QM457 : Midterm Revision Exam

[CH1] Data analytics is a broader term than data analysis, because it covers the management of the complete data lifecycle: collecting, cleansing, organizing, storing, analyzing and governing data.

[CH1] Value and complexity decrease as an organization moves from descriptive analytics towards prescriptive analytics.

[CH1] Diagnostic analytics aims to determine the cause of a phenomenon that occurred in the past, using questions that focus on the reason behind the event.

[CH1] Veracity was part of Doug Laney's original 2001 list of Big Data characteristics, together with volume and velocity.

[CH1] Data with a high signal-to-noise ratio has more veracity than data with a lower signal-to-noise ratio.

[CH1] Semi-structured data, such as JSON and XML, is less complex than structured data but more complex than unstructured data.

[CH1] Human-generated data results from human interaction with systems such as online services, whereas machine-generated data is produced by software programs and hardware devices in response to real-world events.

[CH2] In the DIKW pyramid, wisdom is drawn as an outline because it is typically generated by ICT systems.

[CH2] In the layered view of a business, the middle layer is the tactical or managerial layer, which seeks to steer the organization in alignment with the strategy.

[CH2] The combination of commodity hardware and open-source software has increased the advantage that large enterprises hold over smaller competitors because of their bigger IT budgets.

[CH2] Because of social media, product brands and corporate reputation are now co-created by the company and its customers.

[CH3] Confirmatory data analysis is an inductive approach closely associated with data mining, in which no hypothesis is generated beforehand.

[CH3] Whenever Big Data changes state, for example from data-in-motion to data-at-rest, it should trigger the capture of provenance information that is recorded as metadata.

[CH3] Unlike traditional relational database management systems, NoSQL databases generally provide robust built-in security mechanisms.

[CH3] Business Case Evaluation is the first stage of the Big Data analytics lifecycle.

[CH1] A retail manager asks the analytics team: "Why were Q2 sales less than Q1 sales?" Which type of analytics answers this question?

[CH1] A bank builds a model to estimate the chances that a customer will default on a loan if they have missed a monthly payment. Which type of analytics is the bank using?

[CH1] A hospital system compares three drugs, recommends which one should be given because it provides the best results, and explains why. Which type of analytics is this?

[CH1] Which type of analytics provides "hindsight", answers questions about events that have already occurred, and is often carried out through ad-hoc reporting or dashboards?

[CH1] An ice-cream shop examines how the number of cones sold relates to the daily temperature so it can decide how much to order based on the weather forecast. This activity is best described as:

[CH1] A metric that is linked to an enterprise's strategic goals and acts as a quantifiable reference point for measuring a specific aspect of business performance is called a:

[CH1] A ride-hailing company receives GPS updates from thousands of drivers every second, and its systems struggle to keep up with how fast the data accumulates. Which Big Data characteristic is the main challenge?

[CH1] A company must integrate financial transactions, customer emails and product images coming from different sources into one analytic environment. Which Big Data characteristic does this mainly reflect?

[CH1] A trader notices that a stock quote delayed by 20 minutes is almost useless for making a trade, while a quote that is 20 milliseconds old is very useful. Which Big Data characteristic does this illustrate?

[CH1] A supplier sends product records in files that use tags and semantic markers to separate the data into records and fields, but the files follow no relational or tabular model. This data is best classified as:

[CH1] Organizations and users worldwide create over 2.5 exabytes of data a day from sources such as point-of-sale transactions, scientific experiments, sensors and social media. Which Big Data characteristic does this describe?

[CH2] According to the DIKW pyramid, information becomes knowledge when it is:

[CH2] A manager notices that decision-making takes too long, there are too many meetings, email overload and departments working in silos. In business architecture terms, these are:

[CH2] In the layered view of a business, which layer is occupied by C-level executives and advisory groups?

[CH3] An auditor wants to know where a dataset originally came from and which processing steps and algorithms produced a reported analytic result. Which type of information does the auditor need?

[CH2] A bank replaces paper statements with online banking, then uses the online channel to ask customers to complete surveys and to track the click-through rate of displayed advertisements. Which business driver of Big Data does this reflect?

[CH2] A small Bahraini start-up leases highly scalable storage and processing resources on a pay-as-you-go basis to run large-scale analytics instead of buying servers. Which business driver of Big Data is this?

[CH2] Smart refrigerators, RFID tags, GPS devices and smart meters that continuously stream data to corporations are examples of which business driver?

[CH2] In the Internet of Everything (IoE), which element refers to the devices and objects that feed data into the IoE and use that data to perform actions?

[CH2] In a Business Process Management System (BPMS), the state of an individual process, or of all processes, can be interrogated and visualized through:

[CH3] A telecom company's analysts start with no hypothesis and explore customer data to find any unusual patterns or anomalies that might explain rising churn. Which approach are they using?

[CH3] A logistics company finds that, as its data volumes grow, the time needed to transfer data across its network now exceeds the time needed to actually process it. Which Big Data adoption consideration does this illustrate?

[CH3] A firm jointly analyzes two seemingly harmless datasets, smart meter readings and car GPS logs, and discovers the locations and behavior of specific individuals. Which Big Data adoption consideration is raised?

[CH3] A company's NoSQL database exchanges data in plaintext through simple HTTP-based APIs, making the data prone to network-based attacks. Which Big Data adoption consideration does this concern?

[CH3] Which stage of the Big Data analytics lifecycle is dedicated to integrating multiple datasets into a unified view, for example joining two datasets using a shared Id field?

[CH1] List and explain the four types of data analytics. For each type, give one example question it answers, and state how value and complexity change across the four types.

[CH1] Define and explain the five Vs of Big Data.

[CH1] Compare structured, unstructured and semi-structured data. Give examples of each and explain where each sits in terms of complexity.

[CH2] Explain the DIKW pyramid and show how it aligns with the operational, tactical (managerial) and strategic levels of a business.

[CH2] List the six information and communications technology (ICT) drivers of Big Data adoption, and discuss any four of them.

[CH3] List and briefly explain the nine stages of the Big Data analytics lifecycle in the correct order.

[CH3] Compare confirmatory data analysis and exploratory data analysis.

[CH3] Discuss privacy, security and provenance as Big Data adoption and planning considerations.

[CH1] Classify each of the four data sources collected by Al-Noor Hypermarkets as structured, semi-structured or unstructured data, and justify each classification.

[CH1] Identify the type of data analytics needed to answer each of the operations director's three questions, and explain your choice.

[CH3] Which stage of the Big Data analytics lifecycle should deal with the duplicated records and missing customer IDs? Based on the notes, advise the team on how to treat the sudden spikes in the Muharraq figures.

[CH3] Using the common justifications listed in the notes, explain why Dilmun Logistics should host its Big Data solution in a cloud environment.

[CH3] Identify the privacy issue and the security issue in the Dilmun Logistics case, and recommend how each should be addressed according to the notes.

[CH2] Identify three business drivers of Big Data adoption that are evident in the Dilmun Logistics case, linking each to specific case details.