Introduction to Statistics for Information Systems

Lecture Notes

(By,
Selamysfen D. )

University of Gondar

October 8, 2025

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 1 / 40
 Outline

1 Chapter 1: Introductions
2 Chapter 2: Measures of Central Tendency
3 Chapter 3: Measures of Variation
4 Chapter 4: Introduction to Probability Theory
Counting Rules
Probability
Total Probability and Bayes’ Theorem
Random Variables and Probability Distribution
5 Chapter 5: Discrete Probability Distributions
Mean and Variance of Discrete Probability Distributions
The Binomial Distribution
The Poisson Distribution

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 2 / 40
 Definition & Classifications

Statistics: Science of collecting, organizing, analyzing, and

interpreting data for decision-making.
Relevance to IS: Used in database analysis, user behavior tracking,
and system performance evaluation.
Data Classifications:
Qualitative: Non-numerical (e.g., user roles: admin, guest).
Quantitative: Numerical.
Discrete: Countable (e.g., number of logins).
Continuous: Measurable (e.g., response time in ms).
Levels: Nominal, Ordinal, Interval, Ratio.
Example: Classify “website traffic (hits/day)”.
Answer: Quantitative, discrete, ratio.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 3 / 40
 Data Collection & Organization

Methods:
Surveys: User feedback forms.
Experiments: A/B testing for UI designs.
Observational: Server log analysis.
Existing Records: Database queries.
Organization: Use frequency distributions.
Example: Server response times (ms): 120, 130, 120, 150, 140.

Response Time Frequency

120 2
130 1
140 1
150 1

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 4 / 40
 Data Representation

Graphical Tools:
Histogram: Continuous data (e.g., response times).
Bar Chart: Categorical data (e.g., user types).
Pie Chart: Proportions (e.g., error types).
Box Plot: Summarizes spread and outliers.
Example: Box plot for response times: 120, 130, 120, 150, 140.
Median: 130.
Q1: 120, Q3: 140.
IQR: 140 - 120 = 20.
IS Application: Visualizing system performance metrics in
dashboards.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 5 / 40
 Mathematical Measures

P
xi
Mean: Average, x̄ = n . P
(xi ·wi )
Weighted Mean: Accounts for importance, x̄w = P
wi
.
Example: Mean response time for 120, 130, 120, 150, 140.
120+130+120+150+140
x̄ = 5 = 132 ms.
IS Application: Average user session duration in a system.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 6 / 40
 Positional Measures

Median: Middle value in ordered data.

Midrange: min+max
2 .
Example: For 120, 130, 120, 150, 140.
Ordered: 120, 120, 130, 140, 150.
Median: 130.
Midrange: 120+150
2 = 135.
IS Application: Median response time is robust to extreme values
(e.g., system spikes).

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 7 / 40
 The Mode

Mode: Most frequent value.

Example: For 120, 130, 120, 150, 140.
Mode: 120 (appears twice).
IS Application: Identify most common error code in logs.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 8 / 40
 Positional Measures

Quartiles: Q1 (25th percentile), Q2 (median), Q3 (75th percentile).

IQR: Q3 − Q1.
Example 1: Server response times (ms): 120, 130, 120, 150, 140.
Ordered: 120, 120, 130, 140, 150.
Q1: 120, Q2: 130, Q3: 140.
IQR: 140 − 120 = 20.
Example 2: Database query times (ms): 50, 60, 55, 70, 65, 75, 80.
Ordered: 50, 55, 60, 65, 70, 75, 80.
Q1: 55, Q2: 65, Q3: 75.
IQR: 75 − 55 = 20.
IS Application: Detect variability in query execution times or identify
outliers in system logs.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 9 / 40
 3.2 Mathematical Measures

Range: max − min.

