Presence Of Many Features Research Articles

Predicting UK Housing Price using Machine Learning Algorithms

The development of reliable predictive algorithm for house price as the housing market is a stand-out among the most involved regarding valuing the price and continues to fluctuate, is constantly a need for socio-economic advancement and welfare of citizen. In this paper, we develop machine learning algorithms for forecasting UK housing Price, and find an optimal algorithm that forecasts housing price accurately on the premises of the presence of many features or covariates. After applying correlation analysis to remove correlated variables in order to avoid multicollinearity, thereby increasing the statistical power, a novel method of using regression analysis to first of all understand and select statistically significant features for the various regions in England based on North South divide is adopted. These features are then used in the machine learning algorithm to further increase the statistical power of the algorithm, increase the level of accuracy for each of them and ultimately increase the predictive values for the algorithms.The model construction involves 3 stages: 1- correlation analysis to identify and remove correlated variables thereby avoiding multicollinearity and increasing the statistical power of the linear regression, 2 - using linear regression to determine variables that are statistically significant and 3 - building the machine learning algorithms based on the variables that are statistically significant from the linear regression. A comprehensive dataset of UK Paid housing Price from 2010 to 2019 was linked to a number of other datasets to generate a total 21 variables or features used for the models. Catboost, Gradient Boosting, Bagging, Random Forest, Extra Tree all achieved the excellent model’s performance result in all the regions considered. The comparison of the seven models showed that Extra Tree algorithm consistently achieved the best performance in term of level of accuracy in all the regions. K-Nearest Neighbours (KNN) is the only algorithm with less than 50% level of accuracy. Noticeably, the regions considered had varying or differing insignificant variables, implying that although many variables are common (statistically significant) to all the regions, there are regional differences and impact when modelling or predicting housing prices. This study validates the practicability of developing a machine learning methodology for the prediction of housing price. This research offers a reference for future house price prediction based on machine learning.© 2024 The Author(s). Published by RITHA Publishing. This article is distributed under the terms of the license CC-BY 4.0., which permits any further distribution in any medium, provided the original work is properly cited.

Assessing the reliability of microbial bioerosion features in burnt bones: A novel approach using feature-labelling in histotaphonomical analysis

ObjectivesRecent histotaphonomic studies have focused on the presence of features thought to be caused either by bacteria (microscopic focal destruction/MFD and cyanobacterial tunnelling) or fungal (Wedl tunnelling types 1 and 2) attack on unburnt bone. Identifying these characteristics on burnt bones could indicate the state of decomposition before burning, with important repercussions for both archaeological and forensic contexts. Materials and MethodsFleshed pig (Sus scrofa, N = 25) tibiae were left exposed on a field, then collected at 14-, 34-, 91-, 180-, 365-day intervals before being burnt in an outdoor fire (≤750 °C). Fresh (fleshed) legs (N = 10) acted as unburnt and burnt controls. Thin sections were examined using transmitted light microscopy and backscattered scanning electron microscopy. Diagenetic traits were quantitatively and systematically assessed by a novel data labelling application developed for this study. ResultsFeatures meeting the published characteristics of microbial bioerosion (‘Wedl tunnelling’, ‘lamellate’ and ‘budded MFD’) were significantly correlated with time since deposition on the unburnt bones. Only budded MFD increased significantly over time in the burnt groups. However, the presence features meeting the published characteristics of Wedl 2 tunnelling were present on the fresh burnt bones, indicating that they are an artefact. DiscussionThe presence of many features seemingly indistinguishable from those caused by bioerosion on the freshly burnt control bones suggests that burning is not only able to conceal features thought to be the result of bioerosion but can produce them as well. Thus, such features are not a reliable indication of bioerosion. Budded MFD may be a viable indicator but more research is required.

