arXiv subjects

Xiao-Li Li

Publications and source records attributed to Xiao-Li Li.

15 recordsLinked to original sources

arXiv2021

Global Context Enhanced Graph Neural Networks for Session-based Recommendation

Session-based recommendation (SBR) is a challenging task, which aims at recommending items based on anonymous behavior sequences. Almost all the existing solutions for SBR model user preference only based on the current session without exploiting the other sessions, which may contain both relevant and irrelevant item-transitions to the current session. This paper proposes a novel approach, called Global Context Enhanced Graph Neural Networks (GCE-GNN) to exploit item transitions over all sessions in a more subtle manner for better inferring the user preference of the current session. Specifically, GCE-GNN learns two levels of item embeddings from session graph and global graph, respectively: (i) Session graph, which is to learn the session-level item embedding by modeling pairwise item-transitions within the current session; and (ii) Global graph, which is to learn the global-level item embedding by modeling pairwise item-transitions over all sessions. In GCE-GNN, we propose a novel global-level item representation learning layer, which employs a session-aware attention mechanism to recursively incorporate the neighbors' embeddings of each node on the global graph. We also design a session-level item representation learning layer, which employs a GNN on the session graph to learn session-level item embeddings within the current session. Moreover, GCE-GNN aggregates the learnt item representations in the two levels with a soft attention mechanism. Experiments on three benchmark datasets demonstrate that GCE-GNN outperforms the state-of-the-art methods consistently.

cs.IR↗

arXiv2021

Exploring Global Information for Session-based Recommendation

Session-based recommendation (SBR) is a challenging task, which aims at recommending items based on anonymous behavior sequences. Most existing SBR studies model the user preferences based only on the current session while neglecting the item-transition information from the other sessions, which suffer from the inability of modeling the complicated item-transition pattern. To address the limitations, we introduce global item-transition information to strength the modeling of the dynamic item-transition. For fully exploiting the global item-transition information, two ways of exploring global information for SBR are studied in this work. Specifically, we first propose a basic GNN-based framework (BGNN), which solely uses session-level item-transition information on session graph. Based on BGNN, we propose a novel approach, called Session-based Recommendation with Global Information (SRGI), which infers the user preferences via fully exploring global item-transitions over all sessions from two different perspectives: (i) Fusion-based Model (SRGI-FM), which recursively incorporates the neighbor embeddings of each node on global graph into the learning process of session level item representation; and (ii) Constrained-based Model (SRGI-CM), which treats the global-level item-transition information as a constraint to ensure the learned item embeddings are consistent with the global item-transition. Extensive experiments conducted on three popular benchmark datasets demonstrate that both SRGI-FM and SRGI-CM outperform the state-of-the-art methods consistently.

cs.IR↗

arXiv2021

Exploiting Group-level Behavior Pattern forSession-based Recommendation

Session-based recommendation (SBR) is a challenging task, which aims to predict users' future interests based on anonymous behavior sequences. Existing methods leverage powerful representation learning approaches to encode sessions into a low-dimensional space. However, despite such achievements, all the existing studies focus on the instance-level session learning, while neglecting the group-level users' preference, which is significant to model the users' behavior. To this end, we propose a novel Repeat-aware Neural Mechanism for Session-based Recommendation (RNMSR). In RNMSR, we propose to learn the user preference from both instance-level and group-level, respectively: (i) instance-level, which employs GNNs on a similarity-based item-pairwise session graph to capture the users' preference in instance-level. (ii) group-level, which converts sessions into group-level behavior patterns to model the group-level users' preference. In RNMSR, we combine instance-level user preference and group-level user preference to model the repeat consumption of users, \ie whether users take repeated consumption and which items are preferred by users. Extensive experiments are conducted on three real-world datasets, \ie Diginetica, Yoochoose, and Nowplaying, demonstrating that the proposed method consistently achieves state-of-the-art performance in all the tests.

