Separate Machine Research Articles

Identifying neuroanatomical and behavioral features for autism spectrum disorder diagnosis in children using machine learning

Autism spectrum disorder (ASD) is a neurodevelopmental disorder that can cause significant social, communication, and behavioral challenges. Diagnosis of ASD is complicated and there is an urgent need to identify ASD-associated biomarkers and features to help automate diagnostics and develop predictive ASD models. The present study adopts a novel evolutionary algorithm, the conjunctive clause evolutionary algorithm (CCEA), to select features most significant for distinguishing individuals with and without ASD, and is able to accommodate datasets having a small number of samples with a large number of feature measurements. The dataset is unique and comprises both behavioral and neuroimaging measurements from a total of 28 children from 7 to 14 years old. Potential biomarker candidates identified include brain volume, area, cortical thickness, and mean curvature in specific regions around the cingulate cortex, frontal cortex, and temporal-parietal junction, as well as behavioral features associated with theory of mind. A separate machine learning classifier (i.e., k-nearest neighbors algorithm) was used to validate the CCEA feature selection and for ASD prediction. Study findings demonstrate how machine learning tools might help move the needle on improving diagnostic and predictive models of ASD.

Performance of isobaric adiabatic compressed humid air energy storage system with shared equipment and road-return scheme

• Novel isobaric adiabatic compressed humid-air energy storage system was investigated. • Water spray absorbs compression heat and controls compressed air outlet temperature. • Shared equipment and road-return scheme of energy storage/release system is employed. • Optimal system efficiency occurs at compression outlet parameters of 10 MPa/320 °C. • System efficiency reaches 72.6% and theoretic energy storage density is 17.2 kWh·m −3 . To cope with the inherent problem when intermittent renewable energies of solar and wind are connected to the grid, a novel isobaric adiabatic compressed humid air energy storage system was proposed and investigated. This system adopts a single stage dual-usage compressor-expander with synchronous rotating multi-cylinder to replace two separate multi-stage turbine type machines. The temperature of the compressed air is controlled by water spray before compression. The shared equipment with road-return stratagem of thermal oil energy storage/release system is employed. A hydro-assisted isobaric adiabatic compressed air storage system is formed with an underground cavern such as abandoned coal mine as the lower reservoir and a surface reservoir as the upper reservoir, and the nylon cloth pipe system is arranged in the underground cavern to store compressed air around water. By establishing the thermodynamic model of the system, the influences of the outlet pressure and temperature of compression on the system performance were analyzed determined respectively by the depth of the underground cavern and the working temperature of the thermal oil of the energy storage system. The results show that the round-trip efficiency varies slightly with the compression outlet pressure. Under the conditions of compression outlet parameters of 10 MPa/320 °C and the isentropic efficiencies in compression and expansion are 85%, the round-trip efficiency reaches 66.6% and the theoretical energy storage density is 16.5 kWh·m −3 .

Evaluation of three-phase centrifugal separator machine (Tricanter) for olive oil extraction

One of the most important machines in the olive oil extraction line is the horizontal three-phase centrifugal separator machine or Tricanter. The purpose of this paper is to evaluate the machine designed on the basis of Tricanter and to evaluate the quality of extracted olive oil. For this purpose, four different olive cultivars from Gilan Province in Iran were used. In this research the rotational speed of the Tricanter machine was tested at three levels of 2 500, 3 000, and 3 500 rpm and the content of water added to olive paste was used at three levels of 10, 20, and 30% of the paste mass. Peroxide value (PV) and percentage of acidity were measured for oil extracted from all four olive cultivars. The results of the analysis of variance (ANOVA) test showed that rotational speed and the content of added water had an effect on the acidity and PV for all samples of olive cultivars. The measured values showed that the best speed for the Tricanter machine is 3 500 rpm. At this rotational speed, the peroxide and acidity values are lower than the standard values.