Mitigating the feature fatigue effect for smart products through digital servitization

We analyze the problem of a manufacturer developing smart products and selling them to consumers who can be subject to the feature fatigue effect. Feature fatigue occurs with the presence of many features (smart products are typically over featured goods) that make consumers excited during the purchasing phase while becoming useless and complicated during the consumption phase. The latter negative implications induce frustration and regret among consumers, who can associate these negative feelings and effects with the brand, leading to a detrimental impact on its value. We capture the effect of feature fatigue on brand value through a dynamic model in which the stock of goodwill decreases with excessive investments in technology, which are the main drivers to create fashionable, attractive, and appealing features. We propose a digital servitization strategy to mitigate these negative effects and discover that firms offering digital servitization can end up economically worse off because feature fatigue plays a key role in the optimal strategies. In contrast, firms are economically better off and resolve feature fatigue when digital servitization is independent of other strategies. We identify cases in which digital servitization does not provide any benefit, when it completely nullifies feature fatigue effect, and when its activation complements either short-term or long-term strategies. Our findings are then supported and proofed empirically to increase their generizability.

To choose the use of 2d or 3d images in rehabilitation: review

Different properties of displays, features of visual perception of three-dimensional images and other conditions, probably affect the effectiveness of motor rehabilitation when using a visual feedback channel and virtual reality technology. A brief review presents the latest publications on a choice of 2D or 3D displays. It is concluded that the presence of many features not only creates difficulties in comparing the effects of using various equipment, but also provides the potential for targeted display selection for a particular rehabilitation task.

Spectral-Spatial Classification of Hyperspectral Images: Three Tricks and a New Learning Setting

Spectral-spatial classification of hyperspectral images has been the subject of many studies in recent years. When there are only a few labeled pixels for training and a skewed class label distribution, this task becomes very challenging because of the increased risk of overfitting when training a classifier. In this paper, we show that in this setting, a convolutional neural network with a single hidden layer can achieve state-of-the-art performance when three tricks are used: a spectral-locality-aware regularization term and smoothing- and label-based data augmentation. The shallow network architecture prevents overfitting in the presence of many features and few training samples. The locality-aware regularization forces neighboring wavelengths to have similar contributions to the features generated during training. The new data augmentation procedure favors the selection of pixels in smaller classes, which is beneficial for skewed class label distributions. The accuracy of the proposed method is assessed on five publicly available hyperspectral images, where it achieves state-of-the-art results. As other spectral-spatial classification methods, we use the entire image (labeled and unlabeled pixels) to infer the class of its unlabeled pixels. To investigate the positive bias induced by the use of the entire image, we propose a new learning setting where unlabeled pixels are not used for building the classifier. Results show the beneficial effect of the proposed tricks also in this setting and substantiate the advantages of using labeled and unlabeled pixels from the image for hyperspectral image classification.

Semi‐automated detection of eagle nests: an application of very high‐resolution image data and advanced image analyses to wildlife surveys

AbstractVery high‐resolution (VHR) image data, including from unmanned aerial vehicle (UAV) platforms, are increasingly acquired for wildlife surveys. Animals or structures they build (e.g. nests) can be photointerpreted from these images, however, automated detection is required for more efficient surveys. We developed semi‐automated analyses to map white‐bellied sea eagle (Haliaeetus leucogaster) nests in VHR aerial photographs of the Houtman Abrolhos Islands, Western Australia, an important breeding site for many seabird species. Nest detection is complicated by high environmental heterogeneity at the scale of nests (~1–2 m), the presence of many features that resemble nests and the variability of nest size, shape and context. Finally, the rarity of nests limits the availability of training data. These challenges are not unique to wildlife surveys and we show how they can be overcome by an innovative integration of object‐based image analyses (OBIA) and the powerful machine learning one‐class classifier Maxent. Maxent classifications using features characterizing object texture, geometry and neighborhood, along with limited object color information, successfully identified over 90% of high quality nests (most weathered and unusually shaped nests were also detected, but at a slightly lower rate) and labeled <2% of objects as candidate nests. Although this overestimates the occurrence of nests, the results can be visually screened to rule out all but the most likely nests in a process that is simpler and more efficient than manual photointerpretation of the full image. Our study shows that semi‐automated image analyses for wildlife surveys are achievable. Furthermore, the developed strategies have broad relevance to image processing applications that seek to detect rare features differing only subtly from a heterogeneous background, including remote sensing of archeological remains. We also highlight solutions to maximize the use of imperfect or uncalibrated image data, such as some UAV‐based imagery and the growing body of VHR imagery available in Google Earth and other virtual globes.

