A sketch for the KS test for Big Data

Motivated by the challenges of Big Data, this paper presents an approximative algorithm to assess the Kolmogorov-Smirnov test. This goodness of fit statistical test is extensively used because it is non-parametric. This work focuses on the one-sample test, which considers the hypothesis that a given univariate sample follows some reference distribution. The method allows to evaluate the departure from such a distribution of a input stream, being space and time efficient. We show the accuracy of our algorithm by making several experiments in different scenarios: varying reference distribution and its parameters, sample size, and available memory. The performance of rival methods, some of which are considered the state-of-the-art, were compared. It is demonstrated that our algorithm is superior in most of the cases, considering the absolute error of the test statistic.

Prognosis of severe cytomegalovirus infection in newborns

Objective is to study the features of impaired activation of T and B lymphocytes in order to predicting severe cytomegalovirus infection in newborns. Materials and methods. 133 newborns with cytomegalovirus infection were examined. Immediately after diagnosing cytomegalovirus infection, all patients observed were immunologically ex amined, including assessing count of peripheral blood T and B lymphocytes, as well as their intercellular interaction by using flow cytometry immunostaining for CD3, CD3+CD28–, CD3+CD28+, CD3–CD28+, CD4, CD8, CD20, CD20+CD40+, CD28, CD40. The test was performed by using a Beckman Coulter Epics XL laser flow cytofluorometer. Depending on the condition severity, all children were divided into two groups: 1 — cytomegalovirus infection, severe form — 60 subjects (45.1%); 2 — cytomegalovirus infection, moderate form — 73 subjects (54.9%). Results of the entire set of studied indicators for cellular and humoral arms of immune system revealed statistically significant differences for the prognosis of severe cytomegalovirus infection: CD3+CD28–, CD20, CD20+CD40+, CD4. T lymphocytes with CD3+CD28+ activation markers, through which costimulating signals necessary for the activation of T helper cells are exerted cell-intrinsic features, serving as an important factor ensuring immune response. Using the “classification trees” method, we developed a differentiated approach to forecast severe cytomegalovirus infection in newborns. Systems of inequalities were obtained, four of which classify a subgroup of newborns with severe cytomegalovirus infection. The consistent application of the obtained inequalities makes it possible to isolate from the input stream of sick patients with a prognosis of the development of severe cytomegalovirus infection. The proposed diagnostic rules can be considered as screening markers for predicting a severe cytomegalovirus infection in newborns, which makes possible the timely onset of specific therapy.

On the Time to Buffer Overflow in a Queueing Model with a General Independent Input Stream and Power-Saving Mechanism Based on Working Vacations

A single server GI/M/1 queue with a limited buffer and an energy-saving mechanism based on a single working vacation policy is analyzed. The general independent input stream and exponential service times are considered. When the queue is empty after a service completion epoch, the server lowers the service speed for a random amount of time following an exponential distribution. Packets that arrive while the buffer is saturated are rejected. The analysis is focused on the duration of the time period with no packet losses. A system of equations for the transient time to the first buffer overflow cumulative distribution functions conditioned by the initial state and working mode of the service unit is stated using the idea of an embedded Markov chain and the continuous version of the law of total probability. The explicit representation for the Laplace transform of considered characteristics is found using a linear algebra-based approach. The results are illustrated using numerical examples, and the impact of the key parameters of the model is investigated.

CoCoS: Fast and Accurate Distributed Triangle Counting in Graph Streams

Given a graph stream, how can we estimate the number of triangles in it using multiple machines with limited storage? Specifically, how should edges be processed and sampled across the machines for rapid and accurate estimation? The count of triangles (i.e., cliques of size three) has proven useful in numerous applications, including anomaly detection, community detection, and link recommendation. For triangle counting in large and dynamic graphs, recent work has focused largely on streaming algorithms and distributed algorithms but little on their combinations for “the best of both worlds.” In this work, we propose CoCoS , a fast and accurate distributed streaming algorithm for estimating the counts of global triangles (i.e., all triangles) and local triangles incident to each node. Making one pass over the input stream, CoCoS carefully processes and stores the edges across multiple machines so that the redundant use of computational and storage resources is minimized. Compared to baselines, CoCoS is: (a) accurate: giving up to smaller estimation error; (b) fast : up to faster, scaling linearly with the size of the input stream; and (c) theoretically sound : yielding unbiased estimates.

Metrisable assessment of the course of stream-systemic processes in vector form in industry 4.0

The goal of this paper is to present an innovative conception how to use metrisable vector structure of a manufacturing process, based on quantitative relations between the activity of input streams, features of the product, and effect of losses; all of which are excellent practical solution for Industry 4.0, and in turn intelligent factories. This solution can be a usefull way in the process of building sustainable organization. A vector representation of manufacturing processes was formulated, one which is based in system engineering. Three manufacturing system state vectors were proposed. These are: input stream vector $${\upphi }$$ ϕ , product features vector $$\overrightarrow {{\text{ P}}}$$ P → which is also referred to as quality vector, and losses vector $$\overrightarrow {{\text{ S}}}$$ S → . Scalar, vector, and mixed products of these vectors may form constitutive equations of manufacturing processes. The relations between the vectors $$\upphi$$ ϕ , $$\overrightarrow {{\text{ P}}}$$ P → ,$$\overrightarrow {{\text{ S}}}$$ S → provide a possibility for a metrisable, complex analysis and assessment of a contemporary manufacturing process. The paper shows practical methods for defining the size of the vector values within the process. The demonstrated vector description of stream-systemic processes can also be applied to non-material manufacturing.

An Online Stream Monitoring Algorithm for Fraud Detection in the Transport of Goods

The process of monitoring vehicles used in road transports plays an important role in detecting fraud committed by drivers. Algorithm designers face a number of challenges, including large number of vehicles monitored, demands related to online calculations, and ability to easily explain fraud alarms triggered to supervisors who make final decisions about actions to be taken. In this paper, we propose rather general, lightweight stream, online heuristics. The vehicle’s position is periodically controlled by a GNSS device. The algorithm detects potential illegal activities along the route between the origin and the destination. Anomalies in the vehicle’s trajectory are detected, based on a multi-resolution analysis of the economy of routes. The economy metric is easily understood and verifiable by controllers. The solution is also capable of identifying clearly suspicious trajectories that popular geofencing approaches would overlook. The scale on which the solution may be adopted is obtained thanks to the stream – like nature of the algorithm: essentially, the resources used do not increase along with the size of the input stream (the number of GNSS frames generated for the vehicle). An experiment illustrating the algorithm’s viability is presented as well.

Towards Bayesian Quantification of Permeability in Micro-scale Porous Structures – The Database of Micro Networks

This article develops a Bayesian framework to quantify the absolute permeability of water in a porous structure from the geometry and clustering parameters of its underlying pore-throat network. These parameters include the network`s diameter, transivity, degree, centrality, assortativity, edge density, K-core decomposition, Kleinberg’s hub centrality scores, Kleinberg's authority centrality scores, length, and porosity. In addition, the incorporated clustering aspects of the networks have been determined with respect to several clustering criteria – edge betweenness, greedy optimization of modularity, multi-level optimization of modularity, and short random walks. As such, the article takes the first footsteps of creating a Database of Micro Networks for micro-scale porous structures, to be used as main input stream for the proposed Bayesian scheme. Doi: 10.28991/HIJ-2020-01-04-02 Full Text: PDF

THE CASE FOR LOW-COST, PERSONALIZED VISUALIZATION FOR ENHANCING NATURAL HAZARD PREPAREDNESS

Each year, lives are needlessly lost to floods due to residents failing to heed evacuation advisories. Risk communication research suggests that flood warnings need to be more vivid, contextualized, and visualizable, in order to engage the message recipient. This paper makes the case for the development of a low-cost augmented reality tool that enables individuals to visualize, at close range and in three-dimension, their homes, schools, and places of work and worship subjected to flooding (modeled upon a series of federally expected flood hazard levels). This paper also introduces initial tool development in this area and the related data input stream.