31 results
Search Results
Now showing 1 - 10 of 31
Item Open Access COMPARATIVE ANALYSIS OF FACE RECOGNITION ALGORITHMS IN THE PROBLEM OF VISUAL IDENTIFICATION(СДУ хабаршысы - 2021, 2021) Sadyk U. ; Tursyn D. ; Sarsenbay Sh. ; Aitimov A.Abstract. The article is devoted to two approaches to face recognition implemented in the visual identification software package. The corresponding algorithms are described, flowcharts and fragments of their software implementation are given, and the results of their applicability are analyzed on the basis of the conducted research.Item Open Access THE METHODS AND ALGORITHMS FOR RECOGNIZING KAZAKH LANGUAGE FEATURES(СДУ хабаршысы - 2021, 2021) Kalmurzayev Y. ; Kuanyshbay D.; Othman M.Abstract. Despite the importance of automatic speech recognition (ASR), it is difficult to find freely available models, especially for languages with few speakers. This paper describes a method for training Kazakh models based on end-to-end ASR architecture using open-source data. We put the models to the test, and the results are promising. However, much more training data is required to perform well in noisy environments. We make available to the public our trained Kazakh models and training configurations.Item Open Access OPTIMIZATION ALGORITHMS OVERVIEW FOR COURSE SCHEDULING(СДУ хабаршысы - 2021, 2021) Zhuniskhanov M. ; Suliyev R.Abstract. Today we have a lot of optimization algorithms that can be used for course scheduling, but the main question is which of them fulfills our requirements better and what are the advantages of one from the others. Since course scheduling is an NP-problem, in this article we will look at a few of algorithms that are used to solve different problems and have proved themselves only on the best side and compare their main advantages. When comparing, we will rely not only on the accuracy of the results, but also on the speed of solving complex problems, and on solving complex conditions.Item Open Access ONLINE COURSE RECOMMENDATION SYSTEMS(СДУ хабаршысы - 2021, 2021) Baltabekov Y.S. ; Moldagaliev B.K.Abstract. Today in our world exists many online courses, so today it is hard to students to choose the courses they like. This paper work is about recommendation system based on user’s interests in his profile, this technique called Content Based, and on similarity with other users, this technique called Collaborative filtering, and hybrid method. My aim is to make such system, which will recommend courses that will be interesting to users. By using these techniques it can be improved efficiency of the recommendation system and make the life of the users more easier. Because in the process of study, the user will not be difficult to receive information and he will not be border.Item Open Access ТЕПЛОВЫЕ ПОЛИНОМЫ ДЛЯ ОБРАТНОЙ ЗАДАЧИ СТЕФАНА(СДУ хабаршысы - 2021, 2021) Касабек Д. ; Касабек С. ; Жайлаубек A. ; Телеубек M.Аннотация. Изучения процессов в области контактной техники имеют важную роль во всем мире. Данная статья рассматривает математическую модель замкнутых контактных элементов при нагреве. При нагреве электрического контакта происходят много теплофизических процессов. Задача Стефана является математической моделью процесса с изменением фазы, где тепло либо поглощается, либо выделяется. К примеру, затвердевание металлов, замерзание грунта и воды, таяние льда, рост кристаллов и т.д. Есть два вида задачи Стефана, прямая и обратная. B первом, цель задачи найти распределение температуры по контактному элементу и нахождение положении подвижных границ. B последнем же, предполагая, что подвижные границы известны, составить функцию теплового потока, определяющую распределение температуры. Данная работа дает приближенное решение методом тепловых полиномов.Item Open Access AUTOMATIC ERROR CORRECTION: EVALUATING PERFORMANCE OF SPELL CHECKER TOOLS(СДУ хабаршысы - 2021, 2021) Tolegenova A.Abstract. Spell checking is the task of detecting and correcting spelling errors in text and is one of the most sought-after processes in NLP. There are many open-source toolkits for checking and correcting errors in the text. To test how effective these tools are, in this article I have presented an evaluation of three types of tools as NeuSpell, SymSpell and Hunspell. SymSpell showed a high speed of 2480, this is an indicator of how fast it works than others. And NeuSpell achieved the lowest error rate of 0.80%. The results show the disadvantages and advantages of all algorithms, and that there is still room for improvement.Item Open Access USING THE GINI COEFFICIENT TO BUILD A CREDIT SCORING MODEL(СДУ хабаршысы - 2021, 2021) Sultanova N. ; Tulegenova A. ; Suleimen B.Abstract. The credit scoring model is widely used to predict the likelihood of a customer default. To measure the quality of such scoring models, you can use quantitative indicators such as the GINI index, KolmogorovSmirnov (KS) statistics, Lift, Mahalanobis distance, information statistics. This article discusses and illustrates the practical use of the GINI index.Item Open Access COMPARATIVE ANALYSIS OF SORTING ALGORITHMS USED IN COMPETITIVE PROGRAMMING(СДУ хабаршысы - 2021, 2021) Mamatnabiyev Zh. ; Zhaparov M.Abstract. Sorting is one of the most used and fundamental operation in computer science since the first time it had been tried to arrange list of items in some sequence. A large number of sorting algorithms were designed in order to have best performance in terms of computational complexity, memory usage, stability, and methods. In addition, in algorithmic problem solving, not all algorithms have same efficiency. This paper makes comparison research and discusses four different types of sorting algorithms: bubble sort, quick sort, insertion sort, and merge sort. The algorithms are tested and compared for data usage and time spent for sorting given amount of data.Item Open Access IMPLEMENTING COURSE SCHEDULE GENERATION APPLICATION FOR UNIVERSITY(СДУ хабаршысы - 2021, 2021) Zhuniskhanov M. ; Suliyev R.Abstract. This work is about researching and implementing Course Schedule Generation Application for university. She compares other similar applications in this area and explores important points. Through UML diagrams, important concepts and development progress are explained. Screenshots of using the application will also be provided. Dataset, configuration and rules explained. In the end, we will talk about the importance of such an application and facilitate planning for the university.Item Open Access HANDWRITTEN OPTICAL CHARACTER RECOGNITION: IMPLEMENTATION FOR KAZAKH LANGUAGE(СДУ хабаршысы - 2021, 2021) Kalken M.Abstract. Many documents, including as invoices, taxes, memoranda, and surveys, historical data, and test replies, still require handwriting with the transformation to digital information interchange. Handwritten text recognition (HTR), which is an automatic approach to decode records using a computer, is required in this aspect. For this proposal, I present a study of the implementation of optical recognition algorithms for handwritten text in the Kazakh language, using a recently collected database. The database, called the Kazakh Autonomous Handwritten Text Dataset (KOHTD), contains more than 140,335 segmented images of handwritten exam papers. As an algorithm, I used the proposed model by Harald Scheidl, which consists of several layers of neural networks and an CTC decoder. The trained model by putting an interval of Ir = 0.01 and a batch size of 60 showed effective results with indicators of about 85% accuracy.