Joint Matrix Research Articles

GP-HSI: Human-Scene Interaction with Geometric and Physical Constraints

With the rapid development of AR/VR technologies, achieving natural and seamless human-scene interactions has emerged as a critical challenge in computer vision. Existing methods suffer from low model placement accuracy and unnatural scene interactions. Therefore, we propose a framework called human-scene interaction with geometric and physical constraints (GP-HSI), which places a given pose of a 3D human model in an appropriate position within a 3D scene by establishing geometric and physical constraints, while ensuring interactive fidelity between the human and the scene. Specifically, first, we propose a pose-guided human contact semantic generation method, which generates human semantic labels by classifying the given human poses. Second, we propose a geometrically and semantically constrainted human model placement method, which determines the optimal position of the human model in the scene by constraining the geometric proximity and semantic consistency between models. Third, we propose an inverse kinematics based pose adjustment method, which finds the target human-scene interaction points by constructing a heterogeneous kinematic tree and solves the rotation matrix of human joints to obtain a physically plausible optimal human pose. At last, we develop an interactive system to visualize the generated human-scene interaction. The results of qualitative and quantitative experiments show that our approach is able to place human models at appropriate locations in the scene and generate plausible interactions.

DOME: Directional medical embedding vectors from Electronic Health Records.

The increasing availability of Electronic Health Record (EHR) systems has created enormous potential for translational research. Recent developments in representation learning techniques have led to effective large-scale representations of EHR concepts along with knowledge graphs that empower downstream EHR studies. However, most existing methods require training with patient-level data, limiting their abilities to expand the training with multi-institutional EHR data. On the other hand, scalable approaches that only require summary-level data do not incorporate temporal dependencies between concepts. We introduce a DirectiOnal Medical Embedding (DOME) algorithm to encode temporally directional relationships between medical concepts, using summary-level EHR data. Specifically, DOME first aggregates patient-level EHR data into an asymmetric co-occurrence matrix. Then it computes two Positive Pointwise Mutual Information (PPMI) matrices to correspondingly encode the pairwise prior and posterior dependencies between medical concepts. Following that, a joint matrix factorization is performed on the two PPMI matrices, which results in three vectors for each concept: a semantic embedding and two directional context embeddings. They collectively provide a comprehensive depiction of the temporal relationship between EHR concepts. We highlight the advantages and translational potential of DOME through three sets of validation studies. First, DOME consistently improves existing direction-agnostic embedding vectors for disease risk prediction in several diseases, for example achieving a relative gain of 5.5% in the area under the receiver operating characteristic (AUROC) for lung cancer. Second, DOME excels in directional drug-disease relationship inference by successfully differentiating between drug side effects and indications, correspondingly achieving relative AUROC gain over the state-of-the-art methods by 10.8% and 6.6%. Finally, DOME effectively constructs directional knowledge graphs, which distinguish disease risk factors from comorbidities, thereby revealing disease progression trajectories. The source codes are provided at https://github.com/celehs/Directional-EHR-embedding.

Sonic P-wave attenuation based on fluid pressure differential between joint and matrix

When a P wave compresses in situ rocks saturated with fluid, there is a fluid pressure differential between compliant joints and stiff matrix. The pressure differential automatically drives a mesoscale squirt between joint wall and matrix interior. This mechanism is likely to dominate sonic P-wave attenuation in the field. A new model of P-wave attenuation is established based on mesoscale squirt, in which the first porosity is represented by matrix pores, and the second porosity is represented by joints. The sonic log of a sandstone reservoir in Saudi Arabia is used to illustrate the model. Estimated phase velocity and measured quality factor of sonic P waves are accurately simulated. The results indicate that for the sandstone reservoir, the relative second porosity is 5%, the interjoint distance is 0.1 m, the squirt permeability at very low frequencies is approximately half of the Darcy permeability of the matrix, and the joint gap is 800 μm (quadruple grain diameter of 200 μm). Surprisingly, the model of joint-matrix squirt also is capable of reproducing seismic P-wave attenuation measured in the shallow crust of southern California. Therefore, mesoscale squirt across the joint wall may be the underlying mechanism of attenuation for sonic and seismic P waves shallower than 10 km.

