Course details
Data Storage and Preparation
UPA Acad. year 2020/2021 Winter semester 5 credits
The course focuses on modern database systems as typical data sources for knowledge discovery and further on the preparation of data for knowledge discovery. Discussed are extended relational (object-relational, with support for working with XML and JSON documents), spatial, and NoSQL database systems. The corresponding database model, the way of working with data and some methods of indexing are explained. In the context of the knowledge discovery process, attention is paid to the descriptive characteristics of data and visualization techniques used to data understanding. In addition, approaches to solving typical data pre-processing tasks for knowledge discovery, such as data cleaning, integration, transformation, reduction, etc. are explained. Approaches to information extraction from the web are also presented and several real case studies are presented. As a part of the course, students solve a project focused on ...
Guarantor
Course coordinator
Language of instruction
Completion
Time span
- 26 hrs lectures
- 6 hrs exercises
- 6 hrs pc labs
- 14 hrs projects
Assessment points
- 60 pts final exam (written part)
- 20 pts mid-term test (written part)
- 20 pts projects
Department
Lecturer
Instructor
Subject specific learning outcomes and competences
Students will be able to store and manipulate data in suitable database systems, to explore data and prepare data for modelling within knowledge discovery process.
- Student is better able to work with data in various situations.
- Student improves in solving small projects in a small team.
Learning objectives
The aim of the course is to explain the historical development of database technologies, motivation of knowledge discovery from data and basic steps of knowledge discovery process, to explain essence, properties and the use of extended relational and NoSQL databases as data sources for knowledge discovery and to explain approaches and methods used for data understanding and data pre-processing for knowledge discovery.
Why is the course taught
The aim of this course is to demonstrate how to work with complex data around us, how to store such data, how to get oriented in such data, obtain useful descriptive characteristics from such data, and how to prepare such data for extraction of hidden information/knowledge by application of machine learning methods and other advanced analytical methods.
Prerequisite knowledge and skills
- Fundamentals of relational databases and SQL.
- Object-oriented paradigm.
- Fundamentals of XML.
- Fundaments of computational geometry.
- Fundaments of statistics and probability.
Syllabus of lectures
- History of database technology and knowledge discovery, process of knowledge discovery.
- Object-oriented approach in databases.
- NoSQL databases I - introduction to NoSQL, CAP theorem and BASE, key-value databases, data partitioning and distribution.
- NoSQL databases II -data models in NoSQL databases (column, document, and graph databases), querying and data aggregation, NewSQL databases.
- Web scraping.
- Data preparation - data understanding: descriptive characteristics, visualization techniques, correlation analysis.
- Data preparation - data pre-processing I: data cleaning and integration.
- Data preparation - data pre-processing II: data reduction, imbalanced data, data transformation, other data pre-processing tasks.
- Mid-term exam
- Languages and systems for knowledge discovery, real case studies.
- Support for working with XML and JSON documents in databases.
- Spatial databases.
- Indexing of multidimensional data.
Syllabus of numerical exercises
DEMO excercises
- Object-relational and spatial databases, data definition and manipulation, peculiarities
- Multimedia and XML databases, data indices
- NoSQL databases
Syllabus of computer exercises
- Application binding to object-relational databases, application building in spatial databases
- Multimedia and XML databases, building and exploiting data indices
- NoSQL databases in applications
Syllabus - others, projects and individual work of students
- Creation and feature demonstration of both structured and unstructured data processing, where data may be of various nature.
Progress assessment
- Mid-term exam, for which there is only one schedule and, thus, there is no possibility to have another trial.
- One project should be solved and delivered in a given date during a term.
Exam prerequisites:
At the end of a term, a student should have at least 50% of points that he or she could obtain during the term; that means at least 20 points out of 40.
Plagiarism and not allowed cooperation will cause that involved students are not classified and disciplinary action can be initiated.
Controlled instruction
- Mid-term written exam, there is no resit, excused absences are solved by the guarantor.
- The formulation of the data mining task in the prescribed term, excused absences are solved by the assistent.
- The presentation of the project results in the prescribed term, excused absences are solved by the assistent.
- Final exam, The minimal number of points which can be obtained from the final exam is 20. Otherwise, no points will be assigned to the student. excused absences are solved by the guarantor.
Exam prerequisites
At the end of a term, a student should have at least 50% of points that he or she could obtain during the term; that means at least 20 points out of 40.
Plagiarism and not allowed cooperation will cause that involved students are not classified and disciplinary action can be initiated.
Course inclusion in study plans