Hierarchical Tree Research Articles

Concept Decomposition for Visual Exploration and Inspiration

A creative idea is often born from transforming, combining, and modifying ideas from existing visual examples capturing various concepts. However, one cannot simply copy the concept as a whole, and inspiration is achieved by examining certain aspects of the concept. Hence, it is often necessary to separate a concept into different aspects to provide new perspectives. In this paper, we propose a method to decompose a visual concept, represented as a set of images, into different visual aspects encoded in a hierarchical tree structure. We utilize large vision-language models and their rich latent space for concept decomposition and generation. Each node in the tree represents a sub-concept using a learned vector embedding injected into the latent space of a pretrained text-to-image model. We use a set of regularizations to guide the optimization of the embedding vectors encoded in the nodes to follow the hierarchical structure of the tree. Our method allows to explore and discover new concepts derived from the original one. The tree provides the possibility of endless visual sampling at each node, allowing the user to explore the hidden sub-concepts of the object of interest. The learned aspects in each node can be combined within and across trees to create new visual ideas, and can be used in natural language sentences to apply such aspects to new designs. Project page: https://inspirationtree.github.io/inspirationtree/

Метод косвенных предпочтений формирования весов критериев с многоуровневой структурой

The method of forming criterion weights in tasks of multi-criteria evaluation of objects with a multi-level structure of indicators presented in the form of a hierarchical tree, is shown. The weights values are calculated on the basis of indi-rect measurement of preferences of adjacent pairs of local weights of criteria included in the vertices of a higher level of the hierarchical tree, in the form of expert assessments in the quantitative scale of relations. The calculation of quantitative weights first comes down to lexicographic sorting by descending order of criteria importance at each level of the hierarchy that are included in the vertices of a higher level, as a result of which the number of expert compari-sons of adjacent pairs in the scale of relations is reduced. The formation of local coefficients of the criteria importance is mathematically justified and is based on a matrix with special properties. A comparative analysis of the proposed method for forming quantitative weights with the method of hierarchy analysis, the least squares method and the ap-proximation matrix method based on the matrix of paired comparisons, is given. An example is given of solving the problem of multi-criteria evaluation of combat aircraft that participated in the tender, according to the tactical and technical characteristics presented in various measurement scales.

Bioconversion of citrus waste by long-term DMSO-cryopreserved rumen fluid to volatile fatty acids and biogas is feasible: A microbiome perspective

Preserving rumen fluid as the inoculum for anaerobic digestion of food waste is necessary when access to animal donors or slaughterhouses is limited. This study aims to compare two preservation methods relative to fresh ruminal inoculum: (1) cryoprotected with 5% dimethyl sulfoxide (DMSO) and stored at −20 °C and (2) frozen at −20 °C, both for 6 months. The fermentation activity of different inoculum was evaluated by rumen-based in vitro anaerobic fermentation tests (volatile fatty acids, biomass digestibility, and gas production). Citrus pomace was used as the substrate during a 96-h fermentation. The maximum volatile fatty acids, methane production, and citrus pomace digestibility from fresh rumen fluid were not significantly different from rumen fluid preserved with DMSO. Metagenome analysis revealed a significant difference in the rumen microbial composition and functions between fresh rumen fluid and frozen inoculum without DMSO. Storage of rumen fluid using −20 °C with DMSO demonstrated the less difference compared with fresh rumen fluid in microbial alpha diversity and taxa composition. The hierarchical clustering tree of CAZymes showed that DMSO cryoprotected fluid was clustered much closer to the fresh rumen fluid, showing more similarity in CAZyme profiles than frozen rumen fluid. The abundance of functional genes associated with carbohydrate metabolism and methane metabolism did not differ between fresh rumen fluid and the DMSO–20 °C, whereas the abundance of key functional genes significantly decreased in frozen rumen fluid. These findings suggest that using rumen liquid preserved using DMSO at −20 °C for 180 days is a feasible alternative to fresh rumen fluid. This would reduce the need for laboratories to maintain animal donors and/or reduce the frequency of collecting rumen fluid from slaughterhouses.