Latest Progress in Microsporidian Genome Research

Microsporidia are obligate intracellular pathogens of medical and ecological importance whose genomes have been studied extensively over the last decade. Such studies have focused on the remarkably reduced gene content that characterizes all known species, and some have unraveled the mechanisms that are involved in their extreme genome compaction. In the last year, a large number of new genome sequences from several divergent members of the group have been finally released and analyzed, and these have revealed the presence of many features that were previously unsuspected to exist within the group. This study aims to shortly review the most recent progress in the field of microsporidian genomics, highlighting the importance of the most recently released genome data for our understanding of the biology and evolution of this important group of parasites.

Metastats: an improved statistical method for analysis of metagenomic data

Metagenomic studies were originally focused on exploratory/validation projects but are rapidly being applied in a clinical setting. In this setting, researchers are interested in finding characteristics of the microbiome that correlate with the clinical status of the corresponding sample. Comparatively few computational/statistical tools have been developed that can assist in this process. Rather, most developments in the metagenomics community have focused on methods that compare samples as a whole. Specifically, the focus has been on developing robust methods for determining the level of similarity or difference between samples, rather than on identifying the specific characteristics that distinguish different samples from each other. Metastats [1] was the first statistical method developed specifically to address the questions asked in clinical studies. Metastats allows a comparison of metagenomic samples (represented as counts of individual features such as organisms, genes and functional groups) from two treatment populations (for example, healthy versus disease) and identifies those features that statistically distinguish the two populations. Here, we present major improvements to the Metastats software and the underlying statistical methods. First, we describe new approaches for data normalization that allow a more accurate assessment of differential abundance by reducing the covariance between individual features implicitly introduced by the traditionally used ratio-based normalization. These normalization techniques are also of interest for time-series analyses or in the estimation of microbial networks. A second extension of Metastats is a mixed-model zero-inflated Gaussian distribution that allows Metastats to account for a common characteristic of metagenomic data: the presence of many features with zero counts owing to undersampling of the community. The number of ‘missing features’ (zero counts) correlates with the amount of sequencing performed, thereby biasing abundance measurements and the differential abundance statistics derived from them. Using simulated and real data, we show that these methods significantly improve the accuracy of Metastats. We also describe the addition of several new statistical tests to our code (including presence/absence and the corresponding odds ratio, and penetrance calculations) that improve the usability of our software in clinical practice.

Metastats: an improved statistical method for analysis of metagenomic data

Metagenomic studies were originally focused on exploratory/validation projects but are rapidly being applied in a clinical setting. In this setting, researchers are interested in finding characteristics of the microbiome that correlate with the clinical status of the corresponding sample. Comparatively few computational/statistical tools have been developed that can assist in this process. Rather, most developments in the metagenomics community have focused on methods that compare samples as a whole. Specifically, the focus has been on developing robust methods for determining the level of similarity or difference between samples, rather than on identifying the specific characteristics that distinguish different samples from each other. Metastats [1] was the first statistical method developed specifically to address the questions asked in clinical studies. Metastats allows a comparison of metagenomic samples (represented as counts of individual features such as organisms, genes and functional groups) from two treatment populations (for example, healthy versus disease) and identifies those features that statistically distinguish the two populations. Here, we present major improvements to the Metastats software and the underlying statistical methods. First, we describe new approaches for data normalization that allow a more accurate assessment of differential abundance by reducing the covariance between individual features implicitly introduced by the traditionally used ratio-based normalization. These normalization techniques are also of interest for time-series analyses or in the estimation of microbial networks. A second extension of Metastats is a mixed-model zero-inflated Gaussian distribution that allows Metastats to account for a common characteristic of metagenomic data: the presence of many features with zero counts owing to undersampling of the community. The number of ‘missing features’ (zero counts) correlates with the amount of sequencing performed, thereby biasing abundance measurements and the differential abundance statistics derived from them. Using simulated and real data, we show that these methods significantly improve the accuracy of Metastats. We also describe the addition of several new statistical tests to our code (including presence/absence and the corresponding odds ratio, and penetrance calculations) that improve the usability of our software in clinical practice.

