A novel approach for short-term load forecasting using support vector machines

A support vector machine (SVM) modeling approach for short-term load forecasting is proposed. The SVM learning scheme is applied to the power load data, forcing the network to learn the inherent internal temporal property of power load sequence. We also study the performance when other related input variables such as temperature and humidity are considered. The performance of our proposed SVM modeling approach has been tested and compared with feed-forward neural network and cosine radial basis function neural network approaches. Numerical results show that the SVM approach yields better generalization capability and lower prediction error compared to those neural network approaches.     

Importance of low-flow and high-flow characteristics to restoration of riparian vegetation along rivers in arid south-western United States

1. Riparian vegetation in dry regions is influenced by low‐flow and high‐flow components of the surface and groundwater flow regimes. The duration of no‐flow periods in the surface stream controls vegetation structure along the low‐flow channel, while depth, magnitude and rate of groundwater decline influence phreatophytic vegetation in the floodplain. Flood flows influence vegetation along channels and floodplains by increasing water availability and by creating ecosystem disturbance. 2. On reference rivers in Arizona's Sonoran Desert region, the combination of perennial stream flows, shallow groundwater in the riparian (stream) aquifer, and frequent flooding results in high plant species diversity and landscape heterogeneity and an abundance of pioneer wetland plant species in the floodplain. Vegetation changes on hydrologically altered river reaches are varied, given the great extent of flow regime changes ranging from stream and aquifer dewatering on reaches affected by stream diversion and groundwater pumping to altered timing, frequency, and magnitude of flood flows on reaches downstream of flow‐regulating dams. 3. As stream flows become more intermittent, diversity and cover of herbaceous species along the low‐flow channel decline. As groundwater deepens, diversity of riparian plant species (particularly perennial species) and landscape patches are reduced and species composition in the floodplain shifts from wetland pioneer trees (Populus, Salix) to more drought‐tolerant shrub species including Tamarix (introduced) and Bebbia. 4. On impounded rivers, changes in flood timing can simplify landscape patch structure and shift species composition from mixed forests composed of Populus and Salix, which have narrow regeneration windows, to the more reproductively opportunistic Tamarix. If flows are not diverted, suppression of flooding can result in increased density of riparian vegetation, leading in some cases to very high abundance of Tamarix patches. Coarsening of sediments in river reaches below dams, associated with sediment retention in reservoirs, contributes to reduced cover and richness of herbaceous vegetation by reducing water and nutrient‐holding capacity of soils. 5. These changes have implications for river restoration. They suggest that patch diversity, riparian plant species diversity, and abundance of flood‐dependent wetland tree species such as Populus and Salix can be increased by restoring fluvial dynamics on flood‐suppressed rivers and by increasing water availability in rivers subject to water diversion or withdrawal. On impounded rivers, restoration of plant species diversity also may hinge on restoration of sediment transport. 6. Determining the causes of vegetation change is critical for determining riparian restoration strategies. Of the many riparian restoration efforts underway in south‐western United States, some focus on re‐establishing hydrogeomorphic processes by restoring appropriate flows of surface water, groundwater and sediment, while many others focus on manipulating vegetation structure by planting trees (e.g. Populus) or removing trees (e.g. Tamarix). The latter approaches, in and of themselves, may not yield desired restoration outcomes if the tree species are indicators, rather than prime causes, of underlying changes in the physical environment.     

Knowledge extraction: a comparison between symbolic and connectionist methods

The use of a linguistic representation for expressing knowledge acquired by learning systems is an important issue as regards to user understanding. Under this assumption, and to make sure that these systems will be welcome and used, several techniques have been developed by the artificial intelligence community, under both the symbolic and the connectionist approaches. This work discusses and investigates three knowledge extraction techniques based on these approaches. The first two techniques, the C4.5 and CN2 symbolic learning algorithms, extract knowledge directly from the data set. The last technique, the TREPAN algorithm extracts knowledge from a previously trained neural network. The CN2 algorithm induces if...then rules from a given data set. The C4.5 algorithm extracts decision trees, although it can also extract ordered rules, from the data set. Decision trees are also the knowledge representation used by the TREPAN algorithm.     