Functioning of tourist business enterprises in the context of rural development

The subject of the research is the peculiarities of the functioning of tourism business enterprises in the context of rural development. The purpose of the article is to analyze the resource potential, conditions of organization and foreign experience of rural green tourism enterprises, assessment of the effectiveness of tourism and agricultural activities in rural areas. The methodological basis of the article is the methods and approaches of economic, institutional and sustainable development theory. In particular, systematic and holistic approaches, methods of analysis and synthesis, structural modeling were used; monographic; abstract-logical. The results of the article. It is determined that socio-economic resources of rural green tourism enterprises are understood as labor resources, accommodation facilities, separate utility rooms, machines and equipment or devices, production and technological processes (wine, food, oil and fat, fruit and vegetable canning, etc.), certain productions and activities growing flowers, medicinal herbs, farm animals, picking fruits and berries, mushrooms, honey production, fish farming, etc.) as sightseeing sites or tourist attractions. It is established that the tendency to increase the influence of tourism and resorts on the dynamics of economic and social development of the region is maintained by increasing domestic and foreign tourist flows, as well as increasing revenues from tourism to the local budget. It is substantiated that the assessment of economic efficiency of rural green tourism enterprises should be carried out on the basis of the volume of services provided by them, the costs of their implementation and profits. Field of application of results. The obtained results can be used in the process of further scientific study of an important problem – improving the efficiency of the market of tourist services and rural green tourism enterprises. Conclusions. It is established that rural green tourism enterprises in the leading countries of the world have developed significantly, and their number is 10.0-50.0 times higher than the number of domestic ones. Thus, rural green tourism is a mass activity, employment and income of rural residents. Therefore, in European countries, it is seen as an alternative to agriculture, where the number of employees is constantly declining due to automation and digitalization, and its support has become a state program. It also provides recreation for almost 1/3 of all tourists. Therefore, it is important to implement the most effective areas of experience in rural tourism in Ukraine. This is the organization of thematic annual advertising campaigns for rural recreation with state and regional support; indirect state support in the form of non-taxation; development of pilot projects for the development of rural green tourism for individual rural areas; combining rural green tourism with gastronomic, cognitive, etc. types of tourism; strengthening the greening of rural green tourism.

Automatic Building Detection for Multi-Aspect SAR Images Based on the Variation Features

Multi-aspect synthetic aperture radar (SAR) images contain more information available for automatic target recognition (ATR) than images from a single view. However, the sensitivity to aspect angles also makes it hard to extract and integrate information from multi-aspect images. In this paper, we propose a novel method based on the variations features to realize automatic building detection in the image level. First, to get a comprehensive description of target variation patterns, statistical characteristic variances are derived from three representative and complementary categories. Then, these obtained features are fused and put in the K-means classifier for prescreening, whose results are used as the training sets in supervised classification later to avoid manual labeling. Second, for more precise detection performance, finer features in vector forms are obtained by principal component analysis (PCA). The variation patterns of these feature vectors are explored in two different manners of correlation and fluctuation analyses and processed by separate support vector machines (SVMs) after fusion. Finally, the independent SVM detection results are fused according to a maximum probability rule. Experiments conducted on two different airborne data sets demonstrate the robustness and effectiveness of the proposed method, in spite of significant target signature variabilities and cluttered background.

EnsembleFam: towards more accurate protein family prediction in the twilight zone

BackgroundCurrent protein family modeling methods like profile Hidden Markov Model (pHMM), k-mer based methods, and deep learning-based methods do not provide very accurate protein function prediction for proteins in the twilight zone, due to low sequence similarity to reference proteins with known functions.ResultsWe present a novel method EnsembleFam, aiming at better function prediction for proteins in the twilight zone. EnsembleFam extracts the core characteristics of a protein family using similarity and dissimilarity features calculated from sequence homology relations. EnsembleFam trains three separate Support Vector Machine (SVM) classifiers for each family using these features, and an ensemble prediction is made to classify novel proteins into these families. Extensive experiments are conducted using the Clusters of Orthologous Groups (COG) dataset and G Protein-Coupled Receptor (GPCR) dataset. EnsembleFam not only outperforms state-of-the-art methods on the overall dataset but also provides a much more accurate prediction for twilight zone proteins.ConclusionsEnsembleFam, a machine learning method to model protein families, can be used to better identify members with very low sequence homology. Using EnsembleFam protein functions can be predicted using just sequence information with better accuracy than state-of-the-art methods.

Holography of time machines

We use holography to examine the response of interacting quantum fields to the appearance of closed timelike curves in a dynamically evolving background that initially does not contain them. For this purpose, we study a family of two-dimensional spacetimes that model very broad classes of wormhole time machines. The behavior of strongly coupled conformal theories in these spacetimes is then holographically described by three-dimensional AdS bulk geometries that we explicitly construct. The dual bulk spacetime is free from any divergences, but splits into two disconnected components, without and with CTCs, which are joined only through the boundary; then, passages across the chronology horizon are impossible for any field excitations. In dual terms, the strong self-interaction of the CFT decouples the pathological part from the rest of the spacetime. We also find that entangling the CFTs in two separate time machines connects them through a traversable bulk wormhole. Nevertheless, any entanglement-assisted chronology violations will be prevented by quantum bulk corrections, i.e., subleading 1/N effects, again without needing any gravitational backreaction of the CFT. We are led to speculate that chronology may be protected without involving Planck scale physics.

OpenLDAT—A system for the measurement of display latency metrics

AbstractThe OpenLDAT project (short for Open Latency Display and Analysis Tool) is a system composed of a self‐buildable device and an open‐licensed application to measure several display latency metrics. The most interesting metric is total system latency: the time between an action happening in the physical world, like a mouse being clicked, and the result being displayed on the screen, such as a muzzle flash from a weapon in a videogame. There is currently no similar device on the market, and this type of measurement is traditionally done manually using a modified mouse and a high speed camera, but OpenLDAT can measure it automatically using a built‐in test, or interactively, allowing testing of virtually any game or application, potentially on a separate machine. In addition to system latency, OpenLDAT can also measure more traditional metrics, such as pixel response times.

Demonstration of a high-efficiency carbon dioxide capture and methanation system with heat/material integration for power-to-gas and zero carbon dioxide emissions in flue gasses

Carbon dioxide (CO2) capture and methanation are important methods for realizing both power-to-gas and zero CO2 emissions. However, the technology needs to be improved to solve implementation barriers such as energy efficiency, compactness, economics, and CO2 source flexibility, especially for CO2 derived from fuel combustion.The present report describes the development of a high-efficiency integrated system to separate CO2 from flue gas with a high recovery rate and convert it to high-quality CH4 using in-system heat/material integration.A carbon recycling system with a throughput of 5.1-kW CH4 (LHV basis) was integrated with CO2 adsorbers using zeolite and catalytic methanation reactors. System efficiency, defined as the energy ratio of output CH4 and H2 against the sum of input H2 and power for system operation, was then evaluated using exhaust gas from a combustion furnace with a CO2 concentration of 9%. In general, a decrease in the energy required for system operations, such as gas treatment, CO2 separation, and auxiliary machines, improves efficiency. Two challenges were encountered during the study: reduction in CO2 separation energy by implementation of heat/material management using a H2 reactant as a sweep gas and the reaction heat of methanation, and a reduction in gas compression energy by conducting the methanation reaction under the lowest possible pressure. The results indicated that the CO2 conversion rate was greater than 99% under a pressure of 225 kPa-G, and the only energy consumed was by the oil pump for cooling the reactors. Due to the use of H2 sweep gas and methanation heat recovery, the CO2 recovery rate was also greater than 99%, and the additional energy consumption (1.8 GJ/t-CO2) was due to vacuum pumps. In addition, the system achieved an efficiency greater than 70% (LHV basis), while maintaining both a CO2 recovery rate and a conversion of greater than 99%. Because the energy for dehydration of the exhaust gas was the most significant portion of the total system operational energy, further improvement of the system efficiency is thought to be possible by reducing the energy required for dehydration.

Use of Radiation Circuits for Diagnosis of Melanoma Skin Cancer in Images of Skin Lesions Using Convolutional Neural Networks

Melanoma skin cancer is a fatal illness. However, most melanomas can be treated with minimal surgery if found early. In this regard, the addition of image analysis techniques that automate skin cancer diagnosis would support and increase dermatologists’ diagnosis accuracy. As a result, enhanced melanoma detection can benefit patients who are showing indicators of the disease. Convolutional neural networks can learn from features hierarchically. Since the implementation of a neural network requires a large volume of images to achieve high accuracy rates, an insufficient number of skin cancer images represent an additional challenge in the detection of skin lesions; the current work aims to develop an intelligent system that allows, based on the analysis of images of skin lesions and contextual information of the patient, to accurately determine if it represents a case of melanoma‐type skin cancer. The TensorFlow library was used to execute models in the constructed app. The Mobilenet V2 model was used with a collection of 305 pictures retrieved from the Internet. Diagnoses included melanoma, plaque and psoriatic skin conditions, and Kaposi’s sarcoma and atopic dermatitis. There were two separate machines used to conduct the application tests. There was more than 75% acceptable performance in predicting Kaposi’s sarcoma‐like illnesses for melanoma‐like lesions, as well as plaque psoriasis and atopic dermatitis, respectively. Despite the low amount of images used in training, the constructed mobile application performed well.

The quality of coffee bean classification system based on color by using k-nearest neighbor method

Sorting coffee bean nowadays is still done manually, although there is already a support machine for separation through size, but to determine the quality of the seeds remain manual using human power. This coffee bean sorting is in the spotlight to research whether it can be more effective if there is a tool that can directly find out the quality of coffee. This system will make it easier for workers in the field. In addition to saving time, costs will decrease and also the work of workers will be reduced. This paper present the implementation of machine learning method to classify the coffee bean quality. The dataset use 90 coffee bean for three classes and 30 for each class. From the experimental result, the highest accuracy obtain 83%.

MAC based model to Differentiate Flash crowd and Malicious traffic in SDN

The nature of computer network flash crowd traffic, which is generated by legitimate users accessing servers or other network resources are similar to the traffic generated by Distributed Denial of Service (DDoS) like attacks. With advancement in spoof packet generation tools, attacker may generate Multi-source Multi-destination Multi-protocol (MMM) traffic; characteristics of such traffic are very similar to on-going genuine/ flash crowd traffic in the network. In the case of Software Defined Network (SDN), attacker’s target is controller plane. Controller plane in SDN is a centralized processing unit of the underlying network, which manages several data planes. Controller plane frames the policies and pushes forwarding rules to the data planes. Data planes just maintain the forwarding rules. Thus by overloading the SDN controller, functionality of complete computer network will be hampered. In this paper, we have proposed Media Access Control (MAC) address based Model to Differentiate Flash crowd and Malicious traffic in SDN (MDFMS). Novelty of the proposed model is to detect, locate and mitigate the source of Traditional DDoS (T-DDoS) and MMMDDoS traffic. MDFSM has been implemented on separate machine to avoid any additional computing load on SDN controller. It also preserves the original design of the SDN architecture. Proposed model has been evaluated under various scenarios and encouraging results have been obtained to differentiate T-DDoS and MMM-DDoS from benign flash crowd traffic.

Therapeutic Plasma Exchange using Continuous Flow Cell Separator Machine in a Neuromyelitis Optica Patient Outcome and Analysis

Neuromyelitis optica (NMO) is an inflammatory central nervous system syndrome, associated with serum aquaporin-4 immunoglobulin G antibodies. Autoantibodies against aquaporin-4 (AQP4-IgG), the principal water channel on astrocyte foot processes at the blood brain barrier, are pathogenic in NMO. Acute attacks are managed by high-dose intravenous methylprednisolone, if symptoms fail to resolve, plasma exchange is considered. Here, we present a case of NMO spectrum disorder in a 20 year old lady, failed to pharmacotherapy, was managed by plasma exchange using Spectra Optia Cell Separator machine. Detailed plasma exchange protocol, disease progression and outcome was analysed in this study.

Kidney Replacement Therapy for Patients Requiring Cardiopulmonary Bypass Support during Cardiac Surgery.

