Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

ipt > decision support systems > Flashcards

decision support systems Flashcards

1
Q

what is a decision support system ? what does a DSS do

A

a decision support system (DSS) is designed to help managers make decisions that will further their organisations goal.

  • Assist users to make a choice between two or more alternatives by providing information, models and analysis tools
  • a DSS allows users to manipulate data directly, to incorporate data from external sources and to create data models of ‘what if’ scenarios
  • a DSS provides the support required so that correct decisions can be made. for example, a DSS could show trends in product sales over a three-year period so that material supplies and stock levels can be adequately maintained.
  • Supply evidence to assist decision makers determine alternatives and then prioritise one alternative over other possible alternatives
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

what are structured situations

A
    • decisions are automated
    • decision support systems are not required
    The solution to some problems can be clearly and definitely determined – all variables are clearly and thoroughly understood. These structured situations do not require decision support systems as they best alternative can be objectively determined. (think about continuum, structured is machine based). Structured decisions can be totally automated (that is, using machines). eg Which borrowers have overdue library books
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

what are semi-structured situations

A
  • there is a method to follow
  • requirements are clear cut

They follow a method to reach a decision, but the correct decision is not guaranteed. in these situations, the procedure for arriving at a solution is usually known; however, it might involve a degree of subjective judgement. there is some indication of the path to take but some of the information related to the problem may not be available, might lack precision or be uncertain. eg A bank officer deciding how much to lend to a customer. decisions involve some degree of uncertainty. most real-world situations are semi-structured and can use a DSS

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what are unstructured situations

A
  • there is no method to reach the decision
  • judgements are required
  • requires insights into the problem

this requires human intuition as a basis for finding a solution. there is no clear-cut path to the decision. information relevant to the solution might be missing, and few farts of the solution can be tackled using concrete models. No structured method for reaching a decision, too many variables, many are unknown and their interactions are highly complex and poorly understood. eg predicting stock prices

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

what are some examples of semi-structured situations

A
  • a bank officer deciding how much to lend to a customer
  • fingerprint matching
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

how does decision making in a bank office (deciding how much to lend to a customer) work

A

they assess the risk involved in lending the money and make a decision on the most suitable loan

  1. The customer’s income is sufficient to meet the regular repayments
  2. The customer’s income will continue at current levels for the term of the loan
  3. The bank will be able to recover their funds if the customer is unable to meet their repayment obligations
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

how does decision making in fingerprint matching work

A
  • Automated fingerprint identification system used to help fingerprint experts match prints
  • It generates a list of possible matchers , but an expert is still sometimes required to make a decision and use judgement regarding the degree of similarity between the 2 prints
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

what are some examples of unstructured situations

A
  • predicting stock prices
  • disaster relief management
  • weather forecasting
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

how does decision making in predicting stock prices work

A
  • Info. on share prices is stored in a spreadsheet or database used to predict future stock prices
  • Unforeseen events occur which affect decisions made, therefore decisions cannot be guaranteed
  • Data inputs to stock market prediction decision support systems include:
    • Past sale prices and quantity of shares traded for each public company’s shares
    • Data specific to individual companies
    • Industry specific data
    • Overall historical measures of stock market performance
    • Advice and predictions from politicians and stock market expert
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

how does disaster relief management work

A
  • Disasters unexpected and thus relief organisations generally respond after disaster has occurred
  • Decision making in these situations determine the effectiveness of the response to the disaster
  • DSS’s can
    • Store data describing the details of the disaster
    • Actions required to relieve the situation and resources available to perform these actions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

what are some systems used to support decision making

A
  • spreadsheets
  • databases
  • expert systems
  • neural networks
  • data warehouses
  • group decision support systems
  • geographic information systems (GIS)
  • management information systems (MIS)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

what are spreadsheets and what do they do

A
  • Organises data into one or more worksheets within a grid of columns and rows
  • Valuable for performing “what-if” analysis to alter inputs and view the effect on the outputs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