Food web structure in riverine landscapes

1. Most research on freshwater (and other) food webs has focused on apparently discrete communities, in well-defined habitats at small spatial and temporal scales, whereas in reality food webs are embedded in complex landscapes, such as river corridors. Food web linkages across such landscapes may be crucial for ecological pattern and process, however. Here, we consider the importance of large scale influences upon lotic food webs across the three spatial dimensions and through time.
2. We assess the roles of biotic factors (e.g. predation, competition) and physical habitat features (e.g. geology, land-use, habitat fragmentation) in moulding food web structure at the landscape scale. As examples, external subsidies to lotic communities of nutrients, detritus and prey vary along the river corridor, and food web links are made and broken across the land–water interface with the rise and fall of the flood.
3. We identify several avenues of potentially fruitful research, particularly the need to quantify energy flow and population dynamics. Stoichiometric analysis of changes in C : N : P nutrient ratios over large spatial gradients (e.g. from river source to mouth, in forested versus agricultural catchments), offers a novel method of uniting energy flow and population dynamics to provide a more holistic view of riverine food webs from a landscape perspective. Macroecological approaches can be used to examine large-scale patterns in riverine food webs (e.g. trophic rank and species–area relationships). New multivariate statistical techniques can be used to examine community responses to environmental gradients and to assign traits to individual species (e.g. body-size, functional feeding group), to unravel the organisation and trophic structure of riverine food webs.     

Progress and opportunities in advancing near-term forecasting of freshwater quality

Near-term freshwater forecasts, defined as sub-daily to decadal future predictions of a freshwater variable with quantified uncertainty, are urgently needed to improve water quality management as freshwater ecosystems exhibit greater variability due to global change. Shifting baselines in freshwater ecosystems due to land use and climate change prevent managers from relying on historical averages for predicting future conditions, necessitating near-term forecasts to mitigate freshwater risks to human health and safety (e.g., flash floods, harmful algal blooms) and ecosystem services (e.g., water-related recreation and tourism). To assess the current state of freshwater forecasting and identify opportunities for future progress, we synthesized freshwater forecasting papers published in the past 5 years. We found that freshwater forecasting is currently dominated by near-term forecasts of water quantity and that near-term water quality forecasts are fewer in number and in the early stages of development (i.e., non-operational) despite their potential as important preemptive decision support tools. We contend that more freshwater quality forecasts are critically needed and that near-term water quality forecasting is poised to make substantial advances based on examples of recent progress in forecasting methodology, workflows, and end-user engagement. For example, current water quality forecasting systems can predict water temperature, dissolved oxygen, and algal bloom/toxin events 5 days ahead with reasonable accuracy. Continued progress in freshwater quality forecasting will be greatly accelerated by adapting tools and approaches from freshwater quantity forecasting (e.g., machine learning modeling methods). In addition, future development of effective operational freshwater quality forecasts will require substantive engagement of end users throughout the forecast process, funding, and training opportunities. Looking ahead, near-term forecasting provides a hopeful future for freshwater management in the face of increased variability and risk due to global change, and we encourage the freshwater scientific community to incorporate forecasting approaches in water quality research and management.     

Flood susceptibility mapping using extremely randomized trees for Assam 2020 floods

The year 2020 proved disastrous for the north eastern state of India, Assam. The state witnessed terrible floods in the midst of the pandemic. The current study aims to better understand the role played by various factors that contributed to the deluge. To this end, the current study undertakes a flood susceptibility mapping using a seldom employed decision tree based ensemble machine learning technique of extremely randomized trees (ERT). The model was trained and tested on a flood inventory superimposed with 14 flood influencing factors, namely slope, elevation, aspect, normalized difference vegetation index (NDVI), topographic wetness index (TWI), slope length, land use, geology, soil type, topographic roughness index (TRI), rainfall, distance from rivers, plan and profile curvature. The model was compared against other mapping techniques and produced an area under the receiver operating characteristic curve (AUC) of 0.901 outperforming others. The generated susceptibility map deduced the presence of low elevation, high rainfall and close proximity to rivers as major factors leading up to the disaster. It prophesizes a very high flood risk for approximately 18.32% of the study area concentrated in the northern and western part of the study region.     