P
(xi −x̄)2
Variance: s2 = n−1 .
√
Standard Deviation: s = s2 .
Example 1: Server response times: 120, 130, 120, 150, 140.
Range: 150 − 120 = 30.
Mean: 132. 2 2
+···+(140−132)2
Variance: s2 = (120−132) +(130−132) = 142.
√ 4
Std Dev: s = 142 ≈ 11.92.
Example 2: Query times: 50, 60, 55, 70.
Mean: 50+60+55+70
4 = 58.75.
Range: 70 − 50 = 20.
2 2
Variance: s2 = (50−58.75) +···+(70−58.75) ≈ 67.58.
√ 3
Std Dev: s = 67.58 ≈ 8.22.
IS Application: Measure consistency of system performance or
database efficiency.
(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 10 / 40
 Relative Measures of Variation

Coefficient of Variation: CV = x̄s × 100%.

Example 1: For server response times (mean 132, std dev 11.92).
11.92
CV = 132 × 100 ≈ 9.03%.
Example 2: For user session durations (min): 10, 12, 15, 14, 13.
10+12+15+14+13
Mean: 5 = 12.8.
2 2
Variance: s2 = (10−12.8) +···+(13−12.8) = 3.7.
√ 4
Std Dev: s = 3.7 ≈ 1.92.
CV: 1.92
12.8 × 100 ≈ 15%.
IS Application: Compare variability across systems (e.g., cloud vs.
on-premise) or user engagement metrics.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 11 / 40
 Conclusion

Statistics is critical for IS: data-driven decisions, system optimization,

and user analysis.
Practice: Work on frequency tables, calculate measures, and visualize
data.
Next Steps: Explore probability and distributions (Chapters 4–6).

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 12 / 40
 Introduction to Probability Theory

Definition: Probability theory studies uncertainty and randomness in

events, crucial for decision-making in information systems.
Key Concepts:
Experiment: A process with uncertain outcomes (e.g., server
downtime).
Sample Space: All possible outcomes of an experiment.
Event: A subset of the sample space.
Relevance: Used in risk analysis, system reliability, and data
modeling.
Example: Predicting the probability of a system crash during peak traffic.
Sample Space: {Crash, No Crash}.
Event: System crashes.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 13 / 40
 Counting Rules

Definition: Techniques to count the number of possible outcomes in

a sample space.
Key Rules:
Fundamental Counting Principle: If event A has m outcomes and event
B has n, total outcomes = m × n.
n!
Permutations: Ordered arrangements (P (n, r) = (n−r)! ).
n!
Combinations: Unordered selections (C(n, r) = r!(n−r)! ).
Relevance: Used in password generation, network configurations.
Example: How many ways can 3 servers be assigned to 3 tasks?
Permutation: P (3, 3) = 3! = 6 ways.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 14 / 40
 Probability

Definition: Probability measures the likelihood of an event,

P (A) = NumberTotal
of favorable outcomes
outcomes .
Key Properties:
0 ≤ P (A) ≤ 1.
P (Sample Space) = 1.
For mutually exclusive events: P (A ∪ B) = P (A) + P (B).
Relevance: Used in system failure analysis, cybersecurity risk.
Example: A server fails 2 out of 10 times. Probability of failure?
2
P (Failure) = 10 = 0.2 or 20%.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 15 / 40
 Total Probability and Bayes’ Theorem

Total Probability:
For mutually exclusive events B1 , B2 , . . . , Bn partitioning the sample
space:
P (A) = P (A|B1 )P (B1 ) + P (A|B2 )P (B2 ) + · · · + P (A|Bn )P (Bn ).
Bayes’ Theorem:
P (A|B)P (B)
Updates probabilities based on new evidence: P (B|A) = P (A) .
Relevance: Used in spam filtering, anomaly detection.
Example: A system has two firewalls (F1, F2). F1 detects 80% of attacks
(P (A|F 1) = 0.8), F2 detects 90% (P (A|F 2) = 0.9). F1 is used 60% of
the time (P (F 1) = 0.6), F2 40% (P (F 2) = 0.4). Probability an attack is
detected?
P (A) = (0.8 × 0.6) + (0.9 × 0.4) = 0.48 + 0.36 = 0.84 or 84%.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 16 / 40
 Bayes’ Theorem: Applications in Information Systems