Topological Network Analysis of Beta‐Amyloid and Tau in Alzheimer’s Disease Using PET Imaging Data

AbstractBackgroundAlzheimer’s disease (AD) is characterized by the presence of beta‐amyloid (Aβ) and tau in the brain. Understanding the interplay between these proteins and their spatial distribution may provide insights into disease progression. This study aims to investigate the topological features of Aβ and tau networks in AD, mild cognitive impairment (MCI), and normal cognition (NC) groups using PET images from the ADNI dataset.MethodPartial correlation matrices between the standardized uptake value ratio of 68 regions of interest (ROI’s) were used to generate networks for the AD, MCI, and NC groups, with nodes representing ROI’s and edges representing correlation coefficients. Topological features, including H0 features (connected components), H1 features (loops), and landscape functions of H1 features, were calculated for each sample’s persistent homology. For each group, 100 bootstrapping samples were used to generate the 95% confidence intervals for landscape functions. The connected components were further studied by representing the samples as hierarchical tree structures. In the tree space, the first few generated connections, representing the most consistent pairs of regions positive for Aβ and tau, were compared across groups. Clustering performances were compared using Wasserstein distance of H0 features, Wasserstein distance of H1 features, landscape function distance, and geodesic distance in tree space respectively.ResultThe geodesic distance in tree space showed the best clustering accuracy in both Aβ and tau analyses. Compared with Aβ, H1 features and the corresponding landscape functions performed better for tau. Unique connections in AD and MCI groups were identified, with AD showing more unique connections than MCI compared to NC samples. Dominant unique connections in the AD group included left‐right isthmus cingulate and left fusiform‐left inferior temporal for Aβ, and left‐right paracentral and left cuneus‐left pericalcarine for tau. These connections indicate region pairs with related concentrations of Aβ and tau in AD.ConclusionTopological network analysis can reveal complex underlying structures in PET images and identify consistent patterns in ROI’s associated with Aβ and tau localization. Tau aggregation exhibits more complex behavior than Aβ and is more strongly associated with AD. The identified unique connections in the AD group may serve as potential biomarkers for disease progression.

Hierarchical Clustering Based on Dendrogram in Sustainable Transportation Systems

Each group in a data-driven automobile network has its cluster head. A group can communicate with each other and members of other groups once it has been founded. Vehicles belonging to each group near the other group allow intergroup communication. Because nodes in automotive networks move so quickly, routing in these networks is a complex problem to solve. Each cluster in hierarchical clustering can be partitioned into multiple sub-clusters. Put another way, and the data is stored in a cluster, which is then divided into more clusters. The data is stored directly in separate clusters in non-hierarchical approaches. A dendrogram is a type of hierarchical tree. We anticipate increasing information sharing in clusters by properly clustering vehicles on the road and establishing clusters of the desired size in the relevant dendrogram. We can select clusters of the necessary extent and compare the Quality of Service (QoS) network’s outcomes by breaking the dendrogram at different levels. The findings reveal that the suggested method outperforms AIVISN in delay, PDR, overhead, and Drooped packets compared to AIVISN, 7.12%, 12.21%,8.32%, and 7.34%, respectively.

WebDraw: A machine learning-driven tool for automatic website prototyping

To overcome the time-consuming nature and improve the cost-effectiveness of classical web development, being automatic is the most convenient alternative recent researchers suggest. Over the years, researchers have been working on inventing new approaches to generating websites automatically. In this paper, a novel approach is presented that automates the website generation process by incorporating web designer best practices and driving new prototype websites without the significant effort of redesigning websites. It takes high-fidelity mock-up design artifacts such as screen captures of real-world websites, and generates functional websites similar to the input websites, which involves three steps: GUI element detection, classification, and code generation. First, image processing techniques are applied to detect atomic web GUI elements from a mock-up design artifact of a real-world website. Second, a Convolutional Neural Network (CNN) is trained to classify the extracted web GUI elements into their domain-specific types such as headings, paragraphs, images, etc. A Graphical User Interface (GUI) is typically represented in code as a hierarchical tree, with nested GUI elements bundled together within one another to construct a tree. A recursive algorithm is proposed that constructs the appropriate Document Object Model (DOM) hierarchy for a website by recursively grouping classified web GUI elements within each other. Finally, the constructed DOM is converted to the accurate native code. The approach was implemented as a tool called WebDraw. Design science research evaluation shows that WebDraw achieves an average of 90% web GUI element classification and generates website prototypes that are visually similar to the target website design mock-up artifact while producing functional GUI code. Furthermore, interviews with industry professionals illustrate WebDraw's industry relevance for improving their industrial web design and development workflows.