Ni mesh-reinforced ultrasonic-assisted Cu/Sn58Bi/Cu joint performance: Experiments and first-principles calculations

The Ni mesh was incorporated into the Cu/Sn58Bi/Cu bonding as a reinforcing skeleton to achieve an enhancement effect analogous to steel reinforcement in concrete. Ultrasonic-assisted soldering (UAS) improved the metallurgical bond among the solder, Ni mesh, and substrate. It facilitated the formation of (Cu, Ni)6Sn5 intermetallic compounds (IMCs) layers, increasing the joint strength. Observations indicated that ultrasonic treatment effectively refined the (Cu, Ni)6Sn5 grains and induced a uniform preferred orientation of β-Sn and Bi grains in the joint matrix adjacent to the Ni mesh. The shear strength of the joint reached 72.23 MPa when the ultrasonic application was sustained for 15 s, achieving the fabrication of a high-strength point with low energy consumption. First-principles calculations have confirmed that changes in the Ni content within (Cu, Ni)6Sn5 IMCs improved the stability of the crystal structure. Furthermore, the variations in content could potentially improve the mechanical and electrical properties of the (Cu, Ni)6Sn5. Enhancements in ultrasonic efficiency and the reinforcement of IMC structures offer new avenues for research in green and high-performance electronic packaging material joining technologies.

Cross-domain recommender system with embedding- and mapping-based knowledge correlation

A knowledge transfer-based cross-domain recommender system is currently a research hotspot. Existing research has reached a high level of maturity in mining potential knowledge and establishing transfer mechanisms. However, most of them ignore the impact of the dissimilarity of potential knowledge on the transfer performance. Herein, a cross-domain recommender system based on knowledge correlation-induced the embedding and mapping approach is proposed, denoted by KCEM-CDRS. First, we propose a knowledge correlation measure, which captures the consistency of knowledge between the target and source domains to build the bridge for knowledge transfer. Second, we construct a joint matrix triple factorization model to solve the data sparsity in the target domain while introducing graph regularization to solve the problem of negative knowledge transfer. Results of extensive experiments on real Amazon metadata indicate that compared with three existing cross-domain recommendation methods, KCEM-CDRS shows performance improvements of 0.05–9.55 % and 0.02–2.63 % on mean absolute error and root mean square error, respectively. Additionally, the results of the ablation experiments indicate that consideration of the knowledge correlation between domains is beneficial for knowledge transfer when the density of the source domain is rich.

A Curated Compendium of Transcriptomic Data for the Exploration of Neocortical Development.

Vast quantities of multi-omic data have been produced to characterize the development and diversity of cell types in the cerebral cortex of humans and other mammals. To more fully harness the collective discovery potential of these data, we have assembled gene-level transcriptomic data from 188 published studies of neocortical development, including the transcriptomes of ~30 million single-cells, extensive spatial transcriptomic experiments and RNA sequencing of sorted cells and bulk tissues: nemoanalytics.org/landing/neocortex. Applying joint matrix decomposition (SJD) to mouse, macaque and human data in this collection, we defined transcriptome dynamics that are conserved across mammalian neurogenesis and which elucidate the evolution of outer, or basal, radial glial cells. Decomposition of adult human neocortical data identified layer-specific signatures in mature neurons and, in combination with transfer learning methods in NeMO Analytics, enabled the charting of their early developmental emergence and protracted maturation across years of postnatal life. Interrogation of data from cerebral organoids demonstrated that while broad molecular elements of in vivo development are recapitulated in vitro, many layer-specific transcriptomic programs in neuronal maturation are absent. We invite computational biologists and cell biologists without coding expertise to use NeMO Analytics in their research and to fuel it with emerging data (carlocolantuoni.org).

Clinical-Genomic Analysis of 1261 Patients with Ehlers-Danlos Syndrome Outlines an Articulo-Autonomic Gene Network (Entome).