Definition: Bayes’ Theorem updates probabilities based on new

evidence:
P (A|B)P (B)
P (B|A) =
P (A)
where
P (A) = P (A|B1 )P (B1 ) + P (A|B2 )P (B2 ) + · · · + P (A|Bn )P (Bn ).
Key Concepts:
Prior Probability (P (B)): Initial belief before new evidence.
Likelihood (P (A|B)): Probability of evidence given the hypothesis.
Posterior Probability (P (B|A)): Updated belief after evidence.
Normalizing Constant (P (A)): Total probability of evidence.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 17 / 40
 Applications in Information Systems

Spam Filtering:
Classifies emails as spam or not based on keywords and prior data.
Updates probability of spam given new email content.
Network Intrusion Detection:
Identifies malicious activity by updating probabilities based on network
traffic patterns.
Fault Diagnosis in Systems:
Determines the likelihood of hardware/software failure given error logs.
Recommendation Systems:
Predicts user preferences by updating probabilities based on user
behavior.
Risk Assessment:
Evaluates system vulnerabilities by incorporating new threat data.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 18 / 40
 Example 1: Spam Filtering
Scenario: An email system detects a keyword "win" in an email. 60%
of spam emails contain "win" (P (Win|Spam) = 0.6), but only 10% of
non-spam emails do (P (Win|Non-Spam) = 0.1). 30% of emails are
spam (P (Spam) = 0.3), and 70% are non-spam
(P (Non-Spam) = 0.7).
Question: If an email contains "win," what’s the probability it’s
spam?
Solution:
Calculate total probability of "win":
P (Win) = P (Win|Spam)P (Spam) + P (Win|Non-Spam)P (Non-Spam)
P (Win) = (0.6 × 0.3) + (0.1 × 0.7) = 0.18 + 0.07 = 0.25
Apply Bayes’ Theorem:
P (Win|Spam)P (Spam) 0.6 × 0.3 0.18
P (Spam|Win) = = = = 0.72
P (Win) 0.25 0.25
Result: 72% chance the email is spam.
(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 19 / 40
 Example 2: Network Intrusion Detection

Scenario: A system flags unusual traffic. 5% of traffic is malicious

(P (Malicious) = 0.05), and 95% is normal (P (Normal) = 0.95). The
system flags 90% of malicious traffic (P (Flag|Malicious) = 0.9) but
falsely flags 2% of normal traffic (P (Flag|Normal) = 0.02).
Question: If traffic is flagged, what’s the probability it’s malicious?
Solution:
Total probability of a flag:

P (Flag) = (0.9 × 0.05) + (0.02 × 0.95) = 0.045 + 0.019 = 0.064

Bayes’ Theorem:
0.9 × 0.05 0.045
P (Malicious|Flag) = = ≈ 0.703
0.064 0.064
Result: 70.3% chance the flagged traffic is malicious.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 20 / 40
 Example 3: Fault Diagnosis in Systems

Scenario: A server crashes. Crashes are caused by hardware failure

20% of the time (P (Hardware) = 0.2) or software failure 80% of the
time (P (Software) = 0.8). Hardware failures cause crashes 95% of
the time (P (Crash|Hardware) = 0.95), software failures cause crashes
60% of the time (P (Crash|Software) = 0.6).
Question: If a crash occurs, what’s the probability it’s due to
hardware?
Solution:
Total probability of a crash:

P (Crash) = (0.95 × 0.2) + (0.6 × 0.8) = 0.19 + 0.48 = 0.67

Bayes’ Theorem:
0.95 × 0.2 0.19
P (Hardware|Crash) = = ≈ 0.284
0.67 0.67
Result: 28.4% chance the crash is due to hardware.
(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 21 / 40
 Why Bayes’ Theorem Matters

Practical Impact: Enables dynamic decision-making by updating

probabilities with new data.
Information Systems Relevance:
Improves accuracy in spam detection, intrusion detection, and fault
diagnosis.
Supports predictive analytics in system performance and user behavior.
Study Tips for Mid-Exam:
Practice calculating posterior probabilities with real-world scenarios.
Understand the role of prior and likelihood in updating beliefs.
Use Bayes’ Theorem to solve problems in cybersecurity and diagnostics.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 22 / 40
 Random Variables and Probability Distribution

Random Variable: A function mapping outcomes to numbers.

