Fundamentals of Data Science Analytics

Module 4: Data Analysis

Dr Archimedes C Articulo
Central Graduate School – Isabela State University
Mobile No.: 09178911403
Email Add.: [email protected]

In our journey through data analytics, we've covered the foundational concepts,
understood the data process flow, cleaned our data, and learned how to visualize it.
Now, it's time for the core of data analytics: analysis.

This lesson will introduce you to different types of data analysis – Descriptive,
Predictive, and Prescriptive Analytics. You'll learn how each type helps in understanding
past events, forecasting future trends, and even recommending actions, all with a focus
on their application in local governance. We'll also delve into specific models and
techniques, including a look at advanced applications like sentiment analysis.

This lesson is designed for you to learn at your own pace. Engage with the activities to
deepen your understanding and apply what you learn!

Lesson 1: Introduction to Descriptive Analytics

Learning Objectives: By the end of this Lesson, you should be able to:

1. Define descriptive analytics and explain its purpose.

2. Calculate and interpret common measures in descriptive statistics (measures of
central tendency and dispersion).
3. Understand how descriptive analysis provides crucial insights for policy-making
and decision-making in local governance.

1
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
1.1 What is Descriptive Analytics?

Imagine you're managing the City Health Office in Tuguegarao. You have records of all
dengue cases from last year. Your first question might be: "How many cases did we
have?" or "Which barangay had the most cases?"

Descriptive analytics is the process of summarizing and describing the main features
of a dataset. It's about looking at historical data to understand "what happened." It's the
most fundamental type of data analysis and forms the basis for more advanced
analyses.

Think of it as summarizing a book. You're not predicting the ending or recommending a

new plot; you're just describing what already happened in the story.

Key characteristics of Descriptive Analytics:

1. Historical Focus: Works with past or current data.

2. Summarization: Uses metrics, charts, and graphs to condense data.
3. Pattern Identification: Helps spot trends, distributions, and outliers.
4. Foundation: Essential prerequisite for more advanced analytics.

1.2 Measures in Descriptive Statistics

To describe a dataset, we use various statistical measures. These can be broadly

categorized into:

A. Measures of Central Tendency (What's the "typical" value?)

These measures describe the center point or average of a dataset.

1. Mean (Average): The sum of all values divided by the number of values.

a. Calculation (Excel): =AVERAGE(range)

b. LGU Example: Average age of registered voters in Barangay Linao.
c. Insight: Gives a general idea of the "typical" value.

2. Median: The middle value in a dataset when the values are arranged in order
(ascending or descending). If there's an even number of values, it's the average
of the two middle values.

a. Calculation (Excel): =MEDIAN(range)

2
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 b. LGU Example: Median household income in Tuguegarao City.
c. Insight: Less affected by extreme outliers than the mean, useful for
skewed data (e.g., income, property values).

3. Mode: The value that appears most frequently in a dataset.

a. Calculation (Excel): =MODE.SNGL(range) (for a single mode) or

MODE.MULT(range) (for multiple modes).
b. LGU Example: The most common type of business registered in the past
quarter (e.g., Sari-sari store).
c. Insight: Useful for categorical data or discrete numerical data to show the
most popular or frequent item.

B. Measures of Dispersion (How spread out is the data?)

These measures describe the variability or spread of the data points around the central
tendency.

1. Range: The difference between the highest and lowest values in a dataset.

a. Calculation (Excel): =MAX(range) - MIN(range)

b. LGU Example: The range of daily temperatures in Tuguegarao during the
dry season.
c. Insight: Provides a quick sense of the data's span, but highly sensitive to
outliers.

2. Variance: Measures how far each number in the dataset is from the mean. It's
the average of the squared differences from the mean.

a. Calculation (Excel): =VAR.S(range) (for a sample) or =VAR.P(range) (for a

population)
b. LGU Example: Variance in the number of attendees at community events
across different barangays.
c. Insight: A larger variance indicates data points are more spread out.

3. Standard Deviation: The square root of the variance. It's widely used because
it's in the same units as the original data, making it easier to interpret.

a. Calculation (Excel): =STDEV.S(range) (for a sample) or =STDEV.P(range) (for a

population)
b. LGU Example: Standard deviation of response times for emergency
services.
c. Insight: A low standard deviation means data points are generally close to
the mean; a high standard deviation means they are more spread out.

3
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
C. Measures of Frequency/Counts

For categorical data, simply counting occurrences is fundamental.

