Кафедра "Комп'ютерна інженерія та програмування"

Постійне посилання колекціїhttps://repository.kpi.kharkov.ua/handle/KhPI-Press/1095

Офіційний сайт кафедри https://web.kpi.kharkov.ua/cep

Від 26 листопада 2021 року кафедра має назву – "Комп’ютерна інженерія та програмування"; попередні назви – “Обчислювальна техніка та програмування”, “Електронні обчислювальні машини”, первісна назва – кафедра “Математичні та лічильно-вирішальні прилади та пристрої”.

Кафедра “Математичні та лічильно-вирішальні прилади та пристрої” заснована 1 вересня 1961 року. Організатором та її першим завідувачем був професор Віктор Георгійович Васильєв.

Кафедра входить до складу Навчально-наукового інституту комп'ютерних наук та інформаційних технологій Національного технічного університету "Харківський політехнічний інститут". Перший випуск – 24 інженери, підготовлених кафедрою, відбувся в 1964 році. З тих пір кафедрою підготовлено понад 4 тисячі фахівців, зокрема близько 500 для 50 країн світу.

У складі науково-педагогічного колективу кафедри працюють: 11 докторів технічних наук, 21 кандидат технічних наук, 1 – економічних, 1 – фізико-математичних, 1 – педагогічних, 1 доктор філософії; 9 співробітників мають звання професора, 14 – доцента, 2 – старшого наукового співробітника.

Переглянути

Фільтри

2020 - 20213

Налаштування

Автор: search.filters.author.Meleshko, Yelyzaveta

Результати пошуку

Зараз показуємо 1 - 3 з 3
  • Ескіз
    Документ
    Study of influence of quadrocopter design and settings on quality of its work during monitoring of ground objects
    (Національний технічний університет "Харківський політехнічний інститут", 2021) Maidanyk, Oleksandr; Meleshko, Yelyzaveta; Shymko, Serhii
    The subject of the article is methods of reducing quadcopter magnetometer crosstalk by changing the design and settings of the copter to improve the quality of its work during the monitoring of ground objects. The relevance of the development is determined by the need to increase the physical safety of quadcopters when monitoring ground facilities in various industries because the magnetometer is the most noise-sensitive sensor, and its failure leads to the fall and loss of the drone. The purpose of the article is to determine the optimal design and settings of the quadcopter in terms of its physical safety and quality of work during monitoring of ground facilities in various industries. The research task is to check whether it is possible to protect the magnetometer placed inside the drone body from the power cables crosstalk by grounding, shielding and changing the initial settings of the copter, namely by changing the value of the startup power factor of the motors. Research methods are as follows: theory of automatic control, methods of optimal control and hardware design methods. Conclusions. The role of the drone magnetometer in the monitoring of ground objects has been studied. The study has shown that copters at monitoring ground objects must be equipped with a magnetometer and GPS. The magnetometer is the most sensitive to interference of all sensors. If it does not work properly, the entire drone navigation system stops working. We have carried out experimental studies of the influence of quadcopter design and settings on the quality of its magnetometer work, and hence on work of the copter as a whole. In this paper it is proposed to place a magnetometer inside the body of the drone that will increase its physical safety and simplify the design of the drone, but at the same time it will increase the coupling from the power cables of motors, so it is necessary to choose effective methods of protection. It has been tested whether it is possible to protect the magnetometer from interference from power cables when placing it inside the drone body by grounding, shielding and changing the initial settings of the copter, namely by changing the value of the startup power factor of the motors. The results of the experiments showed that to protect against the interference for magnetometer placed inside the drone body, it is necessary to combine shielding of the magnetometer and decreasing of the startup power factor of the motors.
  • Ескіз
    Документ
    Method of identification bot profiles based on neural networks in recommendation systems
    (Національний технічний університет "Харківський політехнічний інститут", 2020) Meleshko, Yelyzaveta; Drieiev, Oleksandr; Drieieva, Hanna
    The subject matter of the article is the process of increased the information security of recommendation systems. The goal of this work is to develop a method of identification bot profiles in recommendation systems. In this work, the basic models of information attacks by the profile-injection method on recommendation systems were researched, the method of identification bot profiles in recommendation systems using the multilayer feedforward neural networkwas developed and the experiments to test the quality of its work were conducted. The developed method is to identify bot profiles that attempt to change item ratings in a recommendation system inorder to increase the occurrence frequency of target items in recommendation lists to all authentic users, or to certain segments of authentic users. When removing bot profiles' data from the database of the recommendation system before generating recommendation lists, the accuracy of the system and the correctness of recommendations are significantly increased, and authentic users get protection from information attacks. Random, Average and Popular attacks were used to model the attacks on a recommendation system. To identify bots, their ratings for system items were analyzed. The experiments have shown that the neural network that analyzes only the numbers of different ratings in a profile, detects bot profiles with high accuracy, that use Random attack regardless of the number of target items foreach bot. At the same time, the developed neural network can detect bots that use Average or Popular attacks only when they have several target items. Also, the results of the experiments show that type I errors, when the system identifies authentic users as bots, is very rarely appear in the developed method. To improve the accuracy of the neural network, there can add to analysis also other data of user profiles, such as the timestamp of each rating and as segments of items, which was rated.
  • Ескіз
    Документ
    The improved model of user similarity coefficients computation for recommendation systems
    (Національний технічний університет "Харківський політехнічний інститут", 2020) Meleshko, Yelyzaveta; Drieiev, Oleksandr; Al-Oraiqat, Anas Mahmoud
    The subject matter of the article is a model of calculating the user similarity coefficients of the recommendation systems. The urgency of the development is determined by the need to improve the quality of recommendation systems by adapting the time characteristics to possible changes in the similarity coefficients of users. The goal is the development of the improved model of user similarity coefficients calculation for recommendation systems to optimize the time of forming recommendation lists. The tasks to be solved are: to investigate the probability of changing user preferences of a recommendation system by comparing their similarity coefficients in time, to investigate which distribution function describes the changes of similarity coefficients of users in time. The methods used are: graph theory, probability theory, radioactivity theory, algorithm theory. Conclusions.In the course of the researches, the modelof user similarity coefficients calculating for the recommendation systems has been improved. The model differs from the known ones in that it takes into account the recalculation period of similarity coefficients for the individual user and average recalculation period of similarity coefficients for all users of the system or a specific group of users. The software has been developed, in which a series of experiments was conducted to test the effectiveness of the developedmethod. The conducted experiments showed that the developed method in general increases the quality of the recommendation system without significant fluctuations of Precision and Recall of the system. Precision and Recall can decrease slightly or increase, depending on the characteristics of the incoming data set. The use of the proposed solutions will increase the application period of the previously calculated similarity coefficients of users for the prediction of preferences without their recalculation and, accordingly, it will shortenthe time of formation and issuance of recommendation lists up to 2 times.