is to provide transmitter with a method which is easier to achieve compared with the existing pattern in a semi-distributed way. The performance of proposed scheme is finally illustrated through the numerical simulation results.

Cooperated multi-point technique in HSPA+system has not been paid attention until the Single Frequency Dual Cell (SF-DC) mode, where one user equipment (UE) may be served by dual NodeBs simultaneously in the single carrier frequency network, is proposed recently. In the previous research, the system-level performance with SF-DC operation had not been studied comprehensively. The main contribution of this paper is to exploit SF-DC transmit mode in the perspective of the system level and evaluate the system-level performance of SF-DC in HSPA+system with different simulation scenes. The impact of SF-DC on legacy UEs and the load balancing capability of SF-DC in asymmetric loading case are investigated properly in different channel conditions. System-level simulation results show that SF-DC transmit mode performs better compared with single point transmission and the influence of SF-DC mode on legacy UEs is negligible. It also reflects that SF-DC has the ability to balance load in asymmetric loading case.

In this paper, we propose a new efficient distributed dynamic resource allocation (DDRA) scheme to coordinate the inter-cell interference and verify the existence of the Nash equilibrium. This new scheme can be utilized in the long term evolution (LTE) uplink and downlink scenarios to mitigate the inter-cell interference effectively. The system level simulations show that the system performance is improved compared with the conventional random resource allocation scheme.

To meet the increasing demand for mobile broadband, multiple carrier features were subsequently introduced into UMTS Release-8 to Release-10. As a consequence, the UE capabilities of rejecting inter-stream interference have increased considerably. To fully utilize the UE capabilities and wireless resources, multi-point transmissions raised concerns in 3GPP High Speed Downlink Packet Access (HSDPA) standardizing process. Multi-point transmission is a promising technique to increase both the average cell throughput and cell edge user throughput. A novel cooperative sector selection algorithm, which takes both channel condition and sector loading into consideration, and an adaptive precoding transmission scheme are proposed for HSPA multi-point transmission. Through the system level simulation, the proposed transmission scheme effectively outperforms the non-cooperative system and non-adaptive cooperative scheme on both the average sector throughput and sector-edge UEs' rates significantly.

Transmit diversity can be regarded as a technology in wireless communication in order to enhance the system performance. Some relevant methods of transmit diversity have been presented on the basis of link-level research but do not consider multiple access interference (MAI) existing among users in the uplink of multi-user system. In this paper, we proposed one uplink transmit diversity scheme, which considers the effect of MAI in HSUPA. System-level simulation shows our strategy can degrade the interference level at receiver, save transmit power of user, and make an improvement to system throughput compared with the traditional method which does not consider MAI to some extent.

Adaptive beam-forming antenna (ABA) has been adopted as one of the key techniques of the TDD system. Because of the spatial variation of UE and imprecise digital signal processing (DSP) in the practical system, the system performance is deteriorated by the estimation errors of the direction of arrival (DOA). We evaluate the LTE TDD downlink performance of ABA without DOA estimation errors. Meanwhile, the impact of imperfect ABA on the system performance for the FTP-specific service as well as mixed services including FTP and video is investigated. Also a simple mixed services scheduler with adaptive emergency threshold (AET) is proposed over LTE TDD system to improve the system throughput while guaranteeing the quality of service (QoS) of video service.

Orthogonal frequency division multiple access (OFDMA), which is known as one of the most promising physical techniques for advanced wireless communications system, generally has a guard interval (GI) to mitigate the inter-symbol interference (ISI) and inter-carrier interference (ICI) caused by delay paths. However, the intra-cell interference (ISI and ICI) may be severe, when GI becomes insufficient in an execrable radio environment. Under fractional load (FL), resource allocator doesn't have to occupy entire system bandwidth due to the lack of traffic in the network, that is, the flexibility of allocator is increased for resource management. Instead of the complicated interference reduction algorithms at the receiver, in this paper we proposed a novel adaptive resource allocation (ARA) scheme, which utilizes the flexibility of radio resource management under FL, to efficiently avoid intra-cell interference in insufficient GI scenario. System-level simulations show that, the proposed ARA improves system performance than the conventional resource allocation (CRA) under FL, whereas the performance of proposed ARA approaches the CRA under the full load.

As a mainstream technology in the B3G TDD systems, TD-HSDPA (LCR TDD HSDPA, High-Speed Downlink Packet Access of Low Chip Rate TDD) adopts some key techniques to achieve high system throughput and high peak data rate, such as AMC (Adaptive Modulation and Coding) and HARQ (Hybrid Automatic Repeat reQuest). However, feedback delay is inevitable in practical TD-HSDPA systems due to the conventional scheduling mechanism, and its impact on wireless scheduling is critical. In this paper, the joint system performances of the key techniques are studied; furthermore, a new efficient scheduling algorithm is proposed. Compared to the conventional scheduling, the delay-restricted scheduling can reduce the feedback delay to improve the system capacity remarkably. Detailed network-performance results are shown from a comprehensive dynamic system-level simulator, and these presented results are obtained for a macrocellular scenario with best effort packet data services to reflect the expected improvement from introducing delay-restricted scheduling in a multi-carrier TD-HSDPA system.

In this paper, a novel effective low-complexity single carrier frequency-domain equalization SC-FDE scheme in multiple-input multiple-output (MIMO) system is introduced under frequency selective channel. It contains the advantages of SC-FDE and the complexity decreasing in MMSE algorithm by frequency-domain QR (FD-QR) decomposition. Also a rearrangement is implemented according to signal to noise (SNR). So performance gain can be obtained from rearrangement, demodulation after detection and equalization on specific antenna. Simulation result shows that the proposed scheme can obtain a significant performance gain to the conventional MMSE algorithm.

In this paper, Smart Antennas and HSDPA technology in TD-SCDMA will be discussed. It is demonstrated to be inefficient when they are combined together due to the shortage of channelization code resources. To use Smart Antennas more efficiently in TD-SCDMA HSDPA, one SDMA algorithm with combination of generic schedule algorithms is proposed. With SDMA, the system can allocate same channelization codes to different users who are spatially separable well enough. To reduce the complexity, the algorithm is simplified based on some reasonable scenario. The performances of the simplified algorithm with 3 different common schedule algorithms are evaluated through a system simulator, and the parameters affecting the algorithm are studied. In the end, the simulation results show that the system throughputs can be increased with optimized parameters.

This paper researches key techniques and the scheduling strategies in multi-carrier LCR TDD (low chip rate time division duplex) systems. We propose a new PF (proportional fairness) scheduling algorithm based on rate-prediction, and different scheduling strategies combined with SA (smart antenna) or JD (joint detection), which improve the throughput of packet traffics are simulated in dynamic system layer simulation. As some curves are obtained, we can get the dependency relationship between different kinds of technologies.