Prediction of Signal Peptides Using Bio-Basis Function Neural Networks and Decision Trees

Signal peptide identification is of immense importance in drug design. Accurate identification of signal peptides is the first critical step to be able to change the direction of the targeting proteins and use the designed drug to target a specific organelle to correct a defect. Because experimental identification is the most accurate method, but is expensive and time-consuming, an efficient and affordable automated system is of great interest. In this article, we propose using an adapted neural network, called a bio-basis function neural network, and decision trees for predicting signal peptides. The bio-basis function neural network model and decision trees achieved 97.16% and 97.63% accuracy respectively, demonstrating that the methods work well for the prediction of signal peptides. Moreover, decision trees revealed that position P(1'), which is important in forming signal peptides, most commonly comprises either leucine or alanine. This concurs with the (P(3)-P(1)-P(1')) coupling model.     

Neural network control systems that learn to perform appropriately

Setting up a neural network with a learning algorithm that determines how it can best operate is an efficient way to formulate control systems for many engineering applications, and is often much more feasible than direct programming. This paper examines three important aspects of this approach: the details of the cost function that is used with the gradient descent learning algorithm, how the resulting system depends on the initial pre-learning connection weights, and how the resulting system depends on the pattern of learning rates chosen for the different components of the system. We explore these issues by explicit simulations of a toy model that is a simplified abstraction of part of the human oculomotor control system. This allows us to compare our system with that produced by human evolution and development. We can then go on to consider how we might improve on the human system and apply what we have learnt to control systems that have no human analogue.     

Physical and Economic Impacts of Sea-Level Rise and Low Probability Flooding Events on Coastal Communities

Conventionally flood mapping typically includes only a static water level (e.g. peak of a storm tide) in coastal flood inundation events. Additional factors become increasingly important when increased water-level thresholds are met during the combination of a storm tide and increased mean sea level. This research incorporates factors such as wave overtopping and river flow in a range of flood inundation scenarios of future sea-level projections for a UK case study of Fleetwood, northwest England. With increasing mean sea level it is shown that wave overtopping and river forcing have an important bearing on the cost of coastal flood events. The method presented converts inundation maps into monetary cost. This research demonstrates that under scenarios of joint extreme surge-wave-river events the cost of flooding can be increased by up to a factor of 8 compared with an increase in extent of up to a factor of 3 relative to “surge alone” event. This is due to different areas being exposed to different flood hazards and areas with common hazard where flood waters combine non-linearly. This shows that relying simply on flood extent and volume can under-predict the actual economic impact felt by a coastal community. Additionally, the scenario inundation depths have been presented as “brick course” maps, which represent a new way of interpreting flood maps. This is primarily aimed at stakeholders to increase levels of engagement within the coastal community.     

Comparisons of likelihood and machine learning methods of individual classification

Classification methods used in machine learning (e.g., artificial neural networks, decision trees, and k-nearest neighbor clustering) are rarely used with population genetic data. We compare different nonparametric machine learning techniques with parametric likelihood estimations commonly employed in population genetics for purposes of assigning individuals to their population of origin ("assignment tests"). Classifier accuracy was compared across simulated data sets representing different levels of population differentiation (low and high F(ST)), number of loci surveyed (5 and 10), and allelic diversity (average of three or eight alleles per locus). Empirical data for the lake trout (Salvelinus namaycush) exhibiting levels of population differentiation comparable to those used in simulations were examined to further evaluate and compare classification methods. Classification error rates associated with artificial neural networks and likelihood estimators were lower for simulated data sets compared to k-nearest neighbor and decision tree classifiers over the entire range of parameters considered. Artificial neural networks only marginally outperformed the likelihood method for simulated data (0-2.8% lower error rates). The relative performance of each machine learning classifier improved relative likelihood estimators for empirical data sets, suggesting an ability to "learn" and utilize properties of empirical genotypic arrays intrinsic to each population. Likelihood-based estimation methods provide a more accessible option for reliable assignment of individuals to the population of origin due to the intricacies in development and evaluation of artificial neural networks.     

