تأثير مفهوم الانجراف على قواعد البيانات المتدفقة لأساتذة قسم علوم الحاسوب
Concept Drift Effect in Streaming Database for Computer Science Department Staff
- الباحث
- منذر يونس شيت ذنون
- المشرف
- عمار ظاهر ياسين طه
- الجامعة
- جامعة الموصل
- الكلية / القسم
- كلية علوم الحاسوب والرياضيات / علوم الحاسوب
- الشهادة
- ماجستير
- السنة
- 2023
- لغة الرسالة
- عربي
- الكلمات المفتاحية
- انجراف المفهوم ; قواعد البيانات المتدفقة ; تعلم الآلة ; البيانات الضخمة
Abstract
Our life never stops evolving and changing, and our computer systems must adapt this change. It has been noted recently that there has been an enormous and unlimited flow of data from various and varied sources. Most of the current systems require processing events in real time, because the data are important at the time they occur, and it must be processed as soon as possible, or else the data will lose their importance. With the many changes that occur on the traditional database as deletion, addition and continuous updates, the need arises for a tool that can manage the data stream and process them in the shortest possible time. This thesis aims to produce a new tool through which the data stream is managed and events are processed in real time in order to achieve an integration between Python V 3.8 and SQL Server, and designing and implementing a streaming database, through which the data stream is processed in real time to make important decisions based on the stream of those data, as well as drift detection obtained on the streaming database. Most computer systems have great performance and scalability, but the installation and the setup require a high level of expertise and involve a very difficult learning curve. Having systems that implement stream concepts in SQL enhances all of its benefits with the addition of stream features. This thesis contributes to providing a digital archive for the Department of Computer Science and a general study on data stream, the design and implementation of a streaming database. A case study was made for the Department of Computer Science to build a streaming database system, and the proposed tool was contacted to extract data on the activities of the teaching staff in the department. The drift on the streaming database was detected by using three algorithms: DDM, EDDM, and ADWIN to analyze the streaming data. The algorithms were tested to detect drift on the streaming database, according to the output results, the EDDM algorithm achieves the most efficient execution time in data analysis. The results show that the tool has achieved a success in performing its functions in a good way in managing and processing data stream in real time and achieving integration with the relational database, with the possibility of applying the streaming database in many departments of the University of Mosul. streaming database, according to the output results, the EDDM algorithm achieves the most efficient execution time in data analysis. The results show that the tool has achieved a success in performing its functions in a good way in managing and processing data stream in real time and achieving integration with the relational database, with the possibility of applying the streaming database in many departments of the University of Mosul.
طلب نسخة من الملف
الملفات غير منشورة على الموقع. اترك بياناتك وسيتواصل معك موظفو قسم الشؤون الثقافية والفنون لإرسال النسخة مجاناً.