Molecular Taxonomy of Myelodysplastic Syndromes and Its Clinical Implications

Background Recurrent genetic abnormalities are responsible for the pathogenesis of myelodysplastic syndromes (MDS) and their biologic heterogeneity, however current classifications of MDS are still largely based on morphology.Here, we propose a molecular taxonomy of MDS and describe its clinical relevance. Methods We assembled a cohort of 3,233 diagnostic MDS or related myeloid neoplasm samples from the International Working Group for the Prognosis of MDS. Gene mutations, copy-number alterations (CNAs) and copy-neutral loss of heterozygosity (cnLOH) events were derived from targeted sequencing of a 152-gene panel. Molecular features were used for unsupervised clustering analysis using Bayesian Dirichlet processes, and the results were manually curated into a rule-based hierarchical classification tree. Results We detected oncogenic mutations, CNAs, and cnLOH events in 91, 43, and 11% of patients, respectively. We validated and extended patterns of co-occurrence and mutual exclusivity between gene mutations and CNAs, and further described (i) specific occurrences in cis between gene mutations and cnLOH events, (ii) distinct patterns of co-mutation between the allelic state of a same locus (haploid LOH or cnLOH associated with -7/del(7q)), or of the same mutated gene ( TP53, TET2), and (iii) distinct patterns of co-mutation between hotspots of the same gene ( U2AF1). These patterns informed feature definition for clustering analysis. We characterized 18 distinct MDS molecular subgroups. The first 16 groups encompassed 86% (n=2,769) of patients and were based on the presence or absence of mutations in 21 genes ( ASXL1, BCOR, BCORL1, DDX41, DNMT3A, EZH2, FLT3, IDH1, IDH2, MLL, MYC, NPM1, SETBP1, SF3B1, SRSF2, STAG2, TET2, TP53, U2AF1, WT1, ZRSR2), and 6 cytogenetic events (inv(3), complex karyotype, der(1;7), -7, del(5q), -Y), as well as LOH at the TP53 and TET2 locus. The molecularly not-otherwise specified (mNOS) group(8%, n=254) corresponded to the presence of other cytogenetic abnormalities and/or mutations in 51 other recurrently mutated genes (>0.5%). The No-event category (6%, n=210) was characterized by the absence of any recurrent drivers in our assay. Groups ranged in size from 0.5% to 14% of patients and were associated with distinct clinical phenotypes ( Fig 1A) and outcomes ( Fig 1B). The median bone marrow blast percentage across groups ranged from 1.5 to 10%, the median overall survival (OS) from 0.9 to 8.2 years, and the 2-year rate of leukemic transformation from 0 to 40%. The representation of molecular groups within blast strata (<5%, 5-9%, 10-19% blast percentages) was highly heterogeneous: 7, 8, and 9 different groups accounted for more than 5% of patients within each blast stratum, respectively. The prognostic impact of blast counts on outcomes depended on the molecular subgroups. Among the 12 molecular groups with at least 10 patients with outcome data in all 3 blast strata, blast count was prognostically significant in 8 of those groups ( SF3B1, CCUS-like, del(5q), bi-TET2, mNOS, BCOR/L1, IDH-STAG2, TP53-complex) and was not significant in 4 groups ( EZH2-ASXL1, -7/SETBP1, DDX41, AML-like). The molecular groups validated established ( TP53-complex, del(5q), SF3B1) and well characterized entities ( DDX41, AML-like), added further evidence to support previously reported subsets ( bi-TET2, der(1;7), CCUS-like), and described novel genetic subsets (-7/ SETBP1, EZH2-ASXL1, IDH-STAG2, BCOR/L1, U2AF157, U2AF134, SRSF2, ZRSR2). Finally, the MDS molecular groups were used to investigate the genetic heterogeneity of secondary/therapy-related MDS (s/t-MDS) and of myelodysplastic/myeloproliferative neoplasms (MDS/MPN). There were 7 and 6 distinct groups that accounted for more that 5% of patients with s/t-MDS or MDS/MPN, respectively. Within each genetic subgroup, s/t-MDS and MDS/MPN had similar clinical and outcome profiles to primary MDS. Conclusions The genetic complexity of MDS can be organized into 18 molecular groups that reflect the underlying genetic basis of the disease. The prognostic influence of bone marrow blasts varied in the individual genetic subgroups, suggesting that the clinical impact of increased blasts depends on the genetic context. The molecular taxonomy derived in this study is clinically relevant and will inform future classification schemas.

