Enabling Reproducibility for Small and Large Scale Research Data Sets

The machine learning method, now widely used for predicting well performance from unconventional reservoirs in the industry, generally needs large data sets for model development and training. The large data sets, however, are not always available, especially for newly developed unconventional plays. The objective of this work is to develop an innovative machine learning method for predicting well performance in unconventional reservoirs with a relatively small data set. For a small training data set, the corresponding machine learning model can significantly suffer from so-called overfitting meaning that the model can match the training data but has poor predictivity. To overcome this, our new method averages predictions from multiple models that are developed with the same model input, but different initial guesses of model parameters that are unknowns in a machine learning algorithm and determined in the model training. The averaged results are used for the final model prediction. Unlike traditional ensemble learning methods, each prediction in the new method uses all the input data rather than its subset. We mathematically prove that the averaged prediction provides less model uncertainty and under certain conditions the optimum prediction. It is also demonstrated that the method practically minimizes the overfitting and gives relatively unique prediction. The usefulness of the method is further confirmed by its successful application to the data set collected from less than 100 wells in an unconventional reservoir. Sensitivity results with the trained machine learning model show that the model results are consistent with the domain knowledge regarding the production from the reservoir.

Improving classification accuracy using data augmentation on small data sets

Foam flow in vertical gas wells under liquid loading: Critical velocity and pressure drop prediction

Unstructured text data, such as emails, blogs, contracts, academic publications, organizational documents, transcribed interviews, and even tweets, are important sources of data in Information Systems research. Various forms of qualitative analysis of the content of these data exist and have revealed important insights. Yet, to date, these analyses have been hampered by limitations of human coding of large data sets, and by bias due to human interpretation. In this paper, we compare and combine two quantitative analysis techniques to demonstrate the capabilities of computational analysis for content analysis of unstructured text. Specifically, we seek to demonstrate how two quantitative analytic methods, viz., Latent Semantic Analysis and data mining, can aid researchers in revealing core content topic areas in large (or small) data sets, and in visualizing how these concepts evolve, migrate, converge or diverge over time. We exemplify the complementary application of these techniques through an examination of a 25-year sample of abstracts from selected journals in Information Systems, Management, and Accounting disciplines. Through this work, we explore the capabilities of two computational techniques, and show how these techniques can be used to gather insights from a large corpus of unstructured text.

Comparative analysis of deep convolution neural network models on small scale datasets

Two significant obstacles hinder the advancement of Radiology AI. The first is the challenge of overfitting, where small training data sets can result in unreliable outcomes. The second challenge is the need for more generalizability, the lack of which creates difficulties in implementing the technology across various institutions and practices. A recent innovation, deep neuroevolution (DNE), has been introduced to tackle the overfitting issue by training on small data sets and producing accurate predictions. However, the generalizability of DNE has yet to be proven. This paper strives to overcome this barrier by demonstrating that DNE can achieve satisfactory results in diverse external validation sets. The main innovation of the work is thus showing that DNE can generalize to varied outside data. Our example use case is predicting brain metastasis from neuroblastoma, emphasizing the importance of AI with limited data sets. Despite image collection and labeling advancements, rare diseases will always constrain data availability. We optimized a convolutional neural network (CNN) with DNE to demonstrate generalizability. We trained the CNN with 60 MRI images and tested it on a separate diverse collection of images from over 50 institutions. For comparison, we also trained with the more traditional stochastic gradient descent (SGD) method, with the two variants of (1) training from scratch and (2) transfer learning. Our results show that DNE demonstrates excellent generalizability with 97% accuracy on the heterogeneous testing set, while neither form of SGD could reach 60% accuracy. DNE's ability to generalize from small training sets to external and diverse testing sets suggests that it or similar approaches may play an integral role in improving the clinical performance of AI.

We inquire about the properties of 2D Navier-Stokes turbulence simultaneously forced at small and large scales. The background motivation comes by observational results on atmospheric turbulence. We show that the velocity field is amenable to the sum of two auxiliary velocity fields forced at large and small scale and exhibiting a direct enstrophy and an inverse-energy cascade, respectively. Remarkably, the two auxiliary fields reconcile universal properties of fluxes with positive statistical correlation in the inertial range.