Network Environment and Financial Risk Using Machine Learning and Sentiment Analysis

Under the network environment, the trading volume and asset price of a financial commodity or instrument are affected by various complicated factors. Machine learning and sentiment analysis provide powerful tools to collect a great deal of data from the website and retrieve useful information for effectively forecasting financial risk of associated companies. This article studies trading volume and asset price risk when sentimental financial information data are available using both sentiment analysis and popular machine learning approaches: artificial neural network (ANN) and support vector machine (SVM). Nonlinear GARCH-based mining models are developed by integrating GARCH (generalized autoregressive conditional heteroskedasticity) theory and ANN and SVM. Empirical studies in the U.S. stock market show that the proposed approach achieves favorable forecast performances. GARCH-based SVM outperforms GARCH-based ANN for volatility forecast, whereas GARCH-based ANN achieves a better forecast result for the volatility trend. Results also indicate a strong correlation between information sentiment and both trading volume and asset price volatility.     

Population networks: a large-scale framework for modelling cortical neural networks

Artificial neural networks are usually built on rather few elements such as activation functions, learning rules, and the network topology. When modelling the more complex properties of realistic networks, however, a number of higher-level structural principles become important. In this paper we present a theoretical framework for modelling cortical networks at a high level of abstraction. Based on the notion of a population of neurons, this framework can accommodate the common features of cortical architecture, such as lamination, multiple areas and topographic maps, input segregation, and local variations of the frequency of different cell types (e.g., cytochrome oxidase blobs). The framework is meant primarily for the simulation of activation dynamics; it can also be used to model the neural environment of single cells in a multiscale approach. Received: 9 January 1996 / Accepted in revised form: 24 July 1996     

Improved Neural Networks with Random Weights for Short-Term Load Forecasting

An effective forecasting model for short-term load plays a significant role in promoting the management efficiency of an electric power system. This paper proposes a new forecasting model based on the improved neural networks with random weights (INNRW). The key is to introduce a weighting technique to the inputs of the model and use a novel neural network to forecast the daily maximum load. Eight factors are selected as the inputs. A mutual information weighting algorithm is then used to allocate different weights to the inputs. The neural networks with random weights and kernels (KNNRW) is applied to approximate the nonlinear function between the selected inputs and the daily maximum load due to the fast learning speed and good generalization performance. In the application of the daily load in Dalian, the result of the proposed INNRW is compared with several previously developed forecasting models. The simulation experiment shows that the proposed model performs the best overall in short-term load forecasting.     

Brain networks of explicit and implicit learning

Are explicit versus implicit learning mechanisms reflected in the brain as distinct neural structures, as previous research indicates, or are they distinguished by brain networks that involve overlapping systems with differential connectivity? In this functional MRI study we examined the neural correlates of explicit and implicit learning of artificial grammar sequences. Using effective connectivity analyses we found that brain networks of different connectivity underlie the two types of learning: while both processes involve activation in a set of cortical and subcortical structures, explicit learners engage a network that uses the insula as a key mediator whereas implicit learners evoke a direct frontal-striatal network. Individual differences in working memory also differentially impact the two types of sequence learning.     

Neuro-fuzzy decision trees

Fuzzy decision trees are powerful, top-down, hierarchical search methodology to extract human interpretable classification rules. However, they are often criticized to result in poor learning accuracy. In this paper, we propose Neuro-Fuzzy Decision Trees (N-FDTs); a fuzzy decision tree structure with neural like parameter adaptation strategy. In the forward cycle, we construct fuzzy decision trees using any of the standard induction algorithms like fuzzy ID3. In the feedback cycle, parameters of fuzzy decision trees have been adapted using stochastic gradient descent algorithm by traversing back from leaf to root nodes. With this strategy, during the parameter adaptation stage, we keep the hierarchical structure of fuzzy decision trees intact. The proposed approach of applying backpropagation algorithm directly on the structure of fuzzy decision trees improves its learning accuracy without compromising the comprehensibility (interpretability). The proposed methodology has been validated using computational experiments on real-world datasets.     

A discriminative method for family-based protein remote homology detection that combines inductive logic programming and propositional models

Background  

Remote homology detection is a hard computational problem. Most approaches have trained computational models by using either full protein sequences or multiple sequence alignments (MSA), including all positions. However, when we deal with proteins in the "twilight zone" we can observe that only some segments of sequences (motifs) are conserved. We introduce a novel logical representation that allows us to represent physico-chemical properties of sequences, conserved amino acid positions and conserved physico-chemical positions in the MSA. From this, Inductive Logic Programming (ILP) finds the most frequent patterns (motifs) and uses them to train propositional models, such as decision trees and support vector machines (SVM).     