Loss of presenilin function causes Alzheimer's disease‐like neurodegeneration in the mouse

Accumulating evidence has indicated that gain-of-function in beta-amyloid production may be not the necessary mechanism for mutant presenilin-1 (PS1) or PS2 to cause familial Alzheimer's disease (AD). In the present article, we show that conditional knockout of PS1 from the adult stage in the forebrain of mice with the PS2 null mutation triggers robust AD-like neurodegeneration including brain shrinkage, cortical and hippocampal atrophy,ventricular enlargement, severe neuronal loss, gliosis, tau hyperphosphorylation, neurofillament tangle-like structures, and intracellular filaments. Learning and memory functions in these mice are almost completely lost. Notably, there is no beta-amyloid deposition, indicating that presenilin dysfunction can directly cause neurodegeneration without the involvement of beta-amyloid. Furthermore, neurodegeneration occurs in a progressive manner following aging, suggesting that an accumulating effect of presenilin dysfunction over time might be a pathogenic mechanism for the involvement of mutant PS1/PS2 in causing AD. These results validate a mouse model characterized by the presence of many features of AD pathology. Furthermore, the demonstration of AD-like neurodegeneration in the absence of beta-amyloid deposition challenges the long-standing beta-amyloid cascade hypothesis and encourages an open debate on the role of beta-amyloid in causing AD. Most important, our results strongly suggest that to develop gamma-secretase inhibitors for the pharmacological treatment of AD may be not a reasonable strategy because antagonism of presenilin function may worsen neurodegeneration.

Extending quadrature methods to value multi-asset and complex path dependent options

The exposition of the quadrature (QUAD) method (Andricopoulos, Widdicks, Duck, and Newton, 2003. Universal option valuation using quadrature methods. Journal of Financial Economics 67, 447–471 (see also Corrigendum, Journal of Financial Economics 73, 603 (2004)) is significantly extended to cover notably more complex and difficult problems in option valuations involving one or more underlyings. Trials comparing several techniques in the literature, adapted from standard lattice, grid and Monte Carlo methods to tackle particular types of problem, show that QUAD offers far greater flexibility, superior convergence, and hence, increased accuracy and considerably reduced computational times. The speed advantage of QUAD means that, even under the curse of dimensionality, it is not necessary to resort to Monte Carlo methods (certainly for options involving up to five underlying assets). Given the universality and flexibility of the method, it should be the method of choice for pricing options involving multiple underlying assets, in the presence of many features, such as early exercise or path dependency.

Extending Quadrature Methods to Value Multi-asset and Complex Path Dependent Options

The exposition of the quadrature (QUAD) method (Andricopoulos et al., 2003) is significantly extended to cover notably more complex and difficult problems in option valuations involving one or more underlyings. Trials comparing with several techniques in the literature, adapted from standard lattice, grid and Monte Carlo methods to tackle particular types of problem show that QUAD offers far greater flexibility, superior convergence and, hence, increased accuracy and considerably reduced computational times. The speed advantage of QUAD means that, even under the 'curse of dimensionality', it is not necessary to resort to Monte Carlo methods (certainly for options involving up to five underlying assets). Given the universality and flexibility of the method, it should be the method of choice for pricing options involving multiple underlying assets, in the presence of many features, such as early exercise or path dependency.