1. Frequency (Count): The number of times a particular value appears in a

dataset.

a. Calculation (Excel): =COUNTIF(range, criteria)

b. LGU Example: Number of households who received aid, count of specific
crime types.

2. Percentage: The proportion of a category relative to the total.

a. Calculation (Excel): =COUNTIF(range, criteria) / COUNTA(range) (formatted as

percentage)
b. LGU Example: Percentage of the city budget allocated to education.

1.3 How Descriptive Analysis Can Be Used in Policy-Making and Decision-Making

Descriptive analytics is the starting point for almost any data-driven decision. It provides
the foundational understanding of the current state or past performance, which is crucial
for:

1. Situational Awareness: Understanding the current landscape.

a. Policy Example: Knowing the average daily volume of solid waste

collected by barangay informs the budget for waste management
operations.
b. Decision Example: The City Treasurer can see the total amount of
business taxes collected last quarter, assessing if targets were met.

2. Identifying Needs and Gaps: Pinpointing areas requiring attention.

a. Policy Example: If the mean number of police patrols in Barangay A is

significantly lower than Barangay B, despite similar crime rates, a policy
might be adjusted to reallocate resources.
b. Decision Example: The LGU identifies that 70% of reported potholes are
on two specific main roads (descriptive frequency), prompting immediate
repair decisions for those roads.

3. Performance Evaluation: Assessing the effectiveness of existing programs or

policies.

4
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 a. Policy Example: Analyzing the average waiting time for processing
business permits before and after implementing a new online system
helps evaluate the system's success.
b. Decision Example: The LGU can see the median response time for citizen
complaints and decide if it meets internal service level agreements.

4. Baseline Setting: Providing a benchmark for future comparison.

a. Policy Example: Establishing the baseline average flood depth in certain

areas during the monsoon season informs new infrastructure projects'
design goals.
b. Decision Example: The current average water consumption in a city park
can serve as a baseline to measure the impact of new water-saving
initiatives.

5. Resource Allocation: Informing where to best distribute resources.

a. Policy Example: Understanding the population distribution across different

age groups (e.g., high percentage of senior citizens in certain barangays)
informs the policy on senior citizen welfare programs.
b. Decision Example: Based on the number of households below the poverty
line per barangay, the SWDO can decide which barangays need more
targeted food assistance.

6. Communicating Results: Using simple descriptive statistics and visualizations

to present clear findings to stakeholders and the public.
a. Decision Example: A public report showing the total number of participants
in various LGU training programs and the percentage of participants who
completed them.

Activity 1.1: Descriptive Stats for Tuguegarao's Services

Instructions: Imagine you have collected the following hypothetical data from the
Tuguegarao City Public Works and Highways Department regarding reported potholes
and their repair times for a specific month:

1 Report Barangay Severity (1=Minor, 5=Severe) Days to Repair

ID
2 P001 San Roque 3 5
3 P002 Cataggaman 4 8
Pardo
4 P003 Centro 3 2 3
5 P004 San Roque 5 12

5
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 6 P005 Balzain 1 2
7 P006 Cataggaman 3 6
Pardo
8 P007 Centro 3 4 7
9 P008 San Roque 2 4
10 P009 Balzain 5 10
11 P010 Cataggaman 1 3
Pardo

1. Calculate (or write down the Excel formula for):

a. The Mean 'Days to Repair'.

o Excel Formula: =AVERAGE(D2:D11)

b. The Median 'Days to Repair'.

o Excel Formula: =MEDIAN(D2:D11)

c. The Mode 'Barangay' (which barangay appears most frequently).

o Excel Formula: =MODE.SNGL(B2:B11)

d. The Range of 'Days to Repair'.

o Excel Formula: =MAX(D2:D11) - MIN(D2:D11)

e. The Number of reported potholes in "San Roque".

o Excel Formula: =COUNTIF(B2:B11, "San Roque")

f. The Percentage of potholes with 'Severity' of 5.

o Excel Formula: =COUNTIF(C2:C11, 5) / COUNTA(C2:C11) (Remember to format

this cell as a percentage in Excel!)

2. Policy/Decision Application: Based on these hypothetical descriptive findings,

provide two (2) specific ways this analysis could influence policy-making or
decision-making in the Tuguegarao City Public Works and Highways
Department.

This activity will help you apply the concepts of descriptive statistics to a practical LGU
scenario.

6
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
Instruction:
Submit your output in MS Excel file and email them to [email protected]
Save your file as “Your surname_Task1_lesson1_Module4.xlsx”
Attach and send email as instructed above

Lesson 2: Introduction to Predictive Analytics

Learning Objectives: By the end of this Lesson, you should be able to:

1. Define predictive analytics and differentiate it from descriptive analytics.

2. Understand the basic concept of models in predictive analysis.
3. Explain the application of predictive analytics, with emphasis on the Hitachi's
Predictive Analytics Model in crime analysis (and its relevance to LGUs).
4. Describe how predictive analysis can be used in policy-making and decision-
making for local governance.

2.1 What is Predictive Analytics?

While descriptive analytics tells us "what happened," predictive analytics aims to

answer "what will happen?" It uses historical data, statistical models, and machine
learning techniques to forecast future outcomes or probabilities.

Think of it like a weather forecast. Meteorologists use past weather patterns, current
atmospheric conditions, and complex models to predict if it will rain tomorrow. They are
not describing yesterday's rain; they are trying to predict tomorrow's.

Key characteristics of Predictive Analytics:

1. Future Focus: Attempts to forecast future events or trends.

2. Model-Based: Relies on statistical models or machine learning algorithms
trained on historical data.
3. Probabilistic: Predictions often come with a probability or confidence level.
4. Uncertainty: Involves inherent uncertainty, as no prediction is 100% accurate.

2.2 Models in Predictive Analysis

At the heart of predictive analytics are models. A model is essentially a mathematical

formula or algorithm that identifies patterns and relationships in historical data and then
uses those patterns to make predictions on new, unseen data.

7
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
While there are many complex models, some common approaches include:

1. Regression Analysis: Used to predict a continuous numerical value (e.g., future

temperature, future revenue) based on the relationship with one or more
independent variables.

Example: Predicting future electricity consumption based on historical

consumption, average temperature, and population growth.

2. Classification Models: Used to predict a categorical outcome (e.g., whether a

resident will default on a loan, if an area is high-risk for crime).

Example: Predicting whether a business permit application is likely to be

approved or denied based on past applicant characteristics.

3. Time Series Analysis: Specifically designed for data collected over time, to
forecast future values based on past trends and seasonality.

Example: Forecasting the number of dengue cases in Tuguegarao for the

next quarter based on historical monthly case data and seasonal patterns.

Emphasis on Hitachi's Predictive Analytics Model in Application of Crime

Analysis

While specific details of proprietary models like Hitachi's are not publicly exhaustive, we
can understand the principles behind their application in crime analysis. These models
typically employ advanced machine learning algorithms (like clustering, regression, or
even deep learning) to identify patterns in vast amounts of crime data.

How such models generally work in crime analysis:

1. Data Collection: Gathers historical crime data (type of crime, location, date,
time, weapons used, victim/perpetrator demographics, weather conditions, social
events, etc.).

2. Feature Engineering: Transforms raw data into features useful for the model
(e.g., creating "time of day" categories, "day of week" flags, or "proximity to
school" variables).

3. Pattern Recognition: The model learns from this historical data to identify
complex, often non-obvious, patterns.

8
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 Examples: "Burglaries increase on Fridays between 2 PM and 5 PM in
areas with high unemployment rates and low police visibility." or "Certain
types of assaults are more likely near bars during specific weekend
hours."

4. Prediction: Based on current conditions and learned patterns, the model

predicts:

a. Hotspots: Specific areas or blocks where crime is more likely to occur.

b. Time windows: When crime is more likely to happen.
c. Types of crime: What kind of crime is expected.

5. Output: Presents these predictions (e.g., as a heat map of high-risk areas, or a

list of specific patrol routes for a given day).

Relevance to LGUs:

While a full-fledged Hitachi-level system might be beyond typical LGU budgets, the
concept is highly relevant:

1. Resource Optimization: Instead of random patrols, predict where police

presence is most needed, leading to more efficient use of limited personnel.
2. Proactive Prevention: Intervene in high-risk areas before crimes occur, rather
than just reacting.
3. Policy Formulation: Data about crime hotspots can inform policy decisions on
urban planning (e.g., improving lighting in certain areas), social programs (e.g.,
youth engagement in high-risk neighborhoods), or community policing strategies.
4. Citizen Safety: Ultimately, aims to enhance the safety and security of citizens.

2.3 How Predictive Analysis Can Be Used in Policy-Making and Decision-Making

Predictive analytics empowers LGUs to shift from reactive to proactive governance.

1. Resource Forecasting & Planning:

a. Policy Example: Predicting the future population growth in different

barangays informs long-term infrastructure development plans (schools,
hospitals, roads).
b. Decision Example: Forecasting the number of patients expected at city
health centers next month helps in ensuring adequate medical supplies
and staffing.