Discrete: Finite or countable outcomes (e.g., number of system errors).
Continuous: Infinite outcomes (e.g., response time).
Probability Distribution: Describes probabilities of a random
variable’s values.
Discrete: Probability Mass Function (PMF), e.g., P (X = x).
Continuous: Probability Density Function (PDF).
Relevance: Models system performance, network traffic.
Example: Number of errors in a system per hour follows a discrete
distribution:
P (X = 0) = 0.5, P (X = 1) = 0.3, P (X = 2) = 0.2.
Probability of at least one error: P (X ≥ 1) = 0.3 + 0.2 = 0.5.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 23 / 40
 Summary

Probability theory is essential for modeling uncertainty in information

systems.
Key concepts: Counting rules, probability, Bayes’ theorem, and
random variables.
Applications: System reliability, cybersecurity, data analysis.
Next Steps: Practice problems on permutations, conditional probability,
and distributions for the mid-exam!

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 24 / 40
 Mean and Variance of Discrete Probability Distributions
Definition: A discrete probability distribution describes probabilities
of a discrete random variable’s values.
Mean (ExpectedPValue):
µ = E(X) = xi P (X = xi ), where xi are possible values and
P (X = xi ) is the probability.
Represents the average outcome over many trials.
Variance:
σ 2 = V ar(X) = (xi − µ)2 P (X = xi ), or σ 2 = E(X 2 ) − [E(X)]2 .
P
Measures the spread of the distribution.
Relevance: Used to analyze system performance metrics like error
rates or response times.
Example: Number of network errors per hour: P (X = 0) = 0.5,
P (X = 1) = 0.3, P (X = 2) = 0.2.
Mean: µ = (0 × 0.5) + (1 × 0.3) + (2 × 0.2) = 0 + 0.3 + 0.4 = 0.7
errors.
Variance:
E(X 2 ) = (02 × 0.5) + (12 × 0.3) + (22 × 0.2) = 0 + 0.3 + 0.8 = 1.1.
2
(By, Selamysfen D. ) (University of Gondar)Introduction 2 for Information Systems
to Statistics October 8, 2025 25 / 40
 The Binomial Distribution
Definition: Models the number of successes in n independent trials,
each with success probability p.
Probability Mass Function (PMF):
! !
n k n n!
P (X = k) = p (1 − p)n−k , =
k k k!(n − k)!
Mean and Variance:
Mean: µ = np
Variance: σ 2 = np(1 − p)
Relevance: Models events like successful packet transmissions or user
login attempts.
Example: A server processes 10 requests, each with a 90% success rate
(p = 0.9). Probability of exactly 8 successes?
n = 10, k = 8, p
= 0.9, 1 − p = 0.1.
P (X = 8) = 10 8 2 10! 8 2
8 (0.9) (0.1) = 8!2! × (0.9) × (0.1) .
10
8 2
8 = 45, (0.9) ≈ 0.4305, (0.1) = 0.01.
P (X = 8) ≈ 45 × 0.4305 × 0.01 ≈ 0.1937 or 19.37%.
(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 26 / 40
 Binomial Distribution: Overview

Definition: The binomial distribution models the number of

successes in n independent trials, each with success probability p.
Probability Mass Function (PMF):
! !
n k n n!
P (X = k) = p (1 − p)n−k , =
k k k!(n − k)!

Key Properties:
Mean: µ = np
Variance: σ 2 = np(1 − p)
Assumes trials are independent and success probability is constant.
Relevance: Ideal for modeling binary outcomes (success/failure) in
information systems.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 27 / 40
 Applications in Information Systems

Network Packet Transmission:

Models successful packet deliveries over a network.
Cybersecurity Authentication:
Analyzes successful user login attempts or failed authentication
attempts.
Software Testing:
Evaluates the number of successful test cases in a fixed number of
trials.
System Reliability Analysis:
Assesses the number of functioning components in a system.
User Behavior Analysis:
Tracks events like clicks or conversions in a fixed number of user
interactions.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 28 / 40
 Example 1: Network Packet Transmission

Scenario: A network sends 20 packets, each with an 85% chance of

successful transmission (p = 0.85). What is the probability of exactly
15 successful transmissions?
Solution:
Use binomial PMF: P (X = 15) = 20

15 5
15 (0.85) (0.15) .
20 20!


Calculate: 15 = 15!5! = 15504.
(0.85)15 ≈ 0.0874, (0.15)5 ≈ 0.000759.
P (X = 15) ≈ 15504 × 0.0874 × 0.000759 ≈ 0.1028.
Result: 10.28% chance of exactly 15 successful transmissions.
Relevance: Helps optimize network protocols for reliable data
transfer.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 29 / 40
 Example 2: Cybersecurity Authentication

Scenario: A system processes 10 login attempts, with a 95% chance

of successful authentication per attempt (p = 0.95). What is the
probability of at least 9 successful logins?
Solution:
Calculate: P (X
≥ 9) = P (X = 9) + P (X = 10).
P (X = 9) = 10 9 1 9
9 (0.95) (0.05) = 10 × (0.95) × 0.05 ≈
10 × 0.6302 × 0.05
≈ 0.3151.
P (X = 10) = 10 10 0
10 (0.95) (0.05) = (0.95)
10
≈ 0.5987.
P (X ≥ 9) ≈ 0.3151 + 0.5987 ≈ 0.9138.
Result: 91.38% chance of at least 9 successful logins.
Relevance: Informs security policies for authentication systems.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 30 / 40
 Example 3: Software Testing

Scenario: A software module runs 12 test cases, each with an 80%

chance of passing (p = 0.8). What is the probability of exactly 10
tests passing?
Solution:
Use binomial PMF: P (X = 10) = 12

10 2
10 (0.8) (0.2) .
12 12!


Calculate: 10 = 10!2! = 66.
(0.8)10 ≈ 0.1074, (0.2)2 = 0.04.
P (X = 10) ≈ 66 × 0.1074 × 0.04 ≈ 0.2835.
Result: 28.35% chance of exactly 10 tests passing.
Relevance: Guides quality assurance in software development.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 31 / 40
 Why Binomial Distribution Matters

Practical Impact: Models binary outcomes critical to system

performance and reliability.
Information Systems Relevance:
Optimizes network reliability by analyzing packet success rates.
Enhances cybersecurity by evaluating authentication success
probabilities.
Supports software testing and quality assurance processes.
Study Tips for Mid-Exam:
Practice binomial PMF calculations for different n and p.
Apply to IS scenarios like network reliability or software testing.
Compare binomial and Poisson distributions for large n, small p.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 32 / 40
 The Poisson Distribution

Definition: Models the number of events occurring in a fixed interval

of time or space, with average rate λ.
Probability Mass Function (PMF):

λk e−λ
P (X = k) = , k = 0, 1, 2, . . .
k!
Mean and Variance:
Mean: µ = λ
Variance: σ 2 = λ
Relevance: Models rare events like server crashes or network packet
drops.
Example: A network experiences an average of 2 packet drops per hour
(λ = 2). Probability of exactly 3 drops in an hour?
e 3 −2 −2
P (X = 3) = 2 3! = 8×e6 .
e−2 ≈ 0.1353, so P (X = 3) ≈ 8×0.1353
6 ≈ 0.1804 or 18.04%.
(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 33 / 40
 Poisson Distribution: Overview