Kidney Replacement Therapy for Patients Requiring Cardiopulmonary Bypass Support during Cardiac Surgery.

Associations between gut microbiota composition and AD biomarkers

AbstractBackgroundSeveral studies have reported alterations in gut microbiota composition of Alzheimer’s disease (AD) patients. However, the observed differences are not consistent across studies. We aimed to investigate associations between gut microbiota composition and clinical biomarkers of AD using machine learning models in patients with AD dementia, mild cognitive impairment (MCI) and controls.MethodWe included 169 patients from the NUDAD project, comprising 33 with AD dementia (66±8 years, 46%F, MMSE 21[19‐24]), 21 with MCI (64±8 years, 43%F, MMSE 27[25‐29]) and 115 controls (62±8 years, 44%F, MMSE 29[28‐30]). Fecal samples were collected and gut microbiome composition was determined using 16S rRNA sequencing. Clinical parameters of AD included clinical diagnosis, cerebral spinal fluid (CSF) amyloid and phosphorylated tau (pTau) status, positron emission tomography (PET) amyloid status, and MRI visual scores. Associations between gut microbiota composition and dichotomized clinical parameters of AD were assessed with separate machine learning classification models using XGBoost with nested cross‐validation. The model with the highest area under the curve (AUC) was selected for logistic regression, to assess associations between the 20 best predicting microbes (cumulative sum scaled counts) and the outcome measure from this machine learning model while adjusting for age, sex, and BMI.ResultThe machine learning prediction for amyloid status (CSF) from microbiota composition had the highest AUC. Top predicting microbes included several short chain fatty acid (SCFA)‐producing species. In the logistic regression models, these microbes were significantly associated with lower odds of amyloid positive status, and included Eubacterium ventriosum group spp. (OR 0.49 (0.30‐0.76) per SD increase in counts, p = 0.002), Marvinbryantia spp. (OR 0.55 (0.34‐0.85), p = 0.009), Coprococcus catus (OR 0.58 (0.36‐0.89), p = 0.017), Roseburia hominis (OR 0.59 (0.38‐0.90), p = 0.018), Odoribacter splanchnicus (OR 0.51 (0.30‐0.82), p = 0.008), Lachnospiraceae spp. (OR 0.58 (0.36‐0.89), p = 0.014), and Ruminococcaceae spp. (OR 0.44 (0.25‐0.71), p = 0.002).ConclusionGut microbiota composition had the strongest association with amyloid status among the clinical biomarkers examined. We extend on recent studies that observed associations between SCFA levels and AD biomarkers by showing that higher abundances of SCFA‐producing microbes were associated with lower odds of amyloid positive status.

Thermal conductivity prediction of titania-water nanofluid: A case study using different machine learning algorithms

In this study, the thermal conductivity of titania (TiO2)–water nanofluid was predicted using five separate machine learning algorithms with their unique hyperparameters and logical functions. A detailed comparative study on the TiO2–water nanofluid dataset was performed using the artificial neural network (ANN), gradient boosting regression (GBR), support vector regression (SVR), decision tree regression (DTR), and random forest regression (RFR) algorithms. A complete data intelligence analysis of 228 collected data points on different shape and size of titania (TiO2)–water nanofluid was conducted. The dataset consisted of five parameters including size, shape, volume fraction of the nanoparticles, temperature, and thermal conductivity. The shape of the nanoparticles was considered to be altogether a new parameter for predicting the thermal conductivity of the nanofluids. The mean square error (MSE) and R-square (R2) were used to compare these models. From this comparative study of these algorithms, it was found that the gradient boosting is the best algorithm [R2train = 0.99, R2test = 0.99, MSE.train = 0.0003 and MSE.test = 0.0002] for thermal conductivity predictions with test and train accuracies of 99%. It was observed that, the shape of the nanoparticles could influence the thermal conductivity predictions of the nanofluids, substantially.