Number-enhanced representation with hierarchical recursive tree decoding for math word problem solving

Automatic solving math word problems (MWPs) is a number-intensive application in natural language processing (NLP). However, these existing methods are far from achieving acceptable levels of numeracy learning. As a result, the performance of these models is limited for mathematical reasoning. In addition, the mainstream tree decoder suffers from early-stage information loss, resulting in an unsatisfactory performance for complex problems with more operators. In this paper, we propose NERHRT (Number-Enhanced Representation with Hierarchical Recursive Tree Decoding), a simple yet effective number embedding method that produces the numerical reasoning via the decimal notation-based embedding and a dual-direction graph attention network. In addition, a hierarchical recursive tree-structured decoder is introduced to aggregate information from all ancestor nodes. Experiments show that our approach obtains the best performance on four popular benchmark datasets, and beats the state-of-the-art models with a large margin.

Identification of circulating miRNA as early diagnostic molecular markers in malignant glioblastoma base on decision tree joint scoring algorithm.

The lack of clinical markers prevents early diagnosis of glioblastoma (GBM). Many studies have found that circulating microRNAs (miRNAs) can be used as early diagnostic markers of malignant tumours. Therefore, the identification of novel circulating miRNA biomolecular markers could be beneficial to clinicians in the early diagnosis of GBM. We developed a decision tree joint scoring algorithm (DTSA), systematically integrating significance analysis of microarray (SAM), Pearson hierarchical clustering, T test, Decision tree and Entropy weight score algorithm, to screen out circulating miRNA molecular markers with high sensitivity and accuracy for early diagnosis of GBM. DTSA was developed and applied for GBM datasets and three circulating miRNA molecular markers were identified, namely, hsa-miR-2278, hsa-miR-555 and hsa-miR-892b. We have found that hsa-miR-2278 and hsa-miR-892b regulate the GBM pathway through target genes, promoting the development of GBM and affecting the survival of patients. DTSA has better classification effect in all data sets than other classification algorithms, and identified miRNAs are better than existing markers of GBM. These results suggest that DTSA can effectively identify circulating miRNA, thus contributing to the early diagnosis and personalised treatment of GBM.

A Multimodal Single Cell Atlas of Adult Murine Haematopoiesis Reveals Nine Differentiation Trajectories and Epigenetic Regulatory Programs

Single cell genomics has revolutionised our understanding of differentiating systems and led to a reinterpretation of haematopoiesis, moving from a step-wise hierarchical tree model towards more continous differentiation landscapes. Recent single cell RNA sequencing (scRNA-seq) studies of Lineage- ckit+ (LK) mouse bone marrow progenitors have defined a highly granular landscape of differentiation from multipotent stem cells to committed progenitors in 8 different lineages and have served as a valuable reference landscape for comparisons with perturbation states. While these landscapes have permitted exploration and interrogation of early haematopoietic progenitors, as yet, no map of mouse total bone marrow haematopoiesis exists at single cell resolution, limiting efforts to fully understand the gene programs which specify and the cell surface phenotypes which identify maturing haematopoietic lineages nor the connection between early bifurcation decisions in progenitor cells and mature cellular outputs. To answer these questions, we performed droplet-based scRNA-seq and cell surface proteomics on whole bone marrow from 4 male and 4 female 12-week old C57Bl6 mice, using a panel of 138 antibodies against cell surface antigens. To increase representation of rare progenitor populations, we complemented total bone marrow mononuclear cells with c-kit enrichment and FACS sorted LK and LK sca1-positive (LSK) populations. After quality control filtering the single cell transcriptomes and proteomes, over 198,000 single cells were analysed separately before multimodal integration into a single UMAP embedding. Addition of cell surface proteome information to the scRNA-seq data through multimodal integration produced a representation with greater resolution of haematopoietic cell types, including improved resolution of T-cell subtypes and separation of a rare population of eosinophil progenitors which were previously indistinguishable from basophil progenitors. This integrated landscape of mouse haematopoiesis was then clustered and annotated though an iterative process of label transfer and manual annotation into 41 cell type states (Figure 1). Using cell fate analysis, which utilises transcriptional similarity and pseudotemporal ordering to estimate the probability of a single cell's commitment to defined terminal states, we were able to resolve cell fate probabilities towards nine haematopoietic lineages (erythroid, neutrophil, megakaryocyte, lymphoid, monocyte, dendritic cell (DC), plasmacytoid dendritic cell (pDC), basophil and eosinophil (Figure 2)). From these probabilities we defined nine trajectories from the earliest uncommitted progenitors to terminally differentiated populations, including a pDC trajectory involving both lymphoid and common DC progenitors. Analysis of trajectory-based gene expression trends provides a framework for discovery of lineage drivers as well as shared gene expression programs across multiple lineages. Leveraging cell surface proteome data permits the use of ‘in-silico-FACS’ to describe phenotypes of cell populations which exist at cell fate branch-points, for further in-vitro characterisation. To demonstrate the utility of our atlas as a reference, we projected perturbation datasets from 7 pre-leukaemic mouse models onto the atlas, allowing quantitation of mutation-associated cellular and tissue-scale alterations. Finally, we have leveraged the transcription-start site enrichment of 5' scRNA-seq to compute enhancer RNA (eRNA) expression in our atlas, allowing the identification of dynamic usage of specific enhancer elements during haematopoietic specification and thus providing a novel framework for correlation of eRNA expression (as a proxy for active distal chromatin regions) with gene expression trends along specific haematopoietic lineages.