Systematic evaluation of 80 history and 40 history findings diagnosed 1261 patients with Ehlers-Danlos syndrome (EDS) by direct or online interaction, and 60 key findings were selected for their relation to clinical mechanisms and/or management. Genomic testing results in 566 of these patients supported EDS relevance by their differences from those in 82 developmental disability patients and by their association with general rather than type-specific EDS findings. The 437 nuclear and 79 mitochondrial DNA changes included 71 impacting joint matrix (49 COL5), 39 bone (30 COL1/2/9/11), 22 vessel (12 COL3/8VWF), 43 vessel-heart (17FBN1/11TGFB/BR), 59 muscle (28 COL6/12), 56 neural (16 SCN9A/10A/11A), and 74 autonomic (13 POLG/25porphyria related). These genes were distributed over all chromosomes but the Y, a network analogized to an 'entome' where DNA change disrupts truncal mechanisms (skin constraint, neuromuscular support, joint vessel flexibility) and produces a mirroring cascade of articular and autonomic symptoms. The implied sequences of genes from nodal proteins to hypermobility to branching tissue laxity or dysautonomia symptoms would be ideal for large language/artificial intelligence analyses.

Core Collection Formation in Guatemalan Wild Avocado Germplasm with Phenotypic and SSR Data

Guatemala’s wild avocado germplasm holds vital genetic value, but lacking conservation strategies imperils it. Studying its diversity is pivotal for conservation and breeding. The study aimed to comprehensively assess the wild avocado germplasm in Guatemala by combining phenotypic and genotypic data and to create a core collection for conservation and future breeding programs. A total of 189 mature avocado trees were sampled across Guatemala’s northern, southern, and western regions. Morphological characteristics were documented, and genetic diversity was assessed using 12 SSR loci. The investigated germplasm revealed three distinct genetic clusters, exhibiting an average gene diversity of 0.796 and a 7.74% molecular variation among them. The samples showed various morphological characteristics that indicate the presence of three avocado races in Guatemala. The weak correlation between phenotypic and genotypic distances highlighted their independence and complementary nature. The joint matrix effectively integrated and captured genotypic and phenotypic data for comprehensive genetic diversity analysis. A core collection comprising 20% of total accessions that captured maximum genetic diversity was formed. This study exposed wild Guatemalan avocados’ genetic diversity, morphological traits, and conservation significance. Integrated data capture via clustering validates holistic genetic insight for conservation and breeding strategies.

Microstructure and properties of a YG18/40Cr joint vacuum–brazed by Cu–Sn–Ti filler metal

In this study, dissimilar YG18/40Cr joints were prepared using a Cu–Sn–Ti filled metal by a vacuum brazing process, and the microstructural evolution and bonding mechanism of the YG18/Cu–Sn–Ti/40Cr joints were studied. By adjusting the brazing parameters, the formation process of interface structure, the joint structure and matrix internal structure, and the changes in the joint microstructure and properties under different parameters were systematically analyzed. A TiC reaction layer was formed at the interface on both sides of the joint. The brazing performance can be improved by the formation of a reasonable size reaction layer. A TiC–WC cemented carbide was formed near the middle layer inside YG18, and this tenon–and–tenon structure improved the bonding between the YG18 side and the filler alloy. The connector structure was YG18/TiC/CoTi + Co2SnTi/Cu [s, s]/eutectic structure (Cu [s, s] + Cu Sn3Ti5)/CuSn3Ti5/TiC/40Cr, and its eutectic structure changed as the parameters changed. The maximum average shear strength of the YG18/40Cr joint brazed at 940 °C for 5 min reached 216 MPa, which was ≈25% higher than the joint shear strength of the pure copper used in a previous study. The fracture morphology of the joint showed a large number of fossae and tear edges, indicating ductile fracture.

Accuracy Analysis and Experimental Study of a Six-Dimensional Force Sensor with Circular Flexible Spherical Joints.

This paper proposed a circular flexible spherical joint to reduce the processing difficulty and manufacturing cost, and design a six-dimensional force sensor based on it. This paper analyzes the influence of the structural parameters of the flexible spherical joints on the accuracy of the six-dimensional force sensor by comparing the force mapping matrix of flexible spherical and ideal spherical joints. As the force mapping matrix is related to the stiffness matrix, the stiffness matrix of a six-dimensional force sensor based on a circular flexible spherical joint and the force mapping matrix of the generalized external force acting on the sensor was derived. Finally, a set of optimal parameters was selected to create a prototype and was verified through finite elements and experiments. The results show that the six-dimensional force sensor with a circular flexible spherical joint has good accuracy and provides some guidance for the design and analysis of the sensor.