Response Prediction to Isocitrate Dehydrogenase (IDH) Inhibitors in Patients with IDH1- or IDH2-Mutated Acute Myeloid Leukemia Using Clinical and Genomic Data

Background: Ivosidenib (Servier Pharmaceuticals LLC) and enasidenib (Celgene) are inhibitors of mutant IDH1 and IDH2, respectively, approved for the treatment of relapsed/refractory (R/R) AML with an IDH1 or IDH2 mutation (ivosidenib is also approved for front-line use in patients ≥ 75 years or with comorbidities). While these drugs were approved based on durable complete remission (CR) + CR with partial hematologic recovery (CRh) rate, only 23-43% of patients responded. Thus, there is a need for a biomarker to better predict response. We previously presented a genomic model of response to enasidenib using linear discriminate analysis (Ghazanchyan et al., ASH 2018). In the two-part study presented here, we improved this modeling by adding clinical data, using machine learning-based approaches to model response to both enasidenib and ivosidenib, and generated an independent validation cohort using clinical samples.Methods: In part I, models were generated using data submitted to FDA in the marketing applications for ivosidenib and enasidenib (studies AG120-C-001 [NCT02074839] and AG221-C-001 [NCT01915498], respectively). Analysis inclusion criteria were available genomic data, IDH mutation positivity per the companion diagnostics, and receipt of at least the approved dose of IDH inhibitor. Patients in the ivosidenib cohort had untreated (n=33) or R/R AML (n=173); patients in the enasidenib cohort had R/R AML. Patients achieving a best response of CR or CRh were considered responders; all other patients were considered nonresponders. Genomic data for the ivosidenib cohorts were generated with the Foundation Medicine FoundationOne Heme panel (FMI) in dose escalation (n=38) and Brigham and Women's Hospital Rapid Heme Panel (RHP) in dose expansion (n=168); these cohorts were analyzed separately to account for differences between panels. Genomic data for the enasidenib cohort (n=75) were generated with the FMI panel. Machine learning-based feature selection approaches were used prior to modeling to identify important clinical data variables (i.e., baseline demographics and disease characteristics). Each analysis cohort was split into a training set (80% patients) and a testing set (20% patients). Machine learning modeling was performed with the decision-tree based algorithm XGBoost using repeated stratified cross validation.In part II of the study, we generated an independent validation cohort consisting of patients with IDH-mutated myeloid malignancies treated with ivosidenib or enasidenib at collaborating institutions (n=14). Bone marrow or peripheral blood samples were processed for deep whole exome sequencing to an average coverage of 441X. Custom analysis pipelines for filtering and annotation were developed to harmonize exome and panel data.Results: In part I, separate machine learning models were generated for each cohort using clinical data only, genomic data only, and clinical and genomic data together. Use of clinical and genomic data together, compared with either data type alone, resulted in improved model performance with accuracies of 89-100% in the testing sets. The most important clinical and genomic features for each cohort are shown in Figure 1.In part II, the independent validation cohort, collaborators provided response data and clinical data for variables found to be important in the models. Patient demographics and response rates are shown in Table 1. Mutational profiles for model genes mutated in ≥ 2 patients are shown in Figure 2. The models for CR+CRh response using both clinical and genomic data in the validation cohort had accuracies of 50-75% (66.7% ivosidenib BWH; 50% ivosidenib FMI; and 75% enasidenib). The combined accuracy improved to 80% when restricted to AML cases (excluding CMML and MDS) treated with ivosidenib (using the BWH model) or enasidenib (n=10).Conclusions: This study demonstrates the potential of harnessing machine learning for biomarker discovery to improve response prediction for the treatment of AML. The modeling approaches improved with inclusion of multiple data types, highlighting the importance of multi-faceted approaches for biomarker development in AML. The models presented have encouraging accuracies in a small independent validation cohort, suggesting potential clinical utility of machine learning-based biomarkers for identifying patients likely to respond to IDH inhibitors. [Display omitted] DisclosuresKural: Senseonics: Current equity holder in publicly-traded company; NanoDimensions: Current equity holder in publicly-traded company; Bionano Genomics: Current equity holder in publicly-traded company. Ghiaur: Menarini Richerche: Research Funding; Syros Pharmaceuticals: Consultancy. Kazandjian: Arcellx: Honoraria, Membership on an entity's Board of Directors or advisory committees; BMS: Honoraria, Membership on an entity's Board of Directors or advisory committees. Lai: Macrogenics: Consultancy, Membership on an entity's Board of Directors or advisory committees; Jazz Pharma: Consultancy, Membership on an entity's Board of Directors or advisory committees; Genentech: Consultancy, Membership on an entity's Board of Directors or advisory committees; Jazz Pharma: Speakers Bureau; Astellas: Speakers Bureau; Daiichi-Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Membership on an entity's Board of Directors or advisory committees. Wynne: Servier Pharmaceuticals: Honoraria; Carisma Therapeutics: Patents & Royalties. OffLabel Disclosure:This presentation references off label use of enasidenib in patients with MDS and CMML.

Extreme Learning Machine based Pattern Classifiers for Symbolic Interval Data

Interval data are usually applied where inaccuracy and variability must be considered. This paper presents a learning method for Interval Extreme Learning Machine (IELM) in classification. IELM has two steps similar to well known ELM. At first weights connecting the input and the hidden layers are generated randomly and in the second step, ELM uses the Moore–Penrose generalized inverse to determine the weights connecting the hidden and output layers. In order to use Moore–Penrose generalized inverse for determining second layer weights in IELM, this paper proposes four classification methods to handle symbolic interval data based on ELM. The first one uses a midpoint of intervals for each feature value then it applies a classic ELM. The second one considers each feature value as a pair of quantitative features and implements a conjoint for classic extreme learning machine. The third one represents interval features by their vertices and performs a classic extreme learning machine as well. The fourth one takes each interval as a pair of quantitative features after that two separated classic extreme learning machines are performed on these features and combines the results accordingly. Algorithms are tested on the synthetic and real datasets. A synthetic dataset is applied to determine the number of hidden layer nodes in an IELM. The classification error rate is considered as a comparison criterion. The error rate obtained for each proposed methods is 19.1667%, 15% , 6.5358% and 18.3333% respectively. Experiments demonstrate the usefulness of these classifiers to classify symbolic interval data.

A PATH GUIDED METHOD FOR QUICKLY LABELING ROADS IN SATELLITE IMAGES

Determining roads from satellite images has gained more research interest after the recent progress on data-heavy machine learning methods which are also accelerated by the increased amounts of accessible data. An important challenge of learning-based approaches is obtaining labeled data to train the systems. In this study, we propose a method for quickly labeling roads over satellite images of any desired location. Our method exploits the 2D path information obtained from OpenStreetMap, an online community-contributed source of geolocated information. In this environment, roads are roughly described as line segments without their exact shapes and sizes. Using this rough information, we propose a simple interactive user interface where users easily label the road boundaries over presented satellite images. Using our approach, it is possible to rapidly label regions with different road characteristics. Such an approach allows for training separate machine learning systems for different parts of the world which would be advantageous over training a single system to identify all kinds of roads.

Ensemble ecological niche modeling of West Nile virus probability in Florida.

Ecological Niche Modeling is a process by which spatiotemporal, climatic, and environmental data are analyzed to predict the distribution of an organism. Using this process, an ensemble ecological niche model for West Nile virus habitat prediction in the state of Florida was developed. This model was created through the weighted averaging of three separate machine learning models—boosted regression tree, random forest, and maximum entropy—developed for this study using sentinel chicken surveillance and remote sensing data. Variable importance differed among the models. The highest variable permutation value included mean dewpoint temperature for the boosted regression tree model, mean temperature for the random forest model, and wetlands focal statistics for the maximum entropy mode. Model validation resulted in area under the receiver curve predictive values ranging from good [0.8728 (95% CI 0.8422–0.8986)] for the maximum entropy model to excellent [0.9996 (95% CI 0.9988–1.0000)] for random forest model, with the ensemble model predictive value also in the excellent range [0.9939 (95% CI 0.9800–0.9979]. This model should allow mosquito control districts to optimize West Nile virus surveillance, improving detection and allowing for a faster, targeted response to reduce West Nile virus transmission potential.