This course introduces advanced aspects of data warehousing and data mining, encompassing the principles, research results and commercial application of the current technologies.
CSC420Semester VII3 credit hours9 units70 solved questions915 study cardsFull study modes
This course introduces advanced aspects of data warehousing and data mining, encompassing the principles, research results and commercial application of the current technologies.
The free sample is open to everyone. Sign in for 2 free weeks of every answer, flashcards and progress.
This unit introduces the data warehouse as the historical, subject oriented store that feeds analytical decision making. It follows data from its creation through to its use inside a warehouse, contrasts a warehouse with the everyday operational database that runs a business, and covers the multidimensional model, the standard OLAP operations, warehouse architecture, and the schemas used to design one.
The data lifecycle describes the stages any piece of organizational data passes through, from the moment it is created to the moment it is finally disposed of. Five stages are usually distinguished. Creation is when data enters the organization, either through manual entry or through capture devices such as sensors. Storage follows creation: the data is placed under a security policy and a backup and recovery plan so that it survives hardware failure. Usage is the working phase, in which data is viewed, modified, saved and sometimes shared outside the organization to support day to day activity. Archival removes data from the active production environment while keeping a protected copy that can be brought back into use later if it is needed again. Destruction is the final stage, in which every remaining copy of a data item, including any copy held in archive storage, is permanently removed. A data warehouse sits mainly in the storage and usage stages of this lifecycle, since its purpose is to keep large volumes of already-created data available for analysis over a long horizon.
Data that an organization holds falls into three broad types based on how rigidly it is organized. Structured data is the easiest to search and manage because it is held in rows and columns, with every value mapped into a fixed, pre-defined field; relational databases and SQL are the natural tools for it. Unstructured data has no associated data model and cannot be placed into a row-column table, examples being photographs, video, audio and free text; it is harder to search and analyze, and is typically stored in NoSQL databases or in a data warehouse's raw zones rather than a relational schema. Semi-structured data sits between the two: it carries some organizing tags but still has internal variability, the standard example being an email message, whose sender, receiver and timestamp are structured fields while the body of the message is unstructured text.
A data warehouse is a repository that collects information from multiple source systems and stores it, together with its history, to support the decision maker who needs to model and analyze data rather than run day to day transactions. It forms the core of a Business Intelligence system built for reporting and analysis. Data warehousing is the process of building and maintaining that warehouse: it depends on ETL (extract, transform, load), the process that pulls data out of the different source systems, reshapes it into a common form, and loads it into the warehouse, and it requires the warehouse to be refreshed periodically as new source data arrives.
A data warehouse is characterized by four properties, due originally to Bill Inmon. It is subject oriented: rather than modelling the ongoing operations of the whole organization, it is built around a specific subject such as customer, product or sales, including only the data relevant to that subject and excluding what is not useful for understanding it. It is time variant: unlike an operational system, which typically keeps only the most recent value of a record, a warehouse retains historical data so that a value from three, six or twelve months ago can still be retrieved. It is integrated: it draws together heterogeneous sources such as relational databases, flat files and transaction logs, and the data must be cleaned and transformed during loading so that naming conventions and attribute types agree across sources. It is non-volatile: once data lands in the warehouse it is not overwritten in place, a new version is appended instead, so history is never lost, in contrast to an operational database that is constantly updated.
Data Warehousing and Data Mining for TU BSc CSIT is covered here in full, from syllabus notes to solved past questions. This subject has 9 chapters mapped to the official syllabus, and 70 past exam questions with written model answers. The opening of every note and answer is free, and the most-searched questions are free in full. A free account gets 2 free weeks of the whole program plus flashcards and progress tracking so weak topics come back before the exam.
This TU BSc CSIT Data Warehousing and Data Mining guide covers Introduction to Data Warehousing, Introduction to Data Mining, Data Preprocessing, Data Cube Technology, Mining Frequent Patterns, Classification and Prediction, Cluster Analysis, Graph Mining and Social Network Analysis, and more, across 9 chapters. It includes model answers to 70 past exam questions, plus flashcards for quick revision.
Data Warehousing and Data Mining has 9 chapters mapped to the official syllabus.
Yes. Each of the 70 past questions comes with a written model answer; the most-searched ones are free in full, the rest open with a free account's 2 free weeks or a pass.
Free to start. The opening of every note and answer is free, the most-searched questions are free in full, and a free account gets 2 free weeks of the whole program plus flashcards and progress tracking. After that a program pass keeps everything open.
Based on the official syllabus.