Neighborhood Model Research Articles

Functional traits shape plant-plant interactions and recruitment in a hotspot of woody plant diversity.

Understanding and predicting recruitment in species-rich plant communities requires identifying functional determinants of both density-independent performance and interactions. In a common-garden field experiment with 25 species of the woody plant genus Protea, we varied the initial spatial and taxonomic arrangement of seedlings and followed their survival and growth during recruitment. Neighbourhood models quantified how six key functional traits affect density-independent performance, interaction effects and responses. Trait-based neighbourhood models accurately predicted individual survival and growth from the initial spatial and functional composition of species-rich experimental communities. Functional variation among species caused substantial variation in density-independent survival and growth that was not correlated with interaction effects and responses. Interactions were spatially restricted but had important, predominantly competitive, effects on recruitment. Traits increasing the acquisition of limiting resources (water for survival and soil P for growth) mediated trade-offs between interaction effects and responses. Moreover, resprouting species had higher survival but reduced growth, likely reinforcing the survival-growth trade-off in adult plants. Resource acquisition of juvenile plants shapes Protea community dynamics with acquisitive species with strong competitive effects suffering more from competition. Together with functional determinants of density-independent performance, this makes recruitment remarkably predictable, which is critical for efficient restoration and near-term ecological forecasts of species-rich communities.

Forecasting cryptocurrencies volatility using statistical and machine learning methods: A comparative study

Forecasting cryptocurrency volatility can help investors make better-informed investment decisions in order to minimize risks and maximize potential profits. Accurate forecasting of cryptocurrency price fluctuations is crucial for effective portfolio management and contributes to the stability of the financial system by identifying potential threats and developing risk management strategies. The objective of this paper is to provide a comprehensive study of statistical and machine learning methods for predicting daily and weekly volatility of the following four cryptocurrencies: Bitcoin, Ethereum, Litecoin, and Monero. Several models and forecasting methods are compared in terms of their forecasting accuracy, i.e., HAR (heterogeneous autoregressive), ARFIMA (autoregressive fractionally integrated moving average), GARCH (generalized autoregressive conditional heteroscedasticity), LASSO (least absolute shrinkage and selection operator), RR (ridge regression), SVR (support vector regression), MLP (multilayer perceptron), FNM (fuzzy neighbourhood model), RF (random forest), and LSTM (long short-term memory). The realized variance calculated from intraday returns is used as the input variable for the models. In order to assess the predictive power of the models considered, the model confidence set (MCS) procedure is applied. Our experimental results demonstrate that there is no single best method for forecasting volatility of each cryptocurrency, and different models may perform better depending on the specific cryptocurrency, choice of the error metric and forecast horizon. For daily forecasts, the method that is always found in a set of best models is linear SVR, while for weekly forecasts, there are two such methods, namely FNM and RR. Furthermore, we show that simple linear models such as HAR and ridge regression, perform not worse than more complex models like LSTM and RF. The research provides a useful reference point for the development of more sophisticated models.

Contrasted spatial, demographic and genetic structures of a light-demanding African timber species, Cylicodiscus gabunensis Harms – Implications for a sustainable management of its populations

Most Central African rainforest canopies consist of light-demanding tree species that hold high commercial value but also suffer locally from regeneration deficits, raising concerns about the sustainability of logging. Regeneration is influenced by factors such as past perturbations (including human activity), mating systems, and seed/pollen dispersal processes that impact demographic, spatial, and genetic structures within populations. To gain a better understanding of these interactions, we studied the spatial distribution and trunk diameter structure of Cylicodiscus gabunensis (Fabaceae) - a wind-dispersed, insect-pollinated, timber species - in three plots ranging from 400 to 839 ha situated in various environmental contexts (e.g. forest types and elephant densities) across Central Africa. We also genotyped adults and juveniles using microsatellite markers to analyze the spatial genetic structure of each population and infer the selfing rate, seed and pollen dispersal capacities and selection gradients using the ‘neighborhood model’. The selfing rate was low (3 – 4 %), and seed dispersal distances (ds = 184 m) were much shorter than pollen dispersal distances (dp > 2 km). The three populations displayed contrasted spatial, demographic and genetic structures. One population showed no spatial aggregation or genetic structure, and a multimodal diameter structure indicating pulses of regeneration events. Two populations showed strong spatial aggregation and genetic structures. One exhibited a unimodal diameter structure indicating one ancient pulse of regeneration, while the other displayed a 'reverse J-shaped' diameter structure, typical of ongoing regeneration. In the latter, reproductive success appeared leptokurtic, three mother trees accounting for over 90 % of the regeneration and no tree below the minimum cutting diameter implemented by logging companies had offspring. The idiosyncratic nature of population characteristics observed in C. gabunensis suggests that, for sustainable management, a nuanced approach is needed. This involves protecting productive seed trees in areas where natural regeneration is occurring and actively supporting regeneration in areas exhibiting deficits, especially in contexts with low elephant densities.