Kernel structure of the combined English, Dutch, and Polish personality type-nouns, with a critical test against a type-noun based structure in Swahili

We compared three trait-structures based on type-nouns, to find their common kernel structure. We used ratings from 607 participants on 372 English type-nouns, 800 participants on 571 Dutch type-nouns, and 1,325 participants on 454 Polish typenouns. PCA based factor structures were compared using congruence coefficients. SCA was applied on a joint matrix of type-nouns with ratings from a total of 2,737 participants on 331 type-nouns shared by all three languages. The resulting structure reflected versions of the Big Five, yet narrowed to their oratory role. Finally, the results were compared with a type-nouns based structure in Swahili.

MC 2 : Unsupervised Multiple Social Network Alignment

Social network alignment, identifying social accounts of the same individual across different social networks, shows fundamental importance in a wide spectrum of applications, such as link prediction and information diffusion. Individuals more often than not join in multiple social networks, and it is in fact much too expensive or even impossible to acquiring supervision for guiding the alignment. To the best of our knowledge, few method in the literature can align multiple social networks without supervision. In this article, we propose to study the problem of unsupervised multiple social network alignment. To address this problem, we propose a novel unsupervised model of joint Matrix factorization with a diagonal Cone under orthogonal Constraint, referred to as MC 2 . Its core idea is to embed and align multiple social networks in the common subspace via an unsupervised approach. Specifically, in MC 2 model, we first design a matrix optimization to infer the common subspace from different social networks. To address the nonconvex optimization, we then design an efficient alternating algorithm by leveraging its inherent functional property. Through extensive experiments on real-world datasets, we demonstrate that the proposed MC 2 model significantly outperforms the state-of-the-art methods.

Interspecies comparison of temporomandibular joint condylar cartilage extracellular matrix from macro to microscopy

Interspecies comparisons of the extracellular matrix of temporomandibular joint (TMJ) condylar cartilage are necessary to elucidate the mechanisms underlying its superior mechanical properties, to guide the construction of animal models of TMJ-related diseases, and to establish standards for the engineering of TMJ condylar cartilage. Here we characterize and compare TMJ condylar cartilage from six different species from a materials science perspective, including structure, composition and mechanical properties from the macroscopic to the microscopic level. The gross morphology showed obvious interspecies differences in size and shape, which may be related to the different joint motion patterns. Although the condylar cartilage of all species can be divided histologically into a superficial fibrous layer and a deep hyaline layer, there are significant interspecies differences in the microstructure of the fibrils in the two layers, mainly in the diameter of the fibrils. Compositionally, there were no significant differences in collagen composition between species, but the content of glycosaminoglycans (GAGs) decreased progressively with increasing body size, with the same results obtained by Safranin O staining and biochemical analysis. Mechanically, the elastic modulus of mouse condylar cartilage was significantly higher than that of the other species and tended to decrease with increasing body size. This study shows that the TMJ condylar cartilage of different species has its own specific structure-composition-mechanics matching characteristics for their unique masticatory stress dissipation, and differences in fibril diameter and GAGs content may be the two ultimate factors influencing the differences in cartilage mechanical properties between species, while the condylar cartilage of pigs is most similar to that of humans, suggesting that pigs may be a suitable animal model for TMJ studies.

Research on Digital Modeling Method of Triaxial Vibration System

To meet the demand of providing basis for improving triaxial vibration system, it is necessary to extract the overall Frequency Response Function (FRF) matrix of triaxial vibration system from its digital model. The mechanical parts of triaxial vibration system is connected tightly which satisfy rigid connection assumption. The Jetmudsen FRF method is used to construct the overall frequency response function matrix of triaxial vibration system. The electrodynamic shaker (referred to as the shaker) is an important component of the three-axis vibration system, and its function is to provide the test bench with an excitation force in a certain direction. Therefore, it is necessary to define a large number of internal nodes on the shaker. The FRF matrix of shaker (including origin FRF matrix of internal nodes and joints, cross-point FRF matrix of internal nodes and joints) should be obtain to carry out the construction of overall FRF matrix of triaxial vibration system. It is difficult because the size of FRF matrix of shaker is too large. So a way to simplify the distributed loads exerting on the structure into multi-point concentrated loads is proposed, it can be used to obtain the simplified digital model of the electrodynamic shaker. Based study above, the overall frequency response function matrix of the triaxial vibration system is obtained, and the digital modeling of the triaxial vibration system is completed.