Definition: The Poisson distribution models the number of events

occurring in a fixed interval of time or space, with average rate λ.
Probability Mass Function (PMF):

λk e−λ
P (X = k) = , k = 0, 1, 2, . . .
k!
Key Properties:
Mean: µ = λ
Variance: σ 2 = λ
Assumes events are independent and occur at a constant average rate.
Relevance: Ideal for modeling rare or random events in information
systems.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 34 / 40
 Applications in Information Systems

Network Traffic Analysis:

Models the number of incoming packets or requests to a server in a
time period.
System Failure Prediction:
Estimates the frequency of server crashes or hardware failures.
Cybersecurity Incident Detection:
Analyzes the occurrence of security alerts or intrusion attempts.
User Activity Monitoring:
Tracks events like login attempts or database queries per unit time.
Queueing Systems:
Models customer support ticket arrivals or job processing in a system.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 35 / 40
 Example 1: Network Traffic Analysis

Scenario: A server receives an average of 5 connection requests per

minute (λ = 5). What is the probability of exactly 3 requests in a
minute?
Solution:
3 −λ
Use Poisson PMF: P (X = 3) = λ 3!e
.
3 −5 −5
5 e
Substitute λ = 5: P (X = 3) = 3! = 125×e 6 .
e−5 ≈ 0.006738, so P (X = 3) ≈ 125×0.006738
6 ≈ 0.1404.
Result: 14.04% chance of exactly 3 requests.
Relevance: Helps predict server load for resource allocation.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 36 / 40
 Example 2: System Failure Prediction

Scenario: A data center experiences an average of 2 server failures

per day (λ = 2). What is the probability of no failures in a day?
Solution:
0 −λ
Use Poisson PMF: P (X = 0) = λ 0!e
= e−λ .
−2
Substitute λ = 2: P (X = 0) = e ≈ 0.1353.
Result: 13.53% chance of no failures in a day.
Relevance: Aids in planning maintenance schedules and uptime
guarantees.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 37 / 40
 Example 3: Cybersecurity Incident Detection

Scenario: A firewall logs an average of 1 intrusion attempt per hour

(λ = 1). What is the probability of at least 2 attempts in an hour?
Solution:
Calculate: P (X ≥ 2) = 1 − [P (X = 0) + P (X = 1)].
P (X = 0) = e−1 ≈ 0.3679.
1 −1
P (X = 1) = 1 1!e
= e−1 ≈ 0.3679.
P (X ≥ 2) = 1 − (0.3679 + 0.3679) = 1 − 0.7358 ≈ 0.2642.
Result: 26.42% chance of at least 2 intrusion attempts.
Relevance: Guides the tuning of intrusion detection systems.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 38 / 40
 Why Poisson Distribution Matters

Practical Impact: Models rare, random events critical to system

reliability and security.
Information Systems Relevance:
Optimizes resource allocation in networks (e.g., bandwidth for traffic
spikes).
Enhances system reliability by predicting failure rates.
Strengthens cybersecurity by modeling attack frequencies.
Study Tips for Mid-Exam:
Practice Poisson PMF calculations for different λ values.
Apply to IS scenarios like network traffic or error rates.
Understand the difference between Poisson and binomial distributions.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 39 / 40
 Summary

Mean and Variance: Quantify expected outcomes and variability in

system metrics.
Binomial Distribution: Models success/failure scenarios in fixed
trials (e.g., packet transmission).
Poisson Distribution: Models rare events over time/space (e.g.,
network errors).
Applications: Network reliability, system performance, cybersecurity
analysis.
Study Tips: Practice calculating mean, variance, and probabilities for
binomial and Poisson distributions using IS-related scenarios.

(By, Selamysfen D. ) (University of Gondar)Introduction to Statistics for Information Systems October 8, 2025 40 / 40