Create
Learn
Share

Chapter 8

rename
imissyou419's version from 2017-04-17 17:38

Section

Question Answer
Information overloadthe issue that arises when managers have an overabundance of information to use when making decisions
Russel Ackoff 3 assumptions as to how managers make decisions1. managers will have no problem making decisions if they have the data they need,
2. managers may make poor decisions either because of a lack of information or because of an abundance of information,
3. managers know the data they require
Kilobyte 1,000 bytes (100 KB = a low-resolution photograph)
Megabyte1,000,000 bytes (500 MB = CD-ROM)
Gigabyte1x10^9 bytes (100 GB = a library floor of academic journals)
Terabyte1x10^12 bytes (400 TB = national climactic data centre database)
Petabyte1x10^15 bytes (200 PB = all printed material)
Exabyte1x10^18 bytes (5 EB = all words ever spoken by human beings)
Experts suggest that data is growing at a rate of _________% each year30%
Dirty dataproblematic data (data that does not make sense)
Examples of issues with datadirty data, missing values, inconsistent data, wrong granularity (too fine or not fine enough), too much data (too many attributes or too many data points)
Granularity of datarefers to the degree of summation or detail (coarse data are highly summarized, fine data express details that are too precise)
Clickstream datafine data and includes everything on which the customer clicks (tracking their behaviour)
Online transaction processing (OLTP) - real time, batch processinga system that is used to process transactions online, as opposed to paper (transactions can either be processed immediately - real time processing, ticketmaster or they can be grouped and then progressed together - batch processing, gas station)
Data resource challengethe challenge of using collected data effectively in order to improve decision making; occurs when data are collected in OLTP but are not used to improve decision making; use OLAP to solve this challenge
Decision support system (DSSs) or Online Analytical Processing (OLAP)focuses on making OLTP-collected data useful for decision-making; OLAP has the ability to sum, count, average, and perform other simple arithmetic operation on groups of data
OLAP cubean array of data understood in terms of its 0 or more dimensions; same thing as OLAP report
Drill downto further divide the data into more detail; can drill down into data for OLAP report
Business Intelligence (BI) systema system that provides information for improving decision making, 5 categories: GDSSs, reporting system, data-mining system, knowledge management systems, expert systems
Group decision support systems (GDSSs)allow multiple parties to participate in decision making and improve outcome by reducing often inherent biases
Reporting systemsintegrate data from multiple sources and process those data by sorting, grouping, summing, averaging, and comparing; format the results into reports and deliver those reports to users; improve decision making by providing the right info to the right user at right time
Data-mining systemsuses sophisticated statistical techniques to find patterns and relationships, improve decision by discovering patterns and relationships in data to predict future outcomes
Knowledge management systemsshares knowledge of product, product uses, and best practices among employees, managers, customers, and others; improve decision by publishing employee and others' knowledge and create value from existing intellectual capital thereby fostering innovation, improving customer service, increasing organizational responsiveness and reducing costs
Expert systemsencode human knowledge in the form of If/Then rules and process those rules to make a diagnosis or a recommendation; improve decision making by non-experts by encoding, saving, and processing expert knowledge
Market-basket analysisa data-mining system that computes correlations of items on past orders to determine items that are frequently purchased together
Toolsallow data to be processed into information (e.g. a reporting tool can generate a report showing that a customer has cancelled an important order) just think about a tool is just computer, it takes a person to be system that's why to reap improved decision making, must incorporate tools into IS
Systemsallow people to effectively utilize that processed data (e.g. reporting system alert the customer's saleperson to this unwanted news in time for the saleperson to attempt to reverse the decision, use equation from tool to enable banking person to approve or reject loan on the spot)
RFM analysisa way of analyzing and ranking customers according to their purchasing patterns (sort customers based on their most recent purchases, the frequency of their purchases, the money that they spend on each purchase, give values 1-5, smallest RFM values = high value customers)
Data warehouseare used to extract and clean data from operational systems and other sources and to store and catalogue that data for processing by BI tools
programs read operational data and extract, clean, and prepare that data for BI processing
stored in data warehouse database using a data warehouse DBMS (can be different from organization's operation DBMS)
also incl. data purchased from outside source i.e. credit card data
facility for storing data for use by others
Metadatadata about data
Data marta data collection that is created to address the needs of a particular business function, problem, or opportunity (data warehouse as distributor and data mart as retailer in a supply chain) i.e. store sales data mart, web sales data mart, inventory data mart
Data miningthe application of statistical technique to find patterns and relationships among data and make classifications and predictions
Unsupervised data miningis when analysts do not create a model or hypothesis before running the analysis - instead they apply the data mining technique to the data and observe the results; 1 technique is cluster analysis
Cluster analysisstatistical technique (unsupervised) to identify groups of entities that have similar characteristics; common use is to find groups of similar customers from customer order and demographic area
Supervised data miningis when data miners develop a model prior to analysis and apply statistical techniques to data to estimate the parameters of the model
Regression analysisan analysis that measures the impact of a set of variable on another variable; part of supervised data mining i.e. how old they are impacting their phone usage
Neural networksa data mining technique (supervised) used to predict values and make classifications i.e. good prospect or poor prospect customer
Lift - Market Basketshows how much the base probability increases or decreases when other products are purchased
Confidence - Market Basketconditional probability estimate
Support - Market Basketprobability that 2 items will be purchased together
Big datalarge amount of varied data from a variety of sources over a period of time could be used to make better decisions; controversial, hard to define, adds to excessive data collection, expensive, occasionally results in predictions that do not stand test of time, overly vague, or confuse correlation with causation
memorize

Recent badges