A joint matrix factorization and clustering scheme for irregular time series data

Key Performance Indicator (KPI) clustering plays an important role in Artificial Intelligence for IT Operations (AIOps) when the number of KPIs is large. This approach can effectively reduce the overhead by dividing KPIs into several classes, then applying the same anomaly detection or prediction model to all KPIs in a class. However, KPI sampling strategies vary depending on the environment in question, which leads to the production of irregular KPIs. Few existing works have considered the clustering of KPIs with irregular sampling. Matrix factorization (MF) is widely applied in low-rank data recovery and can be used to align and fill irregular KPIs. However, the clustering performance after recovering and filling by MF remains unknown. These two problems interact with each other and should therefore be solved together. Accordingly, we formulate the joint MF and clustering problem for irregular KPIs and design an iterative clustering scheme based on MF. This iterative clustering scheme comprises alignment and pre-filling, the loop of clustering, and subclass filling by MF, and can work with two pre-filling methods. Extensive experiments on two real-world datasets show that the iterative clustering scheme can obtain higher normalized mutual information (NMI) than non-iterative clustering, while also consuming less computational time than Dynamic Time Warping (DTW). The two kinds of pre-filling methods each have their advantages on different datasets.

A Study on the Recommendation Method of Intelligent Media Learning Resources in the Foreign Communication and Teaching of International Communication of Chinese Central Plains Culture

Today, with the development of intelligent media, the foreign communication and teaching activities of the Chinese central plains culture should actively seek experiences that can be learned from, establish a multi-channel foreign communication mode, and then promote the Chinese central plains culture to go out of the country and into the world better. The study improves the collaborative filtering recommendation algorithm and the joint matrix decomposition algorithm based on the theory of migration learning, aiming to improve the learning to optimize the resource recommendation system by calculating the user similarity and establishing the user preference-resource feature matrix. The experimental results show that the average absolute error and root mean square error of the improved algorithms are lower than those of other algorithms, proving that the optimized algorithms improve the accuracy and efficiency of resource recommendation in the foreign communication and teaching activities of the Chinese central plains culture while operating stably and with wide applicability on the recommendation system.

Multivariate Transfer Passenger Flow Forecasting with Data Imputation by Joint Deep Learning and Matrix Factorization

Accurate forecasting of the future transfer passenger flow from historical data is essential for helping travelers to adjust their trips, optimal resource allocation and alleviating traffic congestion. However, current studies have mainly emphasized predicting traffic parameters for a single type of transport, while lacking research into transfer passenger flow influenced by multiple factors across different transport modes. Additionally, efficient traffic prediction relies on high-quality traffic data, yet data loss issues are inevitable but often ignored. To fill these gaps, we present for the first time a reliable joint long short-term memory with matrix factorization deep learning model (i.e., Joint-IF) for accurate imputation and forecasting of transfer passenger flow between metro and bus. This hybrid Joint-IF model uses a repair-before-prediction strategy to deliver the final high-quality outputs. In particular, we simulate a variety of missing combinations under the natural conditions and apply a low-rank matrix factorization to infer those lost values. In addition, we investigate the effects of crucial parameters and spatiotemporal features on transfer flow prediction. To validate the effectiveness of Joint-IF, a large series of experiments are carried out for models’ comparison and validation on the real-world transfer passenger flow dataset of the Shenzhen public transport system, and the results show that the proposed Joint-IF performs better for both imputation and forecasting of transfer passenger flow relative to the baseline models in terms of accuracy and stability.

Effect of gamma irradiation on microstructural evolution and mechanical properties of Sn3Ag0.5Cu solder joints