Seasonal variation of hydrographic and nutrient fields during the US JGOFS Arabian Sea Process Study

Between September 1994 and December 1995, the US JGOFS Arabian Sea Process Experiment collected extensive, high quality hydrographic data (temperature, salinity, dissolved oxygen and nutrients) during all seasons in the northern Arabian Sea. An analysis of this unique data suite suggests the presence of many features that are described in the canonical literature, but these new data provided the following insights. 1. Although the seasonal evolution of mixed-layer depths was in general agreement with previous descriptions, the deepest mixed-layer depths in our data occurred during the late NE Monsoon instead of the SW Monsoon. 2. The region exhibits considerable mesoscale variability resulting in extremely variable temperature-salinity (TS) distributions in the upper 1000 db. This mesoscale variability is readily observed in satellite imaging, in the high resolution data taken by a companion ONR funded project, and in underway ADCP data. 3. The densest water reaching the sea surface during coastal upwelling appeared to have maximum offshore depths of ∼150 m and σ θ’s close to the core value (∼25) for the saline Arabian Sea Water (ASW), but salinities in these upwelling waters were relatively low. The densest water found at the sea surface during late NE Monsoon conditions has σ θ’s>24.8 and relatively high salinities, suggesting that they are a source for the ASW salinity maximum. 4. Persian Gulf Water (PGW) with a core σ θ of 26.6 forms a widespread salinity maximum. Despite the considerable extent of this feature, Persian Gulf outflow water, with a salinity (4) of ∼39 at its source, can only be a minor contributor. Within the standard US JGOFS sampling grid, maximum salinities on this surface are ∼36.8 at stations near the Gulf, falling to values as low as ∼35.3 at the stations farthest removed from its influence. Even at our standard stations closest to the Gulf (N-1 and N-2), the high-salinity, low-nutrient Persian Gulf water has only a modest direct effect on nutrient concentrations. This PGW salinity maximum is associated with the suboxic portions of the Arabian Sea’s oxygen minimum zone. 5. The salinity maximum associated with Red Sea Water (RSW, core σ θ=27.2) in the JGOFS study region is clearly evident at the southermost sampling site at 10′N (S-15). Elsewhere, this signal is weak or absent and salinity on the 27.2 σ θ surface tends to increase towards the Persian Gulf, suggesting that the disappearance of this salinity maximum is due, at least in part, to the influence of the Persian Gulf outflow. 6. Inorganic nitrogen-to-phosphate ratios were lower (frequently much lower) than the standard Redfield ratio of 15/1–16/1 (by atoms) at all times and all depths suggesting that inorganic nitrogen was more important than phosphate as a limiting nutrient for phytoplankton growth, and that the effects of denitrification dominated the effects of nitrogen fixation. 7. The water upwelling off the Omani coast during the SW Monsoon has inorganic nitrogen to silicate ratios that were higher (∼2/1) than the ∼1/1 ratio often assumed as the ratio of uptake during diatom growth. 8. The temporal evolution of inorganic nitrogen-to-silicate ratios suggests major alteration by diatom uptake only during the late SW Monsoon cruise (TN050) in August–September 1995. 9. Widespread moderate surface layer nutrient concentrations occurred during the late NE Monsoon. 10. A zone of high offshore nutrient concentrations was encountered during the SW Monsoon, but instead of being associated with offshore upwelling it may represent offshore advection from the coastal upwelling zone, the influence of an eddy, or both. 11. Although our data do not contradict previous suggestions that the volume of subtoxic water may be reduced the SW Monsoon, they suggest a weaker re-oxygenation than indicated by some previous work. Similarly, they do not confirm results suggesting that secondary nitrite maxima may be common in waters with oxygen concentrations >5 μM.