what are some advantages of electronic spreadsheets

A
  • calculation is quick and accurate using a large range of in-built functions
  • data is easily formatted and presented in a variety of outputs including charts or graphs
  • data can be copied within a spreadsheet or integrated into other spreadsheets or application programs
  • storage of data is faster and takes up less space
  • retrieval of data is fast and efficient
  • the spreadsheet can be used as a calculating tool in a large number of diverse situations
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

why do businesses use spreadsheets

A
  • perform cost-structure analysis
  • prepare break-even analysis
  • produce reports
  • show and forecast trends
  • prepare and store budgets
  • compile profit and loss statements
  • produce and store invoices
  • perform ‘what-if’ trials (such as changing prices or profit levels)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

why do scientists use spreadsheets

A
  • compile statistical research
  • produce graph results
  • test and simulate a range of variable conditions
  • build scientific models of data
  • perform ‘what if’ trials
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

what are the 3 types of data that may be entered into a cell in a spreadsheet (expand, what are they)

A
  1. labels - labels are non-numeric data. labels may be used as a spreadsheet title, column or row headings or any other words or text used to identify or clarify spreadsheet data. labels cannot be used in calculations
  2. values - values are numeric data that are used in calculations. it is possible to change the format of a value so that the data can be displayed in dollar, per cent, scientific or data forms.
  3. formulas - formulas carry out calculations using values in other cells. the formula itself is usually not displayed, only the value calculated is shown
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

what is absolute and relative referencing in formulas

A

formulas - formulas carry out calculations using values in other cells. the formula itself is usually not displayed, only the value calculated is shown
- relative cell reference means that a cell reference in a formula is relative to its position. if the formula is copied or moved to another location in the spreadsheet the formula will change relative to its new position
- absolute cell referencing means a cell reference in a formula is fixed and will not change when it is copied or moved. absolute cell reference is indicated by a $ preceding the cell reference, for example $A$3

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

how do databases help decision making

A
  • They enable users to access large amounts of data to make a decision.
  • Retrieving data to make a decision usually requires the use of a query (QBE / SQL).
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

what is a data warehouse

A

Data warehouse is a database that collects information from different data sources. It is a storage area of raw data that can be analysed to assist organisations to make decisions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

what is a data mart

A
  • Data Mart focuses on a single subject or functional organisation area and summarises the data
    • Meet the needs of an individual system or department in an organisation
21
Q

what is data mining

A

Data Mining is funding in-obvious patterns to make relationships and predictions

22
Q

what is a decision tree algorithm

A

identifies conditions that distinguish between data groups and creates a tree-like model to categorise the data based on these conditions. (view image in notes)

23
Q

what is rule induction

A

generates sets of rules that categorise data, with each rule independent of others. Rules can intersect, and the resulting model helps classify data into overlapping groups. (view image in notes)

24
Q

what is non-linear regression

A

fits a model to data, allowing predictions of unknown values. It goes beyond straight lines and uses curves or complex functions to capture relationships between variables. (view image in notes)

25
Q

what is K-nearest neighbour

A

classifies data by comparing the new item to previously classified items and assigning it to the class that has the majority of its K nearest neighbours. (view image in notes)

26
Q

what are expert systems? what are two important components of expert systems?

A
  • They provide information and solve problems that would otherwise require a person experienced in that field (an expert).
  • A software application that simulates the knowledge and experience of a human expert
  • Their conclusions are not guaranteed so it’s up to the user to accept or reject the decision.
  • Low cost compared with the expense of paying an expert specialist team.
  • Two important components are:
    • The knowledge base and the inference engine – a set of general facts and if-then rules supplied by an expert.
    • The system generates questions using the knowledge base and carries out reasoning using the inference machine
  • Errors can occur as the expert system cannot adapt to a changing environment
27
Q

what are neural networks

A
  • A DSS that uses artificial intelligence to make decisions with the capability of learning
    • An attempt to simulate the complex structure and processes performed by the human brain
    • They process things by going through the input layer, hidden layer, and output layer
    • Trained using sets of sample inputs together with known outputs
    • Once trained the network is able to determine the most likely outputs based on new unseen input data from interacting with new users
    • They have to ability to match, generalise, cope with new situations
    • Well suited to unstructured situations but need lots of processing time
28
Q