<p><span lang="EN-US">Cold (eV) ions of ionospheric origin dominate the number density of most of the volume of the magnetosphere during most of the time. </span><span lang="EN-US">Supersonic flows of cold positive ions are common and can cause a negatively charged wake behind a positively charged spacecraft. The associated induced electric field can be observed and can be used to study the cold ions. We present observations from the Cluster and MMS spacecraft showing how a charged satellite, and also individual charged wire booms of  an electric field instrument, can be used to investigate cold ion populations. </span><span lang="EN-US">Ionospheric ions affect large scales, including the Alfvén velocity and </span><span lang="EN-US"> </span><span lang="EN-US">thus energy transport with waves and the magnetic reconnection rate. These ions also affect small-scale kinetic plasma physics, including the Hall physics and wave instabilities associated with magnetic reconnection. Concerning large scales, we summarize observations from several spacecraft and show that a typical total outflow rate of ionospheric ions is 10<sup>26</sup> ions/s and that many of these ions stay cold also after a long time in the magnetosphere.  Concerning small scales, we show examples of how cold ions modify the Hall physics of thin current sheets, including magnetic reconnection separatrices. On small kinetic scales the cold ions introduce a new length-scale, a gyro radius between the gyro radii of hot (keV) ions and electrons. </span><span lang="EN-US">The Hall currents carried by electrons can be partially cancelled by the cold ions when electrons and the magnetized cold ions ExB drift together. Also, close to a reconnection X-line an additional diffusion region can be formed (regions associated with hot and cold ions, and with electrons, total of three).</span></p>

Aims Anthropogenic activities have drastically increased nutrient availability, resulting in declines in species richness in many plant communities. However, most previous studies have explored only species-loss patterns and mechanisms over small sampling areas, so their results might overestimate species loss at larger spatial scales. The aim of this research was to explore species diversity change patterns and speciesloss rates at multiple scales in alpine meadow communities following nutrient enrichment. Specifically, we asked two closely related questions: (i) do changes in species diversity and species-loss patterns differ among spatial scales? and (ii) how does community compositional dissimilarity and species turnover change among spatial scale? Methods This study was implemented in an alpine meadow community, which is regarded as one of the most sensitive and vulnerable terrestrial ecosystems to anthropogenic nutrient enrichment. We conducted a fertilization experiment that involved the addition of nitrogen (N), phosphorus (P) and a mixture of both to a series of quadrats ranging from 1 to 16 m~2 over 5 years to study the variations in the patterns of species diversity in response to nutrient additions at different spatial scales. Important Findings Our results showed that the changes in species diversity and species loss were dependent on the type of fertilization and the spatial scale. After N and NP fertilization, species diversity significantly decreased at the small scale but not at the large scale, and the rate of species loss decreased as the spatial scale increased. In contrast, the differences between the P addition and control communities were negligible at both the small and large spatial scales. N fertilization caused species to be lost from the small sampling scale, but because different species were lost from different samples, there was an increase in compositional dissimilarity at larger spatial scales, which reduced the total number of species lost when measured at larger scales. These findings highlight spatial scale in evaluating the biodiversity loss after fertilization and suggest that the compositional dissimilarity might play an important role in mediating species loss after fertilization. Our study significantly improved our understanding of changes in species diversity and species loss at different spatial scales under nutrient-enrichment scenarios.

Microplastic particle ingestion or inhalation by humans is a problem of growing concern. Unfortunately, current research methods that use machine learning to understand their potential harms are obstructed by a lack of available data. Deep learning techniques in particular are challenged by such domains where only small or imbalanced data sets are available. Overcoming this challenge often involves oversampling underrepresented classes or augmenting the existing data to improve model performance. This paper proposes GANsemble: a two-module framework connecting data augmentation with conditional generative adversarial networks (cGANs) to generate class-conditioned synthetic data. First, the data chooser module automates augmentation strategy selection by searching for the best data augmentation strategy. Next, the cGAN module uses this strategy to train a cGAN for generating enhanced synthetic data. We experiment with the GANsemble framework on a small and imbalanced microplastics data set. A Microplastic-cGAN (MPcGAN) algorithm is introduced, and baselines for synthetic microplastics (SYMP) data are established in terms of Fréchet Inception Distance (FID) and Inception Scores (IS). We also provide a synthetic microplastics filter (SYMP-Filter) algorithm to increase the quality of generated SYMP. Additionally, we show the best amount of oversampling with augmentation to fix class imbalance in small microplastics data sets. To our knowledge, this study is the first application of generative AI to synthetically create microplastics data.