Machine learning methods without tears: a primer for ecologists

Machine learning methods, a family of statistical techniques with origins in the field of artificial intelligence, are recognized as holding great promise for the advancement of understanding and prediction about ecological phenomena. These modeling techniques are flexible enough to handle complex problems with multiple interacting elements and typically outcompete traditional approaches (e.g., generalized linear models), making them ideal for modeling ecological systems. Despite their inherent advantages, a review of the literature reveals only a modest use of these approaches in ecology as compared to other disciplines. One potential explanation for this lack of interest is that machine learning techniques do not fall neatly into the class of statistical modeling approaches with which most ecologists are familiar. In this paper, we provide an introduction to three machine learning approaches that can be broadly used by ecologists: classification and regression trees, artificial neural networks, and evolutionary computation. For each approach, we provide a brief background to the methodology, give examples of its application in ecology, describe model development and implementation, discuss strengths and weaknesses, explore the availability of statistical software, and provide an illustrative example. Although the ecological application of machine learning approaches has increased, there remains considerable skepticism with respect to the role of these techniques in ecology. Our review encourages a greater understanding of machin learning approaches and promotes their future application and utilization, while also providing a basis from which ecologists can make informed decisions about whether to select or avoid these approaches in their future modeling endeavors.     

Reconstructing muscle activation during normal walking: a comparison of symbolic and connectionist machine learning techniques

One symbolic (rule-based inductive learning) and one connectionist (neural network) machine learning technique were used to reconstruct muscle activation patterns from kinematic data measured during normal human walking at several speeds. The activation patterns (or desired outputs) consisted of surface electromyographic (EMG) signals from the semitendinosus and vastus medialis muscles. The inputs consisted of flexion and extension angles measured at the hip and knee of the ipsilateral leg, their first and second derivatives, and bilateral foot contact information. The training set consisted of data from six trials, at two different speeds. The testing set consisted of data from two additional trials (one at each speed), which were not in the training set. It was possible to reconstruct the muscular activation at both speeds using both techniques. Timing of the reconstructed signals was accurate. The integrated value of the activation bursts was less accurate. The neural network gave a continuous output, whereas the rule-based inductive learning rule tree gave a quantised activation level. The advantage of rule-based inductive learning was that the rules used were both explicit and comprehensible, whilst the rules used by the neural network were implicit within its structure and not easily comprehended. The neural network was able to reconstruct the activation patterns of both muscles from one network, whereas two separate rule sets were needed for the rule-based technique. It is concluded that machine learning techniques, in comparison to explicit inverse muscular skeletal models, show good promise in modelling nearly cyclic movements such as locomotion at varying walking speeds. However, they do not provide insight into the biomechanics of the system, because they are not based on the biomechanical structure of the system.     

River network saturation concept: factors influencing the balance of biogeochemical supply and demand of river networks

River networks modify material transfer from land to ocean. Understanding the factors regulating this function for different gaseous, dissolved, and particulate constituents is critical to quantify the local and global effects of climate and land use change. We propose the River Network Saturation (RNS) concept as a generalization of how river network regulation of material fluxes declines with increasing flows due to imbalances between supply and demand at network scales. River networks have a tendency to become saturated (supply???demand) under higher flow conditions because supplies increase faster than sink processes. However, the flow thresholds under which saturation occurs depends on a variety of factors, including the inherent process rate for a given constituent and the abundance of lentic waters such as lakes, ponds, reservoirs, and fluvial wetlands within the river network. As supply increases, saturation at network scales is initially limited by previously unmet demand in downstream aquatic ecosystems. The RNS concept describes a general tendency of river network function that can be used to compare the fate of different constituents among river networks. New approaches using nested in situ high-frequency sensors and spatially extensive synoptic techniques offer the potential to test the RNS concept in different settings. Better understanding of when and where river networks saturate for different constituents will allow for the extrapolation of aquatic function to broader spatial scales and therefore provide information on the influence of river function on continental element cycles and help identify policy priorities.