Microclimate impacts of neighborhood redesign in a desert community using ENVI-met and MaRTy

Municipalities often consider heat mitigation strategies to address urban overheating, but the location of implementation rarely is co-located with the communities that are carrying the majority of the heat burden in the city. The City of Phoenix, is redeveloping a public housing community with a focus on urban cooling as a desired outcome. This research uses in situ measurements (including the mobile micro-meteorological measurement cart, MaRTy) and ENVI-met microscale modeling of the neighborhood to assess air temperature (Tair) cooling capabilities of the planned redesigns to the neighborhood. After validating the ENVI-met model of the current neighborhood with fixed and mobile measurements with an index of agreement d > 0.9 and d > 0.8, respectively, analysis of the planned urban design shows some cool spots connected to new shade and vegetated corridors with Tair cooling magnitudes as high as 3 °C. Yet, some exposed and building-adjacent areas were identified as potential hot spots in the planned neighborhood. These hotspots underscore the importance of continued collaboration among the City, researchers, and the community to address the needs of the community for the creation of healthier urban environments.

Detection of sparse differential dependent functional brain connectivity.

Functional brain connectivity analysis is an increasingly important technique in neuroscience, psychiatry, and autism research. Functional connectivity can be measured by considering co-activation of brain regions in resting-state functional magnetic resonance imaging (rs-fMRI). We propose a novel Bayesian model to detect differential connections in cross-correlated functional connectivity between region of interest (ROI) pairs. The proposed sparse clustered neighborhood model induces a lower-dimensional sparsity and clustering based on a nonparametric Bayesian approach to model sparse differentially connected ROI pairs. Second, it induces a structured dependence model for modeling potential dependence among ROI pairs. We demonstrate Bayesian inference and performance of the proposed model in simulation studies and compare with a standard model. We utilize the proposed model to contrast functional connectivities between participants with autism spectrum disorder and neurotypical participants using cross-correlated rs-fMRI data from four sites of the Autism Brain Image Data Exchange.

Urbanization Process: A Simulation Method of Urban Expansion Based on RF-SNSCNN-CA Model

One of the focal points in Geographic Information Science (GIS) is to uncover the spatial distribution patterns of geographical phenomena. In response to the insufficient spatial feature learning concerning neighborhoods in traditional machine learning-based Cellular Automata (CA) models for land use change, this study couples the Random Forest (RF) model and the Spatially Non-Stationary Convolutional Neural Network (SNSCNN) model to the CA model. The resulting RF-SNSCNN-CA model considers the issue of spatial non-stationarity by incorporating attention mechanisms. Using observed urban land change data from 2010 to 2017 in the 21 districts of Chongqing’s main city as an example, two sets of experiments comprising eight scenarios were designed to verify the neighborhood effects. The results demonstrate that the proposed RF-SNSCNN-CA model achieves an Overall Accuracy (OA) of 97.82%, Kappa of 0.7683, and Figure of Merit (FoM) of 0.3836. The study reveals the following findings. Firstly, the RF-SNSCNN-CA model integrates the dual advantages of traditional machine learning and deep learning models, in which SNSCNN improves by the combined effect of channel and spatial attention mechanisms improves the learning of neighborhood features; secondly, the machine learning-like urban sprawl CA modeling process, regardless of the approach taken to obtain development suitability, cannot completely replace the learning of the neighborhood part; lastly, the use of traditional neighborhood modeling methods may produce suppression of simulation results and make the model inadequately learn spatial features.

