Constructive-synthesizing modeling and the Process Mining methods in a toolkit to monitor and analyze the software debugging process were applied. Methods for monitoring the development and debugging processes are the basis for improving the level of practical training of students, reducing the time that is used irrationally in the process of software development by a student, and in monitoring the processes of performance of tasks by a teacher. The process of software debugging is seen as a sequence of actions when dealing with relevant tools. Using the methodology of constructive-synthesizing modeling, a constructor for forming a debugging actions log was developed. Based on the constructive model, the extension to the integrated development environment (IDE) Microsoft Visual Studio, in which all debugging actions are recorded in an event log, was designed. During debugging in the IDE, event logs are collected and then a conformance checking of these logs with regard to the reference model is performed. To do this, the ProM (Eindhoven Technical University, Netherlands), a platform for Process Mining methods, is used. By checking compliance, it is possible to compare different debugging processes and recognize behavioral similarities and differences. The main purpose of the developed toolkit is to collect debugging actions from the developer’s IDE. By better understanding how students grasp and deal with errors, one can help novices learn to program. Knowing how programmers debug can encourage researchers to develop more practically directed methods, enable teachers to improve their debugging curricula and allow tool developers to adapt the debugger to the actual needs of users. It is practically suggested to use the prepared tools in the software engineering course

/pdf/development-of-a-toolkit-for-analyzing-software-debugging-7bunixfu5d.pdf

Development of a toolkit for analyzing software debugging processes using the constructive approach

Color models' opportunities were explored for lightning flashes extraction in significant dynamic cloudiness. It was shown that the greatest efficiently in recognition is provided by combining Lab and LCH model-based features. The ranges of color channels are defined for lightning aureoles. Linear and quadratic filters were developed for aureoles detection at the series frames of the video from meteorological satellites. An analysis of their effectiveness was done. The lightning detection filter is based on the processing a current frame of the video and the filtered one containing lightning aureoles. The method and software for lightning extraction were developed based on the image filtering.

Lightning Recognition on the Filtered Videos in the Dynamic Clouds

At first, deep analysis of modeling processes of the lightning flashes dynamic behavior in static background and moving thunderstorm fronts with very mobile clouds at NASA satellite videos was done. Color models’ opportunities were explored for lightning flashes extraction in significant dynamic thunderstorm fronts and cloudiness. It was shown that the greatest efficiently in recognition is provided by combining Lab and LCH model-based features. The ranges of color channels were defined for lightning aureoles. This data was used for their detection at the series frames from meteorological satellites. The lightning detection filtering is based on the processing a current frame of the video with linear and quadratic filters were developed to solve these tasks. An analysis of their effectiveness was done. Modeling of the lightning flashes was implemented using constructive-synthesizing approach. A set of constructors was developed. Implementing parametric multi-character constructors allows forming fractal sequences of characters. Constructor-converter from the character string to time series creates fractal time series, which determine the location, magnitude and decay rate of lightning discharges. Model video images of lightning in the thunderstorm front are formed in accordance with the implementation of the constructor-assembler. The methods and software for lightning extraction from NASA satellites video, and also for realization constructive-synthesizing models were developed.

Constructive-Synthesizing Modeling of Lightning Flashes in the Dynamic Thunderstorm Front

Tools and methods for automatic model generation for the time series’ studied class have been developed. The constructive-synthesizing model is the simplest known one thus it contains a single rule with 4-10 characters. A composite constructor, that consists of four interconnected constructors, was developed using structural synthesizing modeling tools. One of the constructors, according to the given rules, constructs a multi-symbol fractal sequence, the other transforms it into a deterministic fractal time series, the third, using the two preceding ones and genetic algorithms, finds a rule that allows the desired time series to be generated, and the fourth is a control constructor. The appropriate software has been developed. The experiments were conducted on the rules with four characters and a variable number from four to ten on the of the rule's right part. In all cases, the rules found allow accurate recreation of the original time series. The practical application of the proposed approach requires further research into stochastic time series generated by a number of rules.

Modeling of the Deterministic Fractal Time Series by One Rule Constructors

Using the tools of structural-synthesizing modeling, a set of constructors was developed. Implementing parametric multi-character constructors allows to form fractal sequences of characters. Constructor-converter from the character string to time series creates fractal time series, which determine the location, magnitude and decay rate of lightning discharges. Model video images of lightnings in the thunderstorm front are formed in accordance with the implementation of the constructor-assembler. All constructors are developed on the basis of the generalized constructor that was previously presented and repeatedly tested. The model adequacy of the model is confirmed by comparing the video image of the model with the image, what was obtained by NASA satellite. This approach can be the basis for solving the dynamic problems on lightning protection of engineering constructions and civil objects, and development of strategy of aircraft behavior in order to mitigate the risks of lightning strokes in the conditions of movement in the thunderstorm front.

Modeling of Lightning Flashes in Thunderstorm Front by Constructive Production of Fractal Time Series

Using the tools of constructive-synthesizing modeling, constructors of fractal time series which determine the location, magnitude and rate of damping of lighting discharges are developed. Model video images of lightning in the thunderstorm front are formed according to constructors' implementation. The adequacy of the model is verified by comparison of the model video image with the same produced by NASA satellite.

Kostiantyn Lytvynenko

Papers

Modeling of Lightning Flashes in Thunderstorm Front by Constructive Production of Fractal Time Series

Constructive Modeling of Lightning Activity in Thunderstorm Front