Divide and Conquer: A Quick Scheme for Symbolic Regression

Symbolic regression (SR), as a special machine learning method, can produce mathematical models with explicit expressions. It has received increasing attention in recent years. However, finding a concise, accurate expression is still challenging because of its huge search space. In this work, a divide and conquer (D&C) scheme is proposed. It tries to divide the search space into a number of orthogonal sub-spaces based on the separability feature inferred from the sample data (dividing process). For each sub-space, a sub-function is learned (conquering process). The target model function is then reconstructed with the sub-functions according to their separability patterns. To this end, a separability pattern detecting technique, bi-correlation test (Bi-CT), is also proposed. Note that the sub-functions could be determined by any of the existing SR methods, which makes D&C easy to use. The D&C powered SR has been tested on many symbolic regression problems, and the study shows that D&C can help SR to get the target function more quickly and reliably.

A Deep-Learning Based Visual Sensing Concept for a Robust Classification of Document Images under Real-World Hard Conditions

This paper’s core objective is to develop and validate a new neurocomputing model to classify document images in particularly demanding hard conditions such as image distortions, image size variance and scale, a huge number of classes, etc. Document classification is a special machine vision task in which document images are categorized according to their likelihood. Document classification is by itself an important topic for the digital office and it has several usages. Additionally, different methods for solving this problem have been presented in various studies; their respectively reached performance is however not yet good enough. This task is very tough and challenging. Thus, a novel, more accurate and precise model is needed. Although the related works do reach acceptable accuracy values for less hard conditions, they generally fully fail in the face of those above-mentioned hard, real-world conditions, including, amongst others, distortions such as noise, blur, low contrast, and shadows. In this paper, a novel deep CNN model is developed, validated and benchmarked with a selection of the most relevant recent document classification models. Additionally, the model’s sensitivity was significantly improved by injecting different artifacts during the training process. In the benchmarking, it does clearly outperform all others by at least 4%, thus reaching more than 96% accuracy.

Experimental Modeling of Large Diameter Holes Based on Materials of Bronze Age Stone Axes

In the Urals, there are more than a hundred products with large-diameter holes, some of which are made using a copper tube. Experiments on hollow drilling in the expedition of S. A. Semenov have shown great complexity of this method. In experiments on drilling it was planned to identify labor costs for the manufacture of copper tube, and also reveal various aspects of drilling techniques. The article presents the results of experiments on drilling different types of stone (soapstone, marble, serpentine and jade) with a copper tube. Experiments have shown that drilling with a brace is 1.5 times more effective than drilling with a borer. And using a copper tube as a drill increased the speed of making holes by 4–7 times. It also turned out that the abrasive used is very important for drilling efficiency. The most effective abrasive was emery, whose deposits are known in the southern Urals. The results obtained allow the authors to establish with great accuracy what drills and with application of what equipment stone axes of Bronze Age were drilled. A comparison of the experimental drills with bronze age drills shows that drilling with a copper tube was often, but not always, performed using a special machine tool. Labor-intensive making stone axes with large diameter holes especially made of strong row materials attest about high social status of their owners.

Application and comparative study of SWAT and LSTNet models on runoff simulation in the Atsuma River basin

Accurate runoff simulation is of great importance to understand watershed hydrologic cycle process, effective utilize water resources and respond flood disaster. Hydrologic model is one of the main tools for runoff simulation research and the continuous improvement in Machine Learning offers powerful tools for modeling of hydrologic process. This research took the runoff process of the Atsuma River basin in Hokkaido from 2015 to 2019 as object, proposed a special machine learning framework: Long-and Short-term Time-series Network (LSTNet) for runoff simulation, discussed the accuracy for runoff simulation of LSTNet model with (multivariate LSTNet Model) or without (univariate LSTNet Model) meteorological factors and Soil and Water Assessment Tool (SWAT) model respectively, analyzed the model selection for runoff simulation under different data conditions in the basin. The Nash-Sutcliffe efficiency coefficients (NSE) of the runoff simulation results in the validation (test) period were 0.633 (SWAT model), 0.643 (multivariate LSTNet model), and 0.716 (univariate LSTNet model) respectively. The results show that the accuracies of the two models for runoff simulation in the Atsuma River basin are all very high. SWAT model has prominent advantages in runoff simulation and shortcomings. LSTNet model shows great advantages and potential in runoff simulation. In summary, when target basin’ s data is accurate and complete, the accuracy of SWAT model in runoff simulation is high and stable. When the target basin lacks data or the quality of data is poor, LSTNet model can realize high-precision runoff simulation only based on the measured runoff data, which has a strong application.