A particle swarm optimization algorithm based on diversity-driven fusion of opposing phase selection strategies

Opposition-based learning (OBL) is often embedded in intelligent optimization algorithms to solve practical engineering and mathematical problems, but the combinatorial problems among different OBL variants are rarely studied. To this end, we propose a novel OBL variant based on the principle of optical imaging, which combines two novel types of quasi-opposite learning and extended opposite learning, called diversity-driven fused opposition learning (SQOBL). First, a density center based on a neighborhood model is proposed. Based on the rapid convergence of the centroid, combined the advantages of density and centroid to construct a double mean center (DMC) to replace the original center point in quasi-opposite learning based on the principle of refraction. Secondly, an extended opposite learning method based on optical refraction imaging is proposed. Diversity is then exploited to drive different opposing learning strategies at different stages of evolution, thus controlling the exploration and utilization of the algorithm. Finally, SQOBL was embedded in the PSO with eight others representative OBL variants to find the most optimal solution for a test suite. In addition, 8 novel intelligent optimization algorithms and the first three algorithms were selected to evaluate the performance of the latest CEC2022 benchmark test set and realistic constrained optimization problems. Experiments with 56 test functions and 3 real-world constraint optimization problems show that the proposed SQOBL has good integrative properties in CEC2015, CEC2017, CEC2020, and CEC2022 test suites.

The neighborhood as an energy-generating community: the case of the El Banafseg residential neighborhood, East Cairo

Nowadays, energy consumption is a global challenge due to the shortage of some energy sources and, consequently, their increasing prices. This paper argues that by using an integrative approach for using various energy harvesting techniques, a typical residential neighborhood can be operated as a self-generator of some or all of its needed energy. The paper follows mixed qualitative, experimental, and numerical simulation approaches to examine the practicality of the hypothesis. It uses Space syntax modeling theory (socio-spatial software) to analyze the pedestrian flow to indicate potential locations of kinetic energy harvesting piezoelectric tiles in the selected residential neighborhood. In addition, the possibility of installing photovoltaic electrical cells on rooftops of the neighborhood residential blocks is studied. The numerical simulation quantifies the capacity of the harvested energy to achieve a model of a green neighborhood. The paper measures the percentage of the residences’ energy needs that can be harvested using renewable energy resources. It reveals that 30% of the energy consumed in the residential blocks can be generated from photovoltaic electrical cells, and 38% of the energy consumed by street lighting of the entire neighborhood could be generated using the kinetic energy harvested by piezoelectric tiles.

An intelligent modeling framework to optimize the spatial layout of ocean moored buoy observing networks

This research is motivated by the practical requirements in the sustainable deployment of ocean moored buoy observing networks. Ocean moored buoys play an important role in the global marine environment monitoring. Ocean buoy station layout planning is a typical multiple-objective spatial optimization problem that aims to reduce the spatial correlation of buoy stations and improve their spatial monitoring efficiency. In this paper, we develop a multi-objective mathematical model for allocating ocean buoy stations (MOLMofOBS) based on Tobler’s first law of geography. A spatial neighborhood model based on a Voronoi diagram is built to represent the spatial proximity of distributed buoy stations and delimit the effective monitoring region of every station. Then, a heuristic method based on a multiple-objective particles swarm optimization (MOPSO) algorithm is developed to calculate the MOLMofOBS via a dynamic inertia weight strategy. Meanwhile, a series of experiments is conducted to verify the efficiency of the proposed model and algorithms in solving single- and multiple-buoy station location problems. Finally, an interactive portal is developed in the Cyberinfrastructure environment to provide decision-making services for online real-time planning of the ocean buoy station locations. The work reported in this paper will provide spatial decision-making support for the sustainable development of ocean buoy observing networks.

Exploring the relationships between perceived neighborhood boundaries and street network orientation

AbstractThe neighborhood is a core unit of analysis in urban research, planning, and policy‐making. However, perceptual and historical processes oftentimes result in neighborhoods that are not tied to sub‐urban jurisdictions. For instance, historic neighborhoods might lack official spatial definitions, hampering neighborhood‐based tasks in local offices. In this case, urban practitioners can benefit from readily available spatial proxies, such as the local street network. In this study, we conducted an exploratory analysis that combines neighborhood mapping and street network modeling. By retrieving participants' sketched boundaries and quantifying spatial orientations of sketched polygons and local network patterns, we were able to measure and compare the relationships between the urban fabric and the perceived extents of two historic neighborhoods in Lisbon, Portugal. The results provided insights for enhancing existing definitions of non‐official neighborhoods, outlining new urban districts as well as for discussions about the role of the urban form in shaping people's perceptions.

