Search

Tensor-oriented multi-view subspace clustering has achieved significant strides in assessing high-order correlations and improving clustering analysis of multi-view data. Nevertheless, most of existing investigations are typically hampered by the two flaws. First, self-representation based tensor subspace learning usually induces high time and space complexity, and is limited in perceiving nonlinear local structure in the embedding space. Second, the tensor singular value decomposition (t-SVD) model redistributes each singular value equally without considering the diverse importance among them. To well cope with the issues, we propose a hyper-Laplacian regularized concept factorization (HLRCF) in low-rank tensor space for multi-view clustering. Specifically, we adopt the concept factorization to explore the latent cluster-wise representation of each view. Further, the hypergraph Laplacian regularization endows the model with the capability of extracting the nonlinear local structures in the latent space. Considering that different tensor singular values associate structural information with unequal importance, we develop a self-weighted tensor Schatten p-norm to constrain the tensor comprised of all cluster-wise representations. Notably, the tensor with smaller size greatly decreases the time and space complexity in the low-rank optimization. Finally, experimental results on eight benchmark datasets exhibit that HLRCF outperforms other multi-view methods, showingcasing its superior performance.

Multi-view clustering aims to discover common patterns from multi-source data, whose generality is remarkable. Compared with traditional methods, deep learning methods are data-driven and have a larger search space for solutions, which may find a better solution to the problem. In addition, more considerations can be introduced by loss functions, so deep models are highly reusable. However, compared with deep learning methods, traditional methods have better interpretability, whose optimization is relatively stable. In this paper, we propose a multi-view spectral clustering model, combining the advantages of traditional methods and deep learning methods. Specifically, we start with the objective function of traditional spectral clustering, perform multi-view extension, and then obtain the traditional optimization process. By partially parameterizing this process, we further design corresponding differentiable modules, and finally construct a complete network structure. The model is interpretable and extensible to a certain extent. Experiments show that the model performs better than other multi-view clustering algorithms, and its semi-supervised classification extension also has excellent performance compared to other algorithms. Further experiments also show the stability and fewer iterations of the model training.

Proteomics is a powerful tool to understand molecular connections between meat proteins and quality traits. In order to research the total proteome difference and molecular mechanisms related to quality traits in different parts of yak meat. This study was designed to investigate the differences in the proteomes of yak different muscles. In this paper, Tandem Mass Tags (TMT) coupled Liquid chromatography-tandem mass spectrometry (LC-MS/MS) to study the variations of proteome in different parts of yak after it is slaughtered. In total, 88 differentially expressed proteins (DEPs) were identified among striplon (WJR), slivsid (HGT) and chuck (JR). The quality analysis of DEPs showed heat shock proteins (HSPs) and structural proteins could be used as tenderness marker proteins for different parts of yak meat. Myosin and troponin-T may be flavour marker protein in yak meat. The bioinformatics analysis revealed that DEPs are involved in glycolysis, protein structure and phosphorylation. PPI analysis revealed thatmyosin, HSPs and metabolic enzymes may have the potential to be biological markers. This study highlights that the DEPs were responsible for meat quality different.

In order to study the tenderness mechanism of difference in different parts of Yak meat, the differentially expressed proteins (DEPs) and its correlation with quality of Yak meat were studied by TMT (Tandem Mass Tags, TMT) quantitative proteomic method. The results showed that 25, 40 and 34 DEPs were identified in JR/HGT, WJR/HGT and WJR/JR groups, respectively. Correlation analysis between DEPs and quality traits of yak meat showed that there were 12 DEPs with significant correlation were correlated closely with quality traits.The relative contents of the 11 DEPs were significantly different in different parts (p < 0.05).There are 8, 9 and 6 key DEPs in JR/HGT, WJR/HGT and WJR/JR groups, respectively.The expression of these DEPs were down-regulated and up-regulated in different parts, and was closely related to the cell growth and death, carbohydrate metabolism, amino acid metabolism, protein binding, motility activity and unfolded protein binding.The significant difference of the expression of these key proteins in different parts is the internal reason of the difference of quality traits among WJR, JR and HGT.This provide a reference for the study on the mechanism of difference in Yak meat.

This study aimed to evaluate the environmental impacts (i.e. global warming potential (GWP) and resource depletion (RD)) of the bio-based levoglucosan production process through fast pyrolysis of cotton straw via life cycle assessment (LCA). An LCA model consisting of feedstock transportation, biomass pretreatment, fast pyrolysis, bio-oil transportation, bio-oil recovery and levoglucosan extraction was developed. Results indicated that GWP and RD of bio-based levoglucosan production were approximately 2 and 32.5 times less than that of the petroleum-based counterpart. Sensitivity analysis showed that the GWP and RD of levoglucosan production were highly sensitive to plant size, hydrochloric acid usage, cooling energy, levoglucosan yield and bio-oil yield. The results of this research could provide a framework for robust decision making at an industrial level, which is useful for the commercial-scale production of levoglucosan.