Sn3Ag0.5Cu (SAC305) Ball Grid Array (BGA) solder joints were exposed to 60Co source gamma irradiation at a dose rate of 1 Gy(Si)/s from 0 to 200 h in air. The effects of gamma irradiation on the microstructure evolution and mechanical properties of the solder joints were investigated. β-Sn interacted with O2 in the air and was oxidized to SnO2 on the solder joint surface after gamma irradiation. The oxidation initiated at the corners of the solder joint/pad interface, extended to the entire solder joint/pad interface, and finally occurred at the solder joint matrix. Microcracks usually grow around SnO2/β-Sn interfaces. The interaction of gamma rays with β-Sn and SnO2 is the primary reason for solder joint oxidation. Gamma irradiation generated atomic vacancies, breaking bonds, and dangling bonds in β-Sn and SnO2 crystals through the Compton Effect, which promoted the physisorption and chemisorption of O2 on their surfaces, as well as the oxidation of β-Sn and growth of SnO2. The interaction between the gamma irradiation and solder joints can be divided into two stages. SnO2 was nucleated at lattice defects induced by the interaction between gamma rays and β-Sn crystals in the first stage. The growth of the SnO2 nuclei resulting from a large number of crystal defects in SnO2 and the potential difference between SnO2 and the β-Sn surface induced by the Compton Effect occurred in the second stage. According to the differential scanning calorimetry (DSC) analysis, the melting temperature of the SAC305 solder alloy decreased with increasing irradiation time. According to the nanoindentation test results, with an increase in irradiation time, the microhardness of the solder joints first increased, then decreased, and finally increased again.

Cooperative driver pathways discovery by multiplex network embedding.

Cooperative driver pathways discovery helps researchers to study the pathogenesis of cancer. However, most discovery methods mainly focus on genomics data, and neglect the known pathway information and other related multi-omics data; thus they cannot faithfully decipher the carcinogenic process. We propose CDPMiner (Cooperative Driver Pathways Miner) to discover cooperative driver pathways by multiplex network embedding, which can jointly model relational and attribute information of multi-type molecules. CDPMiner first uses the pathway topology to quantify the weight of genes in different pathways, and optimizes the relations between genes and pathways. Then it constructs an attributed multiplex network consisting of micro RNAs, long noncoding RNAs, genes and pathways, embeds the network through deep joint matrix factorization to mine more essential information for pathway-level analysis and reconstructs the pathway interaction network. Finally, CDPMiner leverages the reconstructed network and mutation data to define the driver weight between pathways to discover cooperative driver pathways. Experimental results on Breast invasive carcinoma and Stomach adenocarcinoma datasets show that CDPMiner can effectively fuse multi-omics data to discover more driver pathways, which indeed cooperatively trigger cancers and are valuable for carcinogenesis analysis. Ablation study justifies CDPMiner for a more comprehensive analysis of cancer by fusing multi-omics data.

Cross-domain recommendation with Multi-Auxiliary Domains via Consistent and Selective Cluster-Level Knowledge Transfer

Recommender systems are software tools and techniques designed to provide users with useful items suggestions. For most existing recommender systems, data sparsity is a challenging and common problem. The cross-domain recommender systems can mitigate the sparsity problem of data in a target domain by introducing auxiliary domain with relatively dense data. However, there are three challenging issues in the existing cross-domain recommender systems: (1) Most cross-domain recommender systems are designed for a single auxiliary domain, which may lead to insufficient knowledge transfer. However, in practice, there may be multiple auxiliary domains related to the target domain that are available; (2) Due to the distribution divergence between the auxiliary and target domains, inconsistent knowledge may be transferred to the target domain; (3) Because of the diversity among domains, they do not necessarily share exactly the same rating pattern. Moreover, not all cluster-level knowledge of auxiliary domains is beneficial to the target domain, and some cluster-level knowledge may be harmful to the target domain. To deal with the above three issues, we propose a Cross-Domain Recommendation with Multi-Auxiliary Domains via Consistent and Selective Cluster-Level Knowledge Transfer, called MD-CSKT. First, rating matrices from multiple auxiliary domains are decomposed jointly with a single target rating matrix to transfer richer knowledge from multiple auxiliary domains. Secondly, the Maximum Mean Discrepancy (MMD) regularization constraint is incorporated into the joint decomposition to reduce the distribution discrepancy between each auxiliary domain and the target domain, so as to achieve consistent cluster-level knowledge transfer. Thirdly, selective cluster-level knowledge transfer is realized by adaptively selecting of cluster-level rating knowledge from multiple auxiliary domains in the joint matrix factorization. The experimental results on six real-world datasets in three categories demonstrate that the presented MD-CSKT method substantially outperforms nine comparison methods and achieves the improvement of recommendation accuracy in the target domain. Compared with the state-of-the-art comparison methods, the overall performance of the MD-CSKT method is improved by more than 25%.