20 Systems Development

Question 1

Software Development Life Cycle 5 Steps

Answer 1

1. Systems Strategy- Understand Needs
2. Project Initiation - Proposals Assessed
3. In House - chosen for unique needs
4. Commercial - for common needs
5. Maintenance and support

Question 2

Testing

Combinatorial Testing

Answer 2

Identifies Minimal number of tests to get coverage Developers want

Question 3

Testing

Static Testing

Answer 3

Examines: Programs Code and Documentation thru Review. Not actually executing the program

Question 4

Testing

Dynamic Testing

Answer 4

Executing the Code with Test Cases

Question 5

Testing

White Box Testing

Answer 5

Testing INTERNAL workings, not functionality of end user

Question 6

Testing

Black-Box Testing

Answer 6

How well the software Works. Tests FUNCTIONALITY without knowledge of Code

Question 7

Testing

Gray Box Testing

Answer 7

Both: knowledge of internal structure while testing functionality

Question 8

Four Levels of Tests

Answer 8

Unit Testing: functionality of specific section of code.

Integration Testing: Verify interfaces between Components. Expose defects in interfaces/Interactions of modules

System Testing: End to End. the completely integrated system meets requirements

Acceptance Testing: Whether meets org’s Needs/ready for Release

Question 9

Four Strategies for Conversion

Answer 9

Parallel: both new/old run. Safest/Most Expensive

Direct Changeover: shuts old/starts new. Riskiest/Least Expensive

Pilot: One branch/division at a time

Phased: One function at a time (A/P, A/R)

Question 10

Sandbox

Answer 10

When a program is to be changed, copy is saved to test area. Programmer makes change to this copy

Question 11

RPA Robotic Process Automation

Answer 11

Machine Learning Technology that acquire knowledge and mimic people tasks.

Benefits: perform continuously (no time off)
Eliminate human error

Question 12

AI

Types of Artificial Intelligence

Answer 12

Neural Networks: Processing Elements working together to mimic human brain including learning from previous

Case Based Reasoning: Learn from previous

Rule Based: set of rules to arrive at answer

Intelligent Agents: Apply knowledge base to execute specific, repetitive task

Expert System: attempts to imitate human brain, unstructured problems

Question 13

Benefits of AI to Accounting

Answer 13

1. Automate Data Entry
2. Reduce fraud by viewing vast amounts of documents
3. Strengthen Expenditure Disbursements by viewing ENTIRE POPULATION of documents.

Question 14

Categories of Cloud Services

Answer 14

IASS Infrastructure as A Service

PASS Platform as a Service - provides software and hardware tools (typically for application development)

SAAS Software as a Service

Question 15

Business Intelligence

Answer 15

Collectin of applications, tools, best practices that

• transform data into actionable info for
  a. managerial control
  b. strategic planning
  c. making better decisions

Question 16

Data Analytics

Answer 16

• Qualitative and Quantitative Methods to retrieve data

- identify deviations, opportunities

Question 17

Data Analytics

5 stages

Answer 17

1. Questions Defined
2. Relevant Data obtained (Info Discovery)
3. Clean Data- flushing out useless, ID missing
4. Analyze Data
5. Communicate Results

Question 18

Data Analytics
Types

DDPP ANT

Answer 18

Descriptive - most basic. reports actual Results

Diagnostic - provides insight into REASONS RESULTS OCCURRED.

Predictive - apply assumptions/predict future results

Prescriptive - what needs to be done to make predictive happen

Anomaly - detect unusual Patterns

Network Analysis - analyze network data for patterns

Text Analysis - Text mining to find patterns IN UNSTRUCTURED TEXT

Question 19

DATA MINING

Search for unexpected Relationship

Answer 19

1. Anomalies detected
2. Finding Relationships between variables
3. Classifying Data
4. Regression Analysis peformed
5. Summarize Data

Question 20

Regression Analysis

Goodness of Fit

Answer 20

Whether sample is representative of POPULATION

• Confidence Level - % of times sample expected to be representative of population
• Confidence Interval - for given confidence level, range around sample expected to contain TRUE POPULATION

Question 21

SENSITIVITY ANALYSIS

Answer 21

• Trial and Error to determine changes of variables or assumptions on final result

Question 22

Risk

Simulation

Answer 22

Computer used to generate many results based on various assumptions

Question 23

Risk

Monte Carlo

Answer 23

Randomly selecting values based on probability distribution

Question 24

Risk

Delphi Approach

Answer 24

opinions from experts
Summarize Results
Back to experts
Repeat until opinions converge

Question 25

Time Series Analysis (Trend Analysis)

Answer 25

Predict future trends based on PAST EXPERIENCES Seasonal Pattern- within known periods repeats Cyclical: rise/fall not on a fixed patterns Usually at least a couple of years

Question 26

EDA Exploratory Data Analysis

Answer 26

Encourage data to reveal itself rather than prematurely applying a hypothesis or statistical method

Question 27

What-If Analysis (Goal Seeking)

Answer 27

determining outcome thru changed scenarios. Goal seeking - when an outcome is wanted and needs to determine how it can be achieved

Question 28

Big Data

Answer 28

- voluminous amount of Structured, Semi-structured or unstructered data characterized by 4 V's - Volume (large amount of data) - Variety - wide variety of File Types - Velocity - Speed at which it must be analyzed - Veracity - Trustworthiness of data 5th V - Value : only as valuable as biz outcomes made possible Volume Based Value - More data biz have on Customers, greater insights Variety Based Value - Velocity Based Value - faster data processed, more time will have to ask questions and seek answers

Question 29

Key Technologies

Answer 29

Data Management: Data needs to be high quality to analyze Data Mining: examining large amounts of data to discover patterns Hadoop: Open sournce framwork that stores large amts of data In-Memory Analytics: Delivers IMMEDIATE results by removing data preoparation Predictive Analytics: Text Mining: Analyzes data from web, comment fields, other text based to identify new topics and relationships

Question 30

Limitations of Big Data

Answer 30

User-Level results are INCOMPLETE: limited to data available from org's website for example Answer to 'WHY?" difficult Data subject to "noise" > useless info user level data requires INTERPRETATION prior to use

Question 31

Systems Analysis

Answer 31

Learning how current system functions, Determining user needs. Developing logical Requirements of proposed system

Question 32

Feasibility Study

Answer 32

Technical, operational and Economic feasibility

Question 33

Systems Design

Answer 33

Developing Specifications for - Input/Output - Processing - Internal Controls - Programs/Procedures