Data of IOT (Internet of things) have characteristics of heterogeneity, massive, timeliness and other features, which indicate that much of its data is in the form of small files. Cloud computing is used to deal with large data sets, but a large number of small data sets in the system will occupy most of the resources, resulting in a waste of system resources. In this paper, according to the characteristics of the mass of small data sets, and the deficiency of HDFS handle huge amounts of small data sets. This paper uses MapReduce to analysis the numerous small data sets and proposes a cluster strategy for massive small data based on the k-means clustering algorithm. The experimental results show that the proposed strategy can improve the data processing efficiency, and can improve the utilization of system resources. The research fruits will help us to design more practical merger strategy of massive small data to provide research reference.

A paradigm change in scholarly communication is underway. Supporting Open Science, an effort to make scientific research data accessible to all interested parties by openly publishing research and encouraging others to do the same thereby making it easier to communicate scientific knowledge, is a part of the change that has become increasingly important for (digital) libraries. Digital libraries are able to play a significant role in enabling Open Science by facilitating data sharing, discovery and re-use. Because data citation is often mentioned as one incentive for data sharing, enabling data citation is a crucial feature of research data services. In this article we present a case study of data citation services for the High-Energy Physics (HEP) community using digital library technology. Our example shows how the concept of data citation is implemented for the complete research workflow, covering data production, publishing, citation and tracking of data reuse. We also describe challenges faced and distil lessons learnt for infrastructure providers and scholarly communication stakeholders across disciplines.

Neural networks have been constructed to analyze artificial time series derived from the Tent and Henon map as well as population data of the Canadian lynx. Simple three-layer forwardfeed networks, trained on a small sample data set, provided reasonably good fit to the data and performed well on short-term predictions. Simple neural network models trained on small data sets can perform quite well with synthetic chaotic series as well as population data. The accuracy of the predictions is comparable to that obtained using embedding techniques with three embedding dimensions or nonlinear regression models. In relation to embedding techniques, the current forecasting method using neural networks may be regarded as a global rather than a local method using d embedding dimensions with unit delay time. However, once the network has been trained, the calculation of predicted values is very fast and straightforward, without any searching as required by embedding methods.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Shear viscosity, though being a fundamental property of all fluids, is computationally expensive to calculate from equilibrium molecular dynamics simulations. Recently, machine learning (ML) methods have been used to augment molecular simulations in many contexts, thus showing promise to estimate viscosity too in a relatively inexpensive manner. However, ML methods face significant challenges—such as overfitting, when the size of the data set is small, as is the case with viscosity. In this work, we train seven ML models to predict the shear viscosity of a Lennard–Jones fluid, with particular emphasis on addressing issues arising from a small data set. Specifically, the issues related to model selection, performance estimation and uncertainty quantification were investigated. First, we show that the widely used performance estimation procedure of using a single unseen data set shows a wide variability—in estimating the errors on—small data sets. In this context, the common practice of using cross validation (CV) to select the hyperparameters (model selection) can be adapted to estimate the generalization error (performance estimation) as well. We compare two simple CV procedures for their ability to do both model selection and performance estimation, and find that k-fold CV based procedure shows a lower variance of error estimates. Also, these CV procedures naturally lead to an ensemble of trained ML models. We discuss the role of performance metrics in training and evaluation and propose a method to rank the ML models based on multiple metrics. Finally, two methods for uncertainty quantification—Gaussian process regression (GPR) and ensemble method—were used to estimate the uncertainty on individual predictions. The uncertainty estimates from GPR were also used to construct an applicability domain using which the ML models provided even more reliable predictions on an independent viscosity data set generated in this work. Overall, the procedures prescribed in this work, together, lead to robust ML models for small data sets.

For Yeast Protein Hubs, More Data Means More Connections