HIE-EDT: Hierarchical interval estimation-based evidential decision tree

Decision tree algorithm, because of its strong interpretability and high algorithm efficiency, is widely used in the field of pattern recognition and classification. When the number of data samples is small and there is uncertainty in the data, it is difficult for the traditional decision tree algorithm to fully mine the effective information in the data. In this paper, we use the Dempster–Shafer framework to model data uncertainty and propose a hierarchical interval estimation method to improve decision tree algorithms. The proposed method constructs intervals through two methods of attribute boundary and mean square error estimation, which not only utilizes the characteristics of intervals to model the inaccuracy of data, but also constrains intervals from two aspects, narrowing the representation range of available information. By comparing with the classic decision tree algorithm and the decision tree algorithm based on single interval estimation, the proposed method can perform classification tasks robustly and accurately in different types of data under seven data sets.

Recursive neural programs: A differentiable framework for learning compositional part-whole hierarchies and image grammars.

Human vision, thought, and planning involve parsing and representing objects and scenes using structured representations based on part-whole hierarchies. Computer vision and machine learning researchers have recently sought to emulate this capability using neural networks, but a generative model formulation has been lacking. Generative models that leverage compositionality, recursion, and part-whole hierarchies are thought to underlie human concept learning and the ability to construct and represent flexible mental concepts. We introduce Recursive Neural Programs (RNPs), a neural generative model that addresses the part-whole hierarchy learning problem by modeling images as hierarchical trees of probabilistic sensory-motor programs. These programs recursively reuse learned sensory-motor primitives to model an image within different spatial reference frames, enabling hierarchical composition of objects from parts and implementing a grammar for images. We show that RNPs can learn part-whole hierarchies for a variety of image datasets, allowing rich compositionality and intuitive parts-based explanations of objects. Our model also suggests a cognitive framework for understanding how human brains can potentially learn and represent concepts in terms of recursively defined primitives and their relations with each other.

A Design and Its Application of Multi-Granular Fuzzy Model with Hierarchical Tree Structures

This paper is concerned with the design of a context-based fuzzy C-means (CFCM)-based multi-granular fuzzy model (MGFM) with hierarchical tree structures. For this purpose, we propose three types of hierarchical tree structures (incremental, aggregated, and cascaded types) in the design of MGFM. In general, the conventional fuzzy inference system (FIS) has problems, such as time consumption and an exponential increase in the number of if–then rules when processing large-scale multivariate data. Meanwhile, the existing granular fuzzy model (GFM) reduces the number of rules that increase exponentially. However, the GFM not only has overlapping rules as the cluster centers become closer but also has problems that are difficult to interpret due to many input variables. To solve these problems, the CFCM-based MGFM can be designed as a smaller tree of interconnected GFMs. Here, the inputs of the high-level GFMs are taken from the output to the low-level GFMs. The hierarchical tree structure is more computationally efficient and easier to understand than a single GFM. Furthermore, since the output of the CFCM-based MGFM is a triangular fuzzy number, it is evaluated based on a performance measurement method suitable for the GFM. The prediction performance is analyzed from the automobile fuel consumption and Boston housing database to present the validity of the proposed approach. The experimental results demonstrate that the proposed CFCM-based MGFM based on the hierarchical tree structure creates a small number of meaningful rules and solves prediction-related problems by making them explainable.