Implementing a deep-learning model using Google street view to combine social and physical indicators of gentrification

While physical changes have been empirically recognised as a fundamental component of neighbourhood change, data and modelling constraints have limited the quantification of these indicators. Recently however, the proliferation of Big Data and advancements in deep learning (DL) techniques have enabled mass image processing. It is in this context that we build a Siamese Convolutional Neural Network using Google Street View (GSV) images to detect upgrades to properties as evidence of gentrification. The model achieves 84.8% test accuracy and 74.6% AUC. Building upgrades detected using the model are mapped using Kernel Density Estimation (KDE) and validated against Development Approvals. The DL GSV model is subsequently combined with the socioeconomic-based predictions of seven gentrifying suburbs from a prior Sydney-based gentrification study. The maps of predicted social change are validated against the spatial patterns of building upgrades detected by the DL GSV model. Of the five suburbs with sufficient data, the socioeconomic trends were affirmed with physical indicators of gentrification in four and questioned in one. The paper provides the first machine learning approach to combine social and physical indicators of gentrification. This automated, multi-dimensional approach enables the distinction of gentrification from other forms of neighbourhood change, helping to develop a more comprehensive understanding of gentrification occurring within a city.

NMπ 2.0: Software update to minimize the risk of false positives among determinants of reproductive success.

In plant populations, parentage analysis helps understand factors shaping individual reproductive success. However, estimating reproductive success determinants based on parentage counts requires decoupling the effects of individual fecundity and propagule dispersal. The neighbourhood model implemented in the NMπ software provides a standard solution for this problem based on the fixed-effects regression-like approach. Nonetheless, it has been recently shown that the method is prone to false discoveries when important fecundity determinants are omitted. To account for the unexplained variance in fecundity, the Bayesian approach was developed based on the new model (the hierarchical neighbourhood model; HNM). Here, I present the NMπ software update that allows the HNM approach to be used in the framework of a friendly interface. More importantly, the HNM approach is now made available for both dispersed (seedlings) and nondispersed (seeds with known mothers) progeny data. The Bayesian approach, among others, selects significant fecundity determinants, estimates the proportion of variance in reproductive potential explained by selected determinants (R2 ), and provides individual female and male fecundity values. Although the software was designed to handle microsatellite marker data, a solution is proposed for large sets of single nucleotide polymorphisms. The program can be run on Windows (using either a terminal or a graphical interface) as well as (using a terminal) on Linux, or macOS platforms. In any case, NMπ can utilize multicore processors to speed up the analysis. The updated package containing the code, the executable file, the user manual, and example data is available at https://www.ukw.edu.pl/pracownicy/plik/igor_chybicki/3694/.

Individual, interpersonal and environmental correlates of sedentary behaviours in adults with intellectual disabilities.

Sedentary behaviours have adverse health outcomes and adults with intellectual disabilities are at a higher risk of unhealthy lifestyle behaviours. The lack of knowledge relating to sedentary behaviours in adults with intellectual disabilities has impeded the development of effective interventions. This study aimed to investigate individual, interpersonal and environmental correlates that are associated with sedentary behaviours in adults with intellectual disabilities. A secondary analysis of data from The UK Household Longitudinal Study (Understanding Society; collected 2011-2013) was conducted. Twenty-two predictor variables were included in a stepwise logistic regression, with TV hours during weekdays (≤3 and >3h/day) used as a proxy for sedentary behaviours. A sample of 266 adults, with mean age of 37.9 and range from 18 to 49years old, with intellectual disabilities were identified. Because 63.9% were female, 62.4% had children and 28.2% were employed, the sample is likely to be most representative of more able adults with intellectual disabilities. A significant interaction term between having children and neighbourhood status was found in the initial model so separate models for good and poor-quality neighbourhoods are reported. Having children only had a significant effect to lower the odds of high TV time among participants living in good quality neighbourhoods (OR 0.10, 95% CI 0.03, 0.25). However, for people living in poor quality neighbourhoods it was better quality leisure services that was associated with lower odds of high TV time (OR 0.48, 95% CI 0.23, 0.90). Being employed only significantly reduced the odds of high TV time in the good quality neighbourhood model (OR 0.35, 95% CI 0.12, 0.78). These effects highlight the importance of environmental effects on lifestyle behaviours of adults with intellectual disabilities. Future research should aim to expand our understanding of environmental effects on the sedentary behaviours and other lifestyle behaviours of adults with intellectual disabilities.

It's in the social network: The Social Neighbourhood model to unravel local social structures for liveable and safe neighbourhoods

Fast growth of cities decreases the quality of life in these places. In response, Municipalities install policies aiming to improve local livability. While literature suggests social structures to have a defining impact on policy effectiveness, current evaluation metrics are not able to take this into account. This paper presents the Social Neighbourhood model, an agent-based model used to simulate and explore how livability changes in a neighbourhood given various social structures and policies. The model is applied to a neighbourhood in The Hague, Netherlands. The main result of the modelling experiments is that social structures have a very strong influence on whether or not a policy to improve livability is effective. Three hypotheses, concerning this relationship between social structures, livability, and policy interventions are drawn up as a starting point for future research.

L-АЛГОРИТМ МОДЕЛИРОВАНИЯ ПЕРЕМЕННЫХ СОСТОЯНИЯ СТАДИИ ДИФФУЗИИ ПРОИЗВОДСТВА САХАРА

The article describes an algorithm for recurrent hierarchical identification of a regression model for the diffusion stage temperature mode in sugar production based on the trichotomy of residuals of intermediate models (L-algorithm). The algorithm uses a scheme that makes it possible to develop a neighborhood model while identifying the hierarchical partitioning of input data. This algorithm makes it possible to improve the approximation quality of some initial analytical model based on the analysis of its residuals and subsequent recurrent identification of additional terms corresponding to the hierarchical partitioning of the domain of model input definition (input tuples).

Smart neighborhood as a sustainable neighborhood: a comparative study of Al-Ghadeer village (Najaf-Iraq)

The concept of neighbourhood planning has emerged as a new trend in our community and the emphasis has been of utmost importance on districts in cities as the basic independent units of metropolitans in a way similar to that of community formation. Improving neighbourhood structure can improve the standards of living through the introduction of a smart neighbourhood model, making it more suitable for being walked and enhancing the life quality of residents that leading to the upgrading of communal and territory sketch. This work focuses on the problems where neighbourhoods lack aspects that allow the use of technology and affect workable distance and social relations, which are the main pillars of liveability. The hypothesis was to evaluate the neighbourhoods constructed by Najaf Investment Commission, whether the standards for the neighbourhood were applied according to the standards of the United Nations Organization (UN-Habitat) using gap analysis, The study found a gap between the standards adopted in the planning and design of the studied residential complexes (Al Ghadeer Residential Complex) and the criteria and indicators for planning smart neighbourhood model, particularly in the effectiveness of traffic network and the occupied residential area which was greater than the commercial model and less than the specified standards.

Dimensionality-Transformed Remote Sensing Data Application to Map Soil Salinization at Lowlands of the Syr Darya River

The problem of saving soil resources and their reclamation measures under current climate change conditions attracts the world community’s close attention. It is relevant in the Syr Darya River’s lowlands, where the secondary soil salinization processes have intensified. The demand for robust methods to assess soil salinity is high, and the primary purpose of this study was to develop a quantitative analysis method for soil salinity estimation. We found a correspondence between the sum of salts in a topsoil layer to the Landsat 8 data in the Tasseled cap transformation of the image values. After testing several methods, we built a prediction model. The K-nearest neighborhood (KNN) model with a coefficient of determination equal to 0.96 using selected predictors proved to be the most appropriate for soil salinity assessment. We also performed a quantitative assessment of soil salinity. A significant increase in a salt-affected area and the mean soil sum expressing an intensification of secondary soil salinization from 2018 to 2021 was found. The increasing temperature values, decreasing soil moisture, and agricultural use affect the extension of salt-affected ground areas in the study area. Thus, the soil moisture trend in the Qazaly irrigation zone is negative and declining, with the highest peaks in early spring. The maximum temperature has a mean value of 15.6 °C (minimum = −15.1 °C, maximum = 37.4 °C) with an increasing trend. These parameters are evidence of climate change that also affects soil salinization. PCA transformation of the Landsat-8 satellite images helped to remove redundant spectral information from multiband datasets and map soil salinity more precisely. This approach simultaneously extends mapping opportunities involving visible and invisible bands and results in a smaller dataset.