what are some fields that use neural networks

A
  • Neural networks have applications in numerous fields including:
    • Economic forecasting by financial firms
    • Decision support systems for business, government, research and science
    • Weather forecasting systems
    • Voice recognition systems
    • Predicted sales of products
    • Handwriting recognition
29
Q

what are group decision support systems

A
  • Each participant has access to a computer and can contribute to ideas and discussion to the meeting without one or more participants dominating the meeting
  • The computer sorts the ideas and displays the results for more discussion or a vote
  • The GDSS can provide summaries and notes from the meeting
  • Increases productivity and efficiency of the group
30
Q

what are geographical information systems

A

Geographical information systems
- Represents data using maps
- Example, Google Earth
- Collects data from various sources and interprets it and displays the result
- Can assist users to uncover patterns, relationships and trends by letting users look at, understand, question and interpret data in a variety of forms – such as, maps and globes as well as graphs, charts and reports
- GIS software has the ability to:
- Use maps to show features and their relationships to the earth’s surface
- Plot features and landmarks on top of maps
- Represent different information over the maps such as population density
- Allow users to zoom in and out on areas of interest
- Display text when users scroll over certain positions
- Operate with a GPS system
- Extract information through data mining tools

31
Q

what are management information systems

A

Management information system
- A general term for the computer systems that provide information about the operations of a business
- Can summarise the data and provide past, present and prediction information to help in decision making – enabling the organisation to run more efficiently

32
Q

what are some examples of sources of data

A
  • Sources of data include:
    • Internet – web sites and newsgroups
    • Interviews and surveys (people)
    • Observations and measurements
33
Q

a well designed spreadsheet has what four areas

A
  • Instruction/documentation area: provides information about the spreadsheet or directions for the use of the spreadsheet. Usually at the top and includes a title, description, the authors and the creation/revision date.
  • Input area: includes labels for headings and the values on which calculations are based.
  • Calculation area: contains the formulas and functions that complete the work of the spreadsheet.
  • Output area: displays the results of the spreadsheet
34
Q

what is a formula

A
  • A formula is an instruction to perform a calculation.
  • Whenever values are changed, the formulas are recalculated and produce new results.
  • Built-in functions are more effective than user-developed formulas
  • When a formula is copied, the content in its new location may change or it may remain exactly the same depending on the use of absolute and relative referencing
35
Q

what are some examples of formulas or other things used in spreadsheets

A

check notes, DSS

36
Q

what are if then rules in an expert system

A
  • A set of general facts and if-then rules supplied by an expert
  • If the condition is true, then a certain deduction
  • Knowledge base is often constructed using expert system shells that are a ready-made expert system except that they contain no knowledge
  • They provide an interface to assist the user in creating an expert system. A majority of shells represent knowledge using if-then rules.
  • When a set of if-then rules is completed, the shell builds a knowledge base and an inference engine. They allow quick development
  • E.g: IF it has four legs AND it has a tail AND it has fur AND it barks THEN it is a dog
37
Q

what are some things used in an expert system

A
  • knowledge base
  • database of facts
  • inference engine
  • explanation mechanism
  • user interface
38
Q

what is a knowledge base in an expert system

A
  • a knowledge base which holds the coded rules, concepts and models for the system.
  • contains rules for the system in the: IF<condition> THEN <response></response></condition>
  • heuristics is the use of commonsense rules, drawn from experience to solve problems. this is in contrast to algorithmic programming, which is based on mathematically provable procedures
39
Q

what is a database of facts in an expert system

A
  • The data that relates to the specific situation being analysed.
  • It is used by the inference engine when it is working on a problem
  • Stores the known and learnt facts
  • Some of the facts were inputted by the users as they answer questions
  • Records which rules have been fired and in what order, which is useful for the explanation mechanism
40
Q

what is an inference engine in an expert system

A

an inference engine is a processor of coded knowledge within the expert system. within any system there are processes and rules to be followed for a solution to be reached. the inference engine decides how and when the rules are to be applied. it accesses the knowledge base and, through a series of processes, attempts to produce an answer or solution. the inference engine will prompt the user for more information to determine which rules to process next, or for confirmation of its findings. the inference engine may apply the rules through forward or backward chaining.