Comparative Analysis of various Water Pumping Systems

India is a country which is rewarded with a huge amount of solar energy that varies between 2000 hours to 3000 hours yearly. Solar photovoltaic (SPV) based water pumping is the most appreciated application of solar energy. This paper gives information of various solar photovoltaic (PV) based water pumping systems (WPS) with advantages, issues and more prominent solar PV based WPS driven by a special machine with factors affecting whole system performance. For power conditioning in WPS, the power electronics based DC-DC converters, DC-AC converters, microprocessor and microcontroller are used. The Landsman converter permits excellent tracking of solar radiation using maximum power point tracker (MPPT) with changing insolation levels. This paper gives review on the comparative analysis of different motors used based on reliability, maintenance, and cost-effectiveness. Due to systematic operation emphasis is given to the Brushless DC motor.

MOVEMENT STABILITY OF A SECTION OF THE MACHINE FOR BLACK FALLOW CULTIVATION IN A LONGITUDINAL-VERTICAL PLANE

One of the tasks of using the black fallow in agricultural production is the weed control and the moisture conservation in the soil. Application of the most advanced soil cultivation technologies ensures preservation of no more than 75% of precipitations in the soil. To improve the state of this issue, we have developed a special machine for processing the black fallow. A mathematical model has been developed that describes the dynamics of the movement of the harrow section in a longitudinal-vertical plane, and its solution is given, which allows investigation of the impact of this or that design parameter upon the dynamics of the angle of rotation in time. The adequacy of the developed mathematical model is confirmed by special laboratory and field investigations of the created experimental machine. With rational design parameters the rotation angle of the harrow section in a longitudinal-vertical plane will not exceed – 3º, and the time of its exit to the equilibrium position will not exceed 16...17 s.

Using Digital Footprints in Social Research: an Interdisciplinary Approach

The aim of the study is to clarify the concept of the digital footprint in jurisprudence and social sciences and determine its meaning for interdisciplinary research. The subject of this work is the analysis of the concepts of “digital footprint”, “digital reputation”, “digital image of a person”. The article notes that special machine techniques are required to process digital traces. The ways of storage, transmission and use of digital traces are considered. The research identified the main problems of using digital information in public relations. Digital traces can be used in various studies in the humanities, social sciences: sociology, law, economics, psychology using interdisciplinary research methods. The result of the study was the conclusion that digital information does not always mean a digital footprint. Digital information will have a digital footprint only when it is transferred from one user to another in an online environment, uploaded to a social network, etc. (in the case when it is possible to "trace" its movement). Thus, the "digital footprint" is the ability of information recorded in digital form, to leave special "marks" on the route from subscriber to subscriber, the ability to track such marks, receive information about its movement and transformation in order to collect, process and analyze these data.

Unsupervised Neural Dialect Translation with Commonality and Diversity Modeling

As a special machine translation task, dialect translation has two main characteristics: 1) lack of parallel training corpus; and 2) possessing similar grammar between two sides of the translation. In this paper, we investigate how to exploit the commonality and diversity between dialects thus to build unsupervised translation models merely accessing to monolingual data. Specifically, we leverage pivot-private embedding, layer coordination, as well as parameter sharing to sufficiently model commonality and diversity among source and target, ranging from lexical, through syntactic, to semantic levels. In order to examine the effectiveness of the proposed models, we collect 20 million monolingual corpus for each of Mandarin and Cantonese, which are official language and the most widely used dialect in China. Experimental results reveal that our methods outperform rule-based simplified and traditional Chinese conversion and conventional unsupervised translation models over 12 BLEU scores.

Assembly of MLT type permanent resistors

Considered technological processes of the assemblage of the constant resistors of MLT type for unattended installations, circular layout in two types: with the operation of welding the contact pins without welding (with pressure caps with leads); cutting helical grooves on the metallic base resistor on a special machine; varnishing and painting the raised resistors on the automated line. The equipment for straightening of axial contact terminals and laying of resistors from bulk in the pencil cassette is presented.

Research and Analysis of Automated Special Machine for Fastening of U-Frame and Screw

This paper introduces the Design, Modeling and Simulation of a special tool that attaches a screw to attach the screw to a bimetal of MCB (Miniature Circuit Breaker). Problems found in the current process include complex shape and screw size and full connection operation is an intensive manual operation which makes handling difficult for the operator and also the size of the M2 screw and is headless with a cross section. In partnership with the SPM operator works as a tool to increase productivity by reducing operator fat. The system is a testament to automated material handling system, serial transfer, trap detection, attachment, automatic ejecting system, actuator system, sensors and control capabilities etc. Small screw handling and bike prospect are the major challenges in the special development of a purpose screw attachment system. System Modeling & Simulation analysis is done and the results are encouraging.