cs.IR↗

arXiv2020

Recent Advances in Network-based Methods for Disease Gene Prediction

Disease-gene association through Genome-wide association study (GWAS) is an arduous task for researchers. Investigating single nucleotide polymorphisms (SNPs) that correlate with specific diseases needs statistical analysis of associations. Considering the huge number of possible mutations, in addition to its high cost, another important drawback of GWAS analysis is the large number of false-positives. Thus, researchers search for more evidence to cross-check their results through different sources. To provide the researchers with alternative low-cost disease-gene association evidence, computational approaches come into play. Since molecular networks are able to capture complex interplay among molecules in diseases, they become one of the most extensively used data for disease-gene association prediction. In this survey, we aim to provide a comprehensive and an up-to-date review of network-based methods for disease gene prediction. We also conduct an empirical analysis on 14 state-of-the-art methods. To summarize, we first elucidate the task definition for disease gene prediction. Secondly, we categorize existing network-based efforts into network diffusion methods, traditional machine learning methods with handcrafted graph features and graph representation learning methods. Thirdly, an empirical analysis is conducted to evaluate the performance of the selected methods across seven diseases. We also provide distinguishing findings about the discussed methods based on our empirical analysis. Finally, we highlight potential research directions for future studies on disease gene prediction.

q-bio.QM↗

arXiv2020

Understanding angle-resolved polarized Raman scattering from black phosphorus at normal and oblique laser incidences

The selection rule for angle-resolved polarized Raman (ARPR) intensity of phonons from standard group-theoretical method in isotropic materials would break down in anisotropic layered materials (ALMs) due to birefringence and linear dichroism effects. The two effects result in depth-dependent polarization and intensity of incident laser and scattered signal inside ALMs and thus make a challenge to predict ARPR intensity at any laser incidence direction. Herein, taking in-plane anisotropic black phosphorus as a prototype, we developed a so-called birefringence-linear-dichroism (BLD) model to quantitatively understand its ARPR intensity at both normal and oblique laser incidences by the same set of real Raman tensors for certain laser excitation. No fitting parameter is needed, once the birefringence and linear dichroism effects are considered with the complex refractive indexes. An approach was proposed to experimentally determine real Raman tensor and complex refractive indexes, respectively, from the relative Raman intensity along its principle axes and incident-angle resolved reflectivity by Fresnel$'$s law. The results suggest that the previously reported ARPR intensity of ultrathin ALM flakes deposited on a multilayered substrate at normal laser incidence can be also understood based on the BLD model by considering the depth-dependent polarization and intensity of incident laser and scattered Raman signal induced by both birefringence and linear dichroism effects within ALM flakes and the interference effects in the multilayered structures, which are dependent on the excitation wavelength, thickness of ALM flakes and dielectric layers of the substrate. This work can be generally applicable to any opaque anisotropic crystals, offering a promising route to predict and manipulate the polarized behaviors of related phonons.

cond-mat.mtrl-sci↗

arXiv2018

SL$^2$MF: Predicting Synthetic Lethality in Human Cancers via Logistic Matrix Factorization

Synthetic lethality (SL) is a promising concept for novel discovery of anti-cancer drug targets. However, wet-lab experiments for detecting SLs are faced with various challenges, such as high cost, low consistency across platforms or cell lines. Therefore, computational prediction methods are needed to address these issues. This paper proposes a novel SL prediction method, named SL2MF, which employs logistic matrix factorization to learn latent representations of genes from the observed SL data. The probability that two genes are likely to form SL is modeled by the linear combination of gene latent vectors. As known SL pairs are more trustworthy than unknown pairs, we design importance weighting schemes to assign higher importance weights for known SL pairs and lower importance weights for unknown pairs in SL2MF. Moreover, we also incorporate biological knowledge about genes from protein-protein interaction (PPI) data and Gene Ontology (GO). In particular, we calculate the similarity between genes based on their GO annotations and topological properties in the PPI network. Extensive experiments on the SL interaction data from SynLethDB database have been conducted to demonstrate the effectiveness of SL2MF.

cs.LG↗

arXiv2018

Adversarial adaptive 1-D convolutional neural networks for bearing fault diagnosis under varying working condition

Traditional intelligent fault diagnosis of rolling bearings work well only under a common assumption that the labeled training data (source domain) and unlabeled testing data (target domain) are drawn from the same distribution. However, in many real-world applications, this assumption does not hold, especially when the working condition varies. In this paper, a new adversarial adaptive 1-D CNN called A2CNN is proposed to address this problem. A2CNN consists of four parts, namely, a source feature extractor, a target feature extractor, a label classifier and a domain discriminator. The layers between the source and target feature extractor are partially untied during the training stage to take both training efficiency and domain adaptation into consideration. Experiments show that A2CNN has strong fault-discriminative and domain-invariant capacity, and therefore can achieve high accuracy under different working conditions. We also visualize the learned features and the networks to explore the reasons behind the high performance of our proposed model.

eess.SP↗

arXiv2016

Raman spectroscopic characterization of stacking configuration and interlayer coupling of twisted multilayer graphene grown by chemical vapor deposition

Multilayer graphene (MLG) grown by chemical vapor deposition (CVD) is a promising material for electronic and optoelectronic devices. Understanding the stacking configuration and interlayer coupling of MLGs is technologically relevant and of importance for the device applications. Here, we reported a kind of twisted MLGs (tMLGs), which are formed by stacking one graphene monolayer on the top of AB-stacked MLG by rotating a certain angle between them. The twist angle of tMLGs are identified by the twist-related modes, R and R'. With increasing the total layer number of N, the observed interlayer shear modes in the tMLG flake always follow those of AB-stacked (N-1)LG, while the observed interlayer breathing modes always follow those of AB-stacked NLG, independent of its twist angle. The layer breathing coupling of the tMLGs is almost identical to that of mechanically-exfoliated tMLGs, which demonstrates the high quality of MLGs grown by CVD. This study provides an applicable approach to probe the stacking configuration and interlayer coupling of MLGs grown by CVD or related methods. This work also demonstrates the possibility to grow MLG flakes with a fixed stacking configuration, $e.g.$, t(1+$n$)LG, by the CVD method.

cond-mat.mtrl-sci↗

arXiv2016

Learning Optimal Social Dependency for Recommendation

Social recommender systems exploit users' social relationships to improve the recommendation accuracy. Intuitively, a user tends to trust different subsets of her social friends, regarding with different scenarios. Therefore, the main challenge of social recommendation is to exploit the optimal social dependency between users for a specific recommendation task. In this paper, we propose a novel recommendation method, named probabilistic relational matrix factorization (PRMF), which aims to learn the optimal social dependency between users to improve the recommendation accuracy, with or without users' social relationships. Specifically, in PRMF, the latent features of users are assumed to follow a matrix variate normal (MVN) distribution. The positive and negative dependency between users are modeled by the row precision matrix of the MVN distribution. Moreover, we have also proposed an efficient alternating algorithm to solve the optimization problem of PRMF. The experimental results on real datasets demonstrate that the proposed PRMF method outperforms state-of-the-art social recommendation approaches, in terms of root mean square error (RMSE) and mean absolute error (MAE).

cs.IR↗

arXiv2016

Determining layer number of two dimensional flakes of transition-metal dichalcogenides by the Raman intensity from substrate

Transition-metal dichalcogenide (TMD) semiconductors have been widely studied due to their distinctive electronic and optical properties. The property of TMD flakes is a function of its thickness, or layer number (N). How to determine N of ultrathin TMDs materials is of primary importance for fundamental study and practical applications. Raman mode intensity from substrates has been used to identify N of intrinsic and defective multilayer graphenes up to N=100. However, such analysis is not applicable for ultrathin TMD flakes due to the lack of a unified complex refractive index ($\tilde{n}$) from monolayer to bulk TMDs. Here, we discuss the N identification of TMD flakes on the SiO$_2$/Si substrate by the intensity ratio between the Si peak from 100-nm (or 89-nm) SiO$_2$/Si substrates underneath TMD flakes and that from bare SiO$_2$/Si substrates. We assume the real part of $\tilde{n}$ of TMD flakes as that of monolayer TMD and treat the imaginary part of $\tilde{n}$ as a fitting parameter to fit the experimental intensity ratio. An empirical $\tilde{n}$, namely, $\tilde{n}_{eff}$, of ultrathin MoS$_{2}$, WS$_{2}$ and WSe$_{2}$ flakes from monolayer to multilayer is obtained for typical laser excitations (2.54 eV, 2.34 eV, or 2.09 eV). The fitted $\tilde{n}_{eff}$ of MoS$_{2}$ has been used to identify N of MoS$_{2}$ flakes deposited on 302-nm SiO$_2$/Si substrate, which agrees well with that determined from their shear and layer-breathing modes. This technique by measuring Raman intensity from the substrate can be extended to identify N of ultrathin 2D flakes with N-dependent $\tilde{n}$ . For the application purpose, the intensity ratio excited by specific laser excitations has been provided for MoS$_{2}$, WS$_{2}$ and WSe$_{2}$ flakes and multilayer graphene flakes deposited on Si substrates covered by 80-110 nm or 280-310 nm SiO$_2$ layer.

cond-mat.mtrl-sci↗

arXiv2015

Substrate-free layer-number identification of two-dimensional materials: A case of Mo$_{0.5}$W$_{0.5}$S$_2$ alloy

Any of two or more two-dimensional (2D) materials with similar properties can be alloyed into a new layered material, namely, 2D alloy. Individual monolayer in 2D alloys are kept together by Van der Waals interactions. The property of multilayer alloys is a function of their layer number. Here, we studied the shear (C) and layer-breathing (LB) modes of Mo$_{0.5}$W$_{0.5}$S$_2$ alloy flakes and their link to the layer number of alloy flakes. The study reveals that the disorder effect is absent in the C and LB modes of 2D alloys, and the monatomic chain model can be used to estimate the frequencies of the C and LB modes. We demonstrated how to use the C and LB mode frequency to identify the layer number of alloy flakes deposited on different substrates. This technique is independent of the substrate, stoichiometry, monolayer thickness and complex refractive index of 2D materials, offering a robust and substrate-free approach for layer-number identification of 2D materials.

cond-mat.mtrl-sci↗

arXiv2015

Classification and its applications for drug-target interaction identification

Classification is one of the most popular and widely used supervised learning tasks, which categorizes objects into predefined classes based on known knowledge. Classification has been an important research topic in machine learning and data mining. Different classification methods have been proposed and applied to deal with various real-world problems. Unlike unsupervised learning such as clustering, a classifier is typically trained with labeled data before being used to make prediction, and usually achieves higher accuracy than unsupervised one. In this paper, we first define classification and then review several representative methods. After that, we study in details the application of classification to a critical problem in drug discovery, i.e., drug-target prediction, due to the challenges in predicting possible interactions between drugs and targets.

cs.LG↗

arXiv2015

Microbial community pattern detection in human body habitats via ensemble clustering framework

The human habitat is a host where microbial species evolve, function, and continue to evolve. Elucidating how microbial communities respond to human habitats is a fundamental and critical task, as establishing baselines of human microbiome is essential in understanding its role in human disease and health. However, current studies usually overlook a complex and interconnected landscape of human microbiome and limit the ability in particular body habitats with learning models of specific criterion. Therefore, these methods could not capture the real-world underlying microbial patterns effectively. To obtain a comprehensive view, we propose a novel ensemble clustering framework to mine the structure of microbial community pattern on large-scale metagenomic data. Particularly, we first build a microbial similarity network via integrating 1920 metagenomic samples from three body habitats of healthy adults. Then a novel symmetric Nonnegative Matrix Factorization (NMF) based ensemble model is proposed and applied onto the network to detect clustering pattern. Extensive experiments are conducted to evaluate the effectiveness of our model on deriving microbial community with respect to body habitat and host gender. From clustering results, we observed that body habitat exhibits a strong bound but non-unique microbial structural patterns. Meanwhile, human microbiome reveals different degree of structural variations over body habitat and host gender. In summary, our ensemble clustering framework could efficiently explore integrated clustering results to accurately identify microbial communities, and provide a comprehensive view for a set of microbial communities. Such trends depict an integrated biography of microbial communities, which offer a new insight towards uncovering pathogenic model of human microbiome.

q-bio.QM↗

arXiv2014

The Interlayer Shear Modes in Twisted Multi-layer Graphenes: Interlayer coupling, Davydov Splitting and Intensity Resonance

Graphene and other two-dimensional crystals can be combined to form various hybrids and heterostructures, creating materials on demand, in which the interlayer coupling at the interface leads to modified physical properties as compared to their constituents. Here, by measuring Raman spectra of shear modes, we probe the coupling at the interface between two artificially-stacked few-layer graphenes rotated with respect to each other. The strength of interlayer coupling between the two interface layers is found to be only 20% of that between Bernal-stacked layers. Nevertheless, this weak coupling manifests itself in a Davydov splitting of the shear mode frequencies in systems consisting of two equivalent graphene multilayers, and in the intensity enhancement of shear modes due to the optical resonance with several optically allowed electronic transitions between conduction and valence bands in the band structures. This study paves way for fundamental understanding into the interface coupling of two-dimensional hybrids and heterostructures.

cond-mat.mtrl-sci↗

arXiv2009

Magneto-optical properties of Co/ZnO multilayer films

Multilayer films of ZnO with Co were deposited on glass substrates then annealed in a vacuum. The magnetisation of the films increased with annealing but not the magnitude of the magneto-optical signals. The dielectric functions for the films were calculated using the MCD spectra. A Maxwell Garnett theory of a metallic Co/ZnO mixture is presented. The extent to which this explains the MCD spectra taken on the films is discussed.

cond-mat.mtrl-sci↗