9
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 c. LGU Specific: Predicting future flood severity based on rainfall forecasts
and historical flood data helps the City Disaster Risk Reduction and
Management Office (CDRRMO) pre-position relief goods and plan
evacuations.

2. Targeted Interventions:

a. Policy Example: Identifying areas at high risk for dengue outbreaks based
on weather, past cases, and sanitation data allows for targeted clean-up
drives and public health campaigns.
b. Decision Example: Predicting which businesses are at high risk of tax
delinquency allows the City Treasurer's Office to send early reminders or
offer assistance, rather than chasing after defaulters.

3. Risk Management:

a. Policy Example: Predicting areas prone to informal settlements helps in

developing proactive housing and livelihood programs to prevent future
issues.
b. Decision Example: Predicting which bridges or roads are likely to need
major repairs in the next 1-2 years based on age, traffic volume, and
inspection data, allows for proactive budget allocation.

4. Optimizing Operations:

a. Policy Example: Forecasting demand for public transport services at

different times of day can lead to optimized routing and scheduling
policies.
b. Decision Example: Predicting peak hours for citizen inquiries allows the
LGU's public assistance desk to adjust staffing levels, reducing wait times.

5. Program Effectiveness:

a. Policy Example: Predicting the success rate of vocational training

programs based on participant demographics and past outcomes can help
tailor program design.
b. Decision Example: Forecasting potential tourist arrivals based on events,
holidays, and past trends helps the tourism office plan marketing
campaigns and resource allocation.

Important Note: Predictive analytics is about providing probabilities and likelihoods, not
certainties. Human judgment and ethical considerations are always essential when
applying predictive models, especially in sensitive areas like crime.

10
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
Activity 2.1: Predicting for a Better Tuguegarao

Instructions: Think about a specific challenge in Tuguegarao City where predictive

analytics could be highly beneficial.

1. Identify a Challenge: Describe one specific challenge in your town or City that
you believe could be effectively addressed using predictive analytics.

Examples: Traffic congestion, dengue outbreaks, informal settlement

growth, public market sanitation issues, disaster preparedness (other than
general flooding).

2. Hypothesize Data: What kind of historical data would you need to build a
predictive model for this challenge? Be specific about 2-3 types of data.
3. Prediction Goal: What specific outcome or trend would your predictive model
aim to forecast?
4. Policy/Decision Impact: How would the insights from this predictive analysis
directly influence a policy-making decision or an operational decision in the LGU?
Provide one concrete example.

Write down your answers. This will help you connect predictive theory to local practice.

Instruction:

Submit your output in MS Word file and email them to [email protected]

Save your file as “Your surname_Task2_lesson2_Module4.doc”

Attach and send email as instructed above

11
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
Lesson 3: Prescriptive Analytics & Sentiment Analysis

Learning Objectives: By the end of this Lesson, you should be able to:

1. Define prescriptive analytics and differentiate it from descriptive and predictive

analytics.
2. Understand the concept of sentiment analysis and its application to local
governance.
3. Explain the role of social weather and opinion surveys and social media listening
in LGU decision-making.

3.1 Prescriptive Analytics: What Should We Do?

If descriptive analytics tells us "what happened," and predictive analytics tells us "what
will happen," then prescriptive analytics tells us "what should we do?" It's the most
advanced form of analytics, recommending specific actions to achieve desired
outcomes or mitigate potential risks.

Prescriptive analytics often leverages predictive models and then applies optimization or
simulation techniques to determine the best course of action among various options.

Key characteristics of Prescriptive Analytics:

1. Action-Oriented: Provides specific recommendations for action.

2. Optimization: Aims to find the best possible solution given constraints.
3. Complex: Often combines multiple analytical techniques.
4. Decision Support: Directly guides decision-making.

LGU Applications of Prescriptive Analytics:

1. Resource Optimization:

a. Example: Determining the optimal number of garbage trucks and their

routes to minimize fuel consumption and collection time, given daily waste
volume forecasts (from predictive analytics) and traffic conditions.
b. Tuguegarao Specific: Optimizing the deployment of City Hall personnel to
different service counters based on forecasted citizen foot traffic and
inquiry types to minimize waiting times.

2. Scheduling and Planning:

12
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 a. Example: Recommending the best times for road repairs to minimize
traffic disruption, considering traffic patterns (descriptive) and forecasted
traffic volume (predictive).

3