A rapid method for the quantification of levoglucosan in pyrolysis liquids using high-performance liquid chromatography (HPLC) was developed. The method avoids the tedious and time-consuming sample preparation required by current analytical methods In this method, levoglucosan has better separation of xylose, glucose and fructose so as to avoid interference. The calibration curve coefficient of determination on levoglucosan was >0.990 determined by an evaporative light scattering detector (ELSD). The relative standards for the new method were ≤1.34% at low concentration and ≤2.56% at high concentration. The spiked levoglucosan recovery on the pyrolysis liquid samples were between 96.79 and 99.13%. The research demonstrates that it is possible to obtain excellent accuracy and efficiency using HPLC to quantificate levoglucosan found in fast pyrolysis bio-oils.

As one of the fundamental research issues, feature selection plays a critical role in machine learning. By the removal of irrelevant features, it attempts to reduce computational complexities of upstream tasks, usually with computation accelerations and performance improvements. This paper proposes an auto-encoder based scheme for unsupervised feature selection. Due to the inherent consistency, this framework can solve traditional constrained feature selection problems approximately. Specifically, the proposed model takes non-negativity, orthogonality, and sparsity into account, whose internal characteristics are exploited sufficiently. It can also employ other loss functions and flexible activation functions. The former can fit a wide range of learning tasks, and the latter has the ability to play the role of regularization terms to impose regularization constraints on the model. Thereinafter, the proposed model is validated on multiple benchmark datasets, where various activation and loss functions are analyzed for finding better feature selectors. Finally, extensive experiments demonstrate the superiority of the proposed method against other compared state-of-the-arts.

The knowledge about adsorption/desorption of cellulase on/from residual lignin is essential for efficient cellulose hydrolysis by recycling the cellulase from residual solid fraction. Herein, the impact of alkali pretreatment at different alkali strength on the adsorption/desorption of enzymatic residual lignins (ERLs) with cellulase was investigated. ERLs pretreated at stronger alkali strength had higher molecular weights and negative zeta potentials, but lower hydrophobicity. The inhibition of ERL to the enzymatic cellulose digestion was mitigated by alkali pretreatment, and inhibition degrees decreased from 21.5 % (with un-pretreated ERL) to 18.9 %, 12.9 %, and 10.8 % with the addition of ERLs treated with 0.5 %, 1.0 %, and 1.5 % (w/w) sodium hydroxide aqueous solution, respectively. The affinity of ERLs with cellulase diminished after alkali pretreatment, and the binding strength decreased from 183.0 to 130.7 mL/g with the alkali strength increasing, retaining more cellulase activities in supernatant. Desorption capacity of bound-cellulase from ERL increased after alkali pretreatment, and the recovery increased from 65.3% to 73.1% with the alkali strength increasing. The cellulase released after desorption exhibited hydrolytic activity, and higher glucose yields were obtained by the bound-cellulase on alkali-pretreated ERLs compared with that of un-pretreated ERL. The results provide references for the development of alkali pretreatment and recycling cellulase from residual lignin for efficient lignocellulosics digestion.

Clustering has long been an enduring and promising task in machine learning. However, developed one-side clustering is still insufficient to explore the context of data, such as texts and genes. Hence, developing two-way clustering has drawn more attention in recent years, which tends to cluster samples and features simultaneously. This paper proposes a sparse neighbor constrained co-clustering via category consistency learning, for alleviating the misclassification of close points. Following an additional observation, samples often fall into the same category as their neighbors, as do features. Accordingly, the co-clustering problem is formulated as nonnegative matrix tri-factorization appended dual regularizers, considering coherence between data affinity and label assignment. Then, a multiplicative alternating scheme is raised for objective optimization, whose convergence and correctness are theoretically guaranteed. Furthermore, the proposed approach is validated on six datasets using three evaluation metrics, whose parameter sensitivity is analyzed as well. Finally, comprehensive experiments show that our algorithm is competitive against existing ones.

Evaluating the technical feasibility and economic viability of citric acid production at the commercial scale is challenging because it depends on several factors including type of feedstock, process parameters and production routes. In this research, techno-economic and environmental impacts of citric acid production from corn are comparatively evaluated for three scenarios. In the production process Aspergillus niger is used for corn fermentation in a commercial scale of 6400 kg/h citric acid production rate. The scenarios are defined based on application of three recovery methods, including calcium precipitation, solvent extraction, and ion exchange. From three investigated scenarios, the lowest selling price of citric acid is achieved in a scenario using solvent extraction recovery method, followed by the scenario with calcium precipitation recovery method. Sensitivity analysis shows that the minimum selling price of citric acid is highly sensitive to electricity cost, by-product credit, corn supply cost, fermentor cost and wastewater treatment cost. Environmental impacts of citric acid production in three scenarios are assessed using the life cycle assessment approach. Citric acid production with ion exchange recovery method is the most environment-friendly scenario. The results of this research are useful for the development of citric acid production in commercial scale.