Effects of biotic interactions on plant fecundity depend on spatial and functional structure of communities and time since disturbance

Abstract Biotic interactions in plant communities affect individual fitness and community dynamics. Interactions between plants vary in space, over time and with organisational complexity. Yet it is challenging to quantify temporal, spatial and functional determinants of different types of interactions between long‐lived perennial plant species and their effect on lifetime fecundity. We studied how plant–plant, pollinator‐ and seed predator‐mediated interactions affect year‐to‐year variation in three fecundity components (cone production, seed set and seed survival) during post‐fire recovery. Age‐stratified data on the three fecundity components were collected in 19 even‐aged communities comprising 20 serotinous Protea shrub species in the South African Fynbos. We analyse data on these fecundity components with neighbourhood models to infer the sign and strength of interactions throughout post‐disturbance recovery, the neighbour plant traits that shape them and the spatial scale at which interactions take place. For each fecundity component, these models describe how neighbourhood effects change over time and with spatial distance between plants. We then predicted neighbourhood effects on individual fecundity components and cumulative reproductive output at different post‐fire stand ages for each focal plant. Competitive effects on cone production and seed set increased with post‐fire stand age as biomass and floral resources for pollinators build up. In contrast, neighbourhood effects on seed survival were weak throughout post‐disturbance recovery. Plant–plant interactions were shaped by neighbour traits related to resource acquisition, whereas animal‐mediated interactions depended on neighbour traits related to resource availability for pollinators and seed predators. The spatial scale of the interactions increased from plant–plant over predator‐mediated to pollinator‐mediated interactions. The joint effect of these interactions on cumulative reproductive output caused the proportion of focal plants experiencing competition to increase with time since fire. Synthesis. We show that temporal changes in biotic interactions throughout post‐disturbance recovery of perennial plant communities depend on functional traits and can be integrated into neighbourhood effects on lifetime fecundity. Studying the temporal, spatial and functional determinants of neighbourhood effects on lifetime fecundity is important for predicting not only individual plant fitness but also population and community dynamics in changing environments.

An improved decision tree algorithm based on variable precision neighborhood similarity

The decision tree algorithm has been widely used in data mining and machine learning due to its high accuracy, low computational cost and high interpretability. However, when dealing with the continuous data, the classical decision tree algorithm needs to replace continuous attributes with discretized attributes by the strategy of discretization. Discretization may cause a loss of information structure, which will affect the performance of classification. To tackle this problem, many researchers have proposed different decision tree methods based on variable precision neighborhood rough sets. However, these methods do not consider the geometric structure of neighborhood systems, which may lead to a contradiction in the transitivity of the equivalence relation. In this paper, we first define a novel neighborhood geometric similarity in a neighborhood system from the perspective of geometry. Second, by combining the neighborhood geometric similarity and the neighborhood algebraic similarity, we propose four new kinds of neighborhood similarities, which can solve the contradictory transitivity of the equivalence relation. Third, a variable precision neighborhood rough set model is constructed using the new similarities, and a novel decision tree algorithm is proposed based on this model, where the degree of attribute dependence is used as the partition measure. Experimental results on 14 selected datasets from the UCI Machine Learning Repository show that our algorithm is effective. The average accuracy of our algorithm is over 90%, which is 10% higher than the classical decision tree algorithms, and the number of leaf nodes increases slightly.

DynDeepDRIM: a dynamic deep learning model to infer direct regulatory interactions using time-course single-cell gene expression data.

Time-course single-cell RNA sequencing (scRNA-seq) data have been widely used to explore dynamic changes in gene expression of transcription factors (TFs) and their target genes. This information is useful to reconstruct cell-type-specific gene regulatory networks (GRNs). However, the existing tools are commonly designed to analyze either time-course bulk gene expression data or static scRNA-seq data via pseudo-time cell ordering. A few methods successfully utilize the information from multiple time points while also considering the characteristics of scRNA-seq data. We proposed dynDeepDRIM, a novel deep learning model to reconstruct GRNs using time-course scRNA-seq data. It represents the joint expression of a gene pair as an image and utilizes the image of the target TF-gene pair and the ones of the potential neighbors to reconstruct GRNs from time-course scRNA-seq data. dynDeepDRIM can effectively remove the transitive TF-gene interactions by considering neighborhood context and model the gene expression dynamics using high-dimensional tensors. We compared dynDeepDRIM with six GRN reconstruction methods on both simulation and four real time-course scRNA-seq data. dynDeepDRIM achieved substantially better performance than the other methods in inferring TF-gene interactions and eliminated the false positives effectively. We also applied dynDeepDRIM to annotate gene functions and found it achieved evidently better performance than the other tools due to considering the neighbor genes.