Income inequality in later years: occupational trajectories or initial social characteristics?

This study focuses on the constitution of financial reserves in Switzerland from a longitudinal perspective. Personal income after retirement derives from financial reserves whose constitution depends both on positional factors, such as sex and birth cohorts, and processual factors, such as occupational trajectories, in the institutional context of the Swiss pension system (structural factors). We hypothesise that some processual, positional and structural factors interact with each other to shape financial reserves available in old age. We assess this set of factors and their interactions using the occupational trajectory types stemming from optimal matching analysis (OMA) combined with the hierarchical clustering and regression tree methods. We used the retrospective biographic data SHARELIFE gathered during the third wave of the SHARE survey in 2009. The results show that occupational trajectories are influential factors accounting for much of the financial reserves available in later life. However, these processual factors interact with positional factors such as sex and birth cohort. The retirement schemes generalised in Switzerland during the period under consideration add up to the effect of positional factors on the constitution of financial reserves.

An opportunistic energy‐efficient dynamic self‐configuration clustering algorithm in WSN‐based IoT networks

SummaryThe demand for the Internet of Things (IoT) has significantly increased in the current scenario; specific sectors that require IoT include industrial automation, home control, health care applications, military and surveillance applications, habitat monitoring, and nanoscopic sensor applications. The use of optimal wireless sensor networks (WSNs) in transmission techniques has resulted in their involvement. A WSN is made up of thousands of randomly distributed sensor nodes that sense and transmit environmental data such as temperature, pressure, humidity, light, and sound. One of the most important requirements when using these sensor nodes is energy. As a result, it has become a major area of research in recent years; additionally, several design techniques and protocols have been presented in the last decade, particularly for IoT‐based applications. As a result, the systemization of an energy‐optimized WSN in dynamic functional conditions with automatic self‐configuration of sensor nodes is a critical goal. This paper proposed an opportunistic energy‐efficient dynamic self‐configuration routing (OEDSR) algorithm for IoT‐based applications. Initially, the optimal route to the base station (BS) is calculated by using the residual energy and mobility factor of the sensor nodes obtained through a routing tree model based on graph theory. To reduce the number of connections, an optimal path is determined based on dynamic cluster generation through a hierarchical tree architecture. Finally, the network‐related parameters, such as throughput, delay and packet delivery ratio (PDR), are compared with the peer existing routing protocols to depict the efficiency of the OEDSR protocol.

DASH: Dynamic Attention-Based Substructure Hierarchy for Partial Charge Assignment.

We present a robust and computationally efficient approach for assigning partial charges of atoms in molecules. The method is based on a hierarchical tree constructed from attention values extracted from a graph neural network (GNN), which was trained to predict atomic partial charges from accurate quantum-mechanical (QM) calculations. The resulting dynamic attention-based substructure hierarchy (DASH) approach provides fast assignment of partial charges with the same accuracy as the GNN itself, is software-independent, and can easily be integrated in existing parametrization pipelines, as shown for the Open force field (OpenFF). The implementation of the DASH workflow, the final DASH tree, and the training set are available as open source/open data from public repositories.

Generation matrix: An embeddable matrix representation for hierarchical trees

Starting from the local structures to study hierarchical trees is a common research method. However, the cumbersome analysis and description make the naive method challenging to adapt to the increasingly complex hierarchical tree problems. To improve the efficiency of hierarchical tree research, we propose an embeddable matrix representation for hierarchical trees, called Generation Matrix. It can transform the abstract hierarchical tree into a concrete matrix representation and then take the hierarchical tree as a whole to study, dramatically reducing the research complexity. Mathematical analysis shows that Generation Matrix can simulate various recursive algorithms without accessing local structures and provides a variety of interpretable matrix operations to support the research of hierarchical trees. Applying Generation Matrix to differential privacy hierarchical tree release, we propose a Generation Matrix-based optimally consistent release algorithm (GMC). It provides an exceptionally concise process description so that we can describe its core steps as a simple matrix expression rather than multiple complicated recursive processes like existing algorithms. Our experiments show that GMC takes only a few seconds to complete a release for large-scale datasets with more than 10 million nodes. The calculation efficiency is increased by up to 100 times compared with the state-of-the-art schemes.

Алгоритмы предиктивной аналитики параметров инженерных систем зданий

The article discusses the relevance of deep learning methods using in predictive analytics of parameters of engineering systems in construction. The advantages and disadvantages of such methods are described, as well as the architecture of a cyberphysical type system based on machine learning for solving the problem of predictive analytics algorithms for building engineering system parameters. Conclusions are drawn about the applicability of the DBSCAN clustering algorithms, neural networks with the Attention mechanism and hierarchical trees ensembles within the research task.

Automated ICD coding using extreme multi-label long text transformer-based models

Encouraged by the success of pretrained Transformer models in many natural language processing tasks, their use for International Classification of Diseases (ICD) coding tasks is now actively being explored. In this study, we investigated two existing Transformer-based models (PLM-ICD and XR-Transformer) and proposed a novel Transformer-based model (XR-LAT), aiming to address the extreme label set and long text classification challenges that are posed by automated ICD coding tasks. The Transformer-based model PLM-ICD, which currently holds the state-of-the-art (SOTA) performance on the ICD coding benchmark datasets MIMIC-III and MIMIC-II, was selected as our baseline model for further optimisation on both datasets. In addition, we extended the capabilities of the leading model in the general extreme multi-label text classification domain, XR-Transformer, to support longer sequences and trained it on both datasets. Moreover, we proposed a novel model, XR-LAT, which was also trained on both datasets. XR-LAT is a recursively trained model chain on a predefined hierarchical code tree with label-wise attention, knowledge transferring and dynamic negative sampling mechanisms. Our optimised PLM-ICD models, which were trained with longer total and chunk sequence lengths, significantly outperformed the current SOTA PLM-ICD models, and achieved the highest micro-F1 scores of 60.8 % and 50.9 % on MIMIC-III and MIMIC-II, respectively. The XR-Transformer model, although SOTA in the general domain, did not perform well across all metrics. The best XR-LAT based models obtained results that were competitive with the current SOTA PLM-ICD models, including improving the macro-AUC by 2.1 % and 5.1 % on MIMIC-III and MIMIC-II, respectively. Our optimised PLM-ICD models are the new SOTA models for automated ICD coding on both datasets, while our novel XR-LAT models perform competitively with the previous SOTA PLM-ICD models.

A Literature Database Review of the Competitive Factors That Influence the Production and Use of Whey in the Brazilian Dairy Industry.

The consumption of milk and derivatives is part of the diet of a large part of the population. The substantial demand for dairy products has prompted the industry to expand its product range by incorporating whey, a previously disregarded by-product, as a significant ingredient. Consequently, the application of whey within the dairy sector has escalated, fostering novel food trends driven by market, nutritional, technical, environmental, and economic considerations. In this context, the primary objective of this research was to identify, categorise, and analyse the competitive factors influencing the production and use of whey within the dairy products industry and to correlate these factors with Brazil's economic, food, and nutritional landscape. A comprehensive literature review encompassing 41 articles and scientific documents sourced from PubMed, Scopus, and SciELO databases and supplementary research was undertaken to pinpoint these factors. A total of seventeen competitive factors that exert influence over the production and use of whey within the dairy industry were identified. These seventeen factors were systematically classified and modelled into a hierarchical decision tree structure. A meticulous analysis of these factors revealed a spectrum of strengths, weaknesses, opportunities, and threats associated with this sector, considering the Brazilian nutritional, food, and economic context. This research will help elaborate healthy strategies for developing new products with whey in the composition and maintaining the nutritional quality for the consumer. In addition, the research can help companies manage their operations, which can be directed towards improving their performance in the factors discussed in this research, with reflections and impacts on competitiveness in nutritional, environmental, economic, technological, and organisational dimensions.