whether forward or backward chaining is used depends on the problem being solved and how many possible conclusions there are. it is the inference engine that applies the forward or backward chaining, so it controls the strategy used and either confirms or negates the facts input

41
Q

what is forward chaining

A

a forward chaining expert system starts with the known facts and then applies relevant rules until it arrives at a conclusion. by starting with the known facts and applying rules, it can search for a match as various conditions are met. forward chaining could be used in a science experiment where various materials are to be combined

42
Q

what is backward chaining

A

a backward chaining system starts with the conclusion or believed fact. it then works back through the rules to see if all the conditions are satisfied. if the facts all support the conclusion, it will confirm that the conclusion is appropriate

43
Q

what is an explanation mechanism in an expert system

A
  • Expert systems are able to explain how they reached conclusions. Essentially the explanation is a replay of the inferences made by the inference engine. Inferences occur every time a rule fires and new facts are established.
  • Evidence of how it got to the conclusion
44
Q

what is a user interface in a decision support system

A
  • Where the user and software interact and communicate
  • The interface makes use of rules within the system to output questions to the user and to indicate the status of the solution
  • outputs the conclusion when found
45
Q

what are certainty factors

A

certainty factors as a means of dealing with unclear situations
- Describes the level of certainty that a fact or a consequent is correct on the scale of 0 - 1
- Certainty factors are specified directly as part of each consequent and they can also be entered by the user as they answer questions
- E.g: If it is known that Kermit is green it might be concluded with 85% confidence that he is a frog, or if it’s known he is a frog then 95% sure that he hops
- These numbers are meant to imitate the confidences humans use in reasoning rather than to follow the mathematical definitions used in calculating probability

46
Q

what are artificial neural networks (ANNs)

A
  • Artificial neural networks (ANNs) simulate the organisation, analysis and processing information processes performed by the human brain
  • ANNs are useful for pattern matching applications. an image or stock quotes are both examples of patterns
  • ANNs are trained using sample data that includes the desired results
  • Once trained the ANN can recognise the words within bitmaps even when the handwriting and fonts are different

pattern matching in neural networks
- Optical Character Recognition (OCR) is largely a pattern matching exercise – problems that involve such pattern matching decisions are well suited to solution using ANNs

47
Q

what are macros and how might they be used in spreadsheets

A

Macros are used to automate processing in all types of applications including spreadsheets. macro is a sequence of recorded or written instructions that automate repetitive tasks or perform a series of actions with a single command. macros have many uses and can considerably speed up data entry. macros can be used in spreadsheets to automate tasks such as:

  • opening and closing menus
  • entering text, by moving the cursor to the next appropriate cell, or by offering a dialog box
  • moving the cursor around the spreadsheet
  • formatting cells
  • sorting data
  • moving between worksheets
  • creating graphs and charts
48
Q

compare and contrast processing methods used by databases, neural networks and expert systems

A

Databases, neural networks, and expert systems employ distinct processing methods, each suited to their specific purposes. Databases primarily focus on efficient data storage, retrieval, and management. They use structured query language (SQL) to interact with data and perform operations like filtering, sorting, and joining. Databases excel at handling large volumes of structured data and ensuring data integrity.

On the other hand, neural networks leverage interconnected nodes or artificial neurons to simulate human-like learning and pattern recognition. They excel at processing unstructured or complex data, such as images, audio, and text, through layers of mathematical computations. Neural networks can generalise from training data to make predictions or classify new inputs, enabling tasks like image recognition, speech synthesis, and natural language processing.

Expert systems, also known as knowledge-based systems, rely on rules and knowledge bases to mimic human expertise. They employ an inference engine to reason, make deductions, and provide solutions. Expert systems use a collection of rules and facts to analyse problems and deliver recommendations or insights in specific domains. They excel at capturing and applying domain-specific knowledge, aiding in tasks such as medical diagnosis, troubleshooting, or decision-making.

ipt (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions