CS614 Final Term Current Subjective by Masters

2024 Feb

Mcq's 30 sy ziada Junid Wali file sy thy Mostly mcqs from junaid file and some from handouts Mcqs handsout but 10 from past files Mostly from Junaid file. MCQS mostly from Junaid file Mostly mcqs from junaid file and some from handouts

Q. Three dimensions context of web warehouse

1. Three dimensions common in context of Web data warehouse.

Page, Event, Session, Referral OR Sure, here are three dimensions for a web data warehouse:

1. Page Dimension: This refers to the specific webpage where an event occurs. It’s flexible to accommodate both static and dynamic pages.
2. Event Dimension: This captures what action a user takes on a webpage, like clicking a link or filling out a form.
3. Session Dimension: This tracks a user’s entire visit to a website, from the moment they arrive until they leave.
4. Referral: Indicates where users came from before landing on the website, providing insights into traffic sources and marketing effectiveness.

Each dimension provides a different perspective for analyzing user behavior on a website.

Quality validation techniques

1. Referential Integrity (RI): Ensures that relationships between data tables are maintained, preventing orphan records.
2. Attribute domain: Validates that data values fall within predefined ranges or types.
3. Data Quality Rules: Apply specific criteria to data to check for accuracy, completeness, and consistency.
4. Data Histograming: Visualizes data distributions to identify anomalies or patterns for validation purposes.

Project planning and design

1. Project Planning: This is the first step where you assess your organization’s readiness for a data warehouse project, establish the scope, justify the costs vs benefits, and form a team.
2. Design: This involves understanding business requirements, setting up the technical architecture, selecting products, designing the data model, and planning for data staging and analytic applications. It’s all about translating business needs into a functional design.

OR

Project planning involves assessing readiness, scoping the project, justifying costs, assembling the right team, and creating a detailed project plan with milestones and deliverables. Designing the data warehouse includes defining business requirements, creating a technical architecture, selecting

CS614 Final Term Current subjective shared by Students solved by Masters

appropriate products, dimensional modeling, physical design, and developing ETL processes to stage data effectively. These steps ensure alignment with business needs and technical capabilities for a successful data warehouse implementation.

Q. Classes of anomalies (lexical error and missing record) Ek me lexical and missimg record se

Q2. Lexical error and Missing attributes ki definition lekhni thi. (3 marks)

Classes of anomalies mn lexical error and missing records (3 marks)

1. Lexical Error: This is a type of syntactic anomaly where the data does not match the expected format. For example, if a table is supposed to have five columns but some rows only have four, that’s a lexical error.
2. Missing Record: This is a type of coverage anomaly where an entire record (or row in a table) that should be present in the data is missing. This could be due to issues during data collection or processing.

OR Lexical errors refer to mistakes in the format of data, like having fewer columns in a table row than expected. Irregularities involve inconsistencies in the use of values, units, or abbreviations, such as using different currencies to denote salaries. Missing records are instances where expected data entries are absent, leading to gaps in the dataset, requiring careful consideration to determine if they represent real-world absences or errors in data collection.

1. Three parallel track Kimball approach

Q1. kimball parallel tracks types. (3 marks)

Q2. Three tracks name Kimball approach (3 marks)

In Kimball's approach to data warehouse development, there are three parallel tracks:

1. Technology Track: Focuses on designing the technical architecture to support integration of various technologies. It involves evaluating and selecting specific products based on the identified capabilities.
2. Data Track: Translates business requirements into a dimensional model (data model), which is then transformed into a physical structure. Activities include performance tuning strategies like aggregation, indexing, partitioning, and designing ETL processes for data staging.
3. Analytic Applications Track: This track is about designing and developing applications that will analyze the data stored in the data warehouse. This track focuses on creating tools and interfaces for users to analyze and derive insights from the data

Q2. 3 task of DT?

here are three tasks that are typically part of Data Transformation Services (DTS):

1. Establishing Connections: This involves setting up the necessary connections to the source and destination databases. Setting up connections to source and destination databases.
2. Data Import and Export: This task involves extracting data from the source and loading it into the destination. It’s like moving data from one place to another.
3. Data Transformation: This task is about changing the data in some way. For example, it might involve cleaning up the data, or converting it into a different format. It’s like changing the data to make it more useful or easier to work with. Transforming data: Modifying the format or content of data during the transfer process.

Q3. Write names of DTS items (5 marks)

Here are the names of five items in Data Transformation Services (DTS):

1. DTS Packages: These are collections of connections, tasks, and workflows that are executed together. Collections of tasks and transformations for data movement and transformation.
2. DTS Tasks: These are individual units of work within a package, such as establishing connections, importing/exporting data, or transforming data.
3. DTS Transformations: These are operations that change the data in some way during the data movement process. Processes within tasks that modify the format or content of data.
4. DTS Package Workflows: These define the order in which tasks are executed within a package. Sequences of tasks executed in a specific order to accomplish a task or process.
5. DTS Tools: These are utilities or applications used to create, manage, and execute DTS packages. Software interfaces and utilities used to create, manage, and execute DTS packages and tasks.
6. Meta Data: Information about the structure and characteristics of data stored within a DTS package or task.
7. Two parallel Software Architecture

Parallel Software Architectures Shared Memory | Shared everything Shard Disk | Shared everything Shared Nothing

5. Two main types clustering in data mining

Q6. Two types of clustering in mining (3 marks)

Two main types are one way clustering and two way clustering. One way clustering is when only data records (rows) are used. Two way clustering is when all the rows and columns are being used for clustering purpose. OR

1. One-way Clustering: This method uses all the attributes of the data and clusters based on similarity among rows. It gives a global view of the data.
2. Two-way Clustering (Biclustering): This method clusters both rows and columns simultaneously, providing a local view of the data. It’s useful for getting more detailed insights from specific clusters.
3. Cost of performing nested loop join

The cost of performing a nested loop join is calculated based on the number of blocks accessed in both tables involved in the join. It’s the sum of the cost of accessing the first table and the product of the number of qualifying rows in the first table and the blocks accessed in the second table. The order of tables matters, as changing the order can significantly affect the cost. For example, if Table A has 500 blocks and Table B has 700 blocks, the join cost of A&B could be 35,500 I/Os, while the join cost of B&A could be 50,700 I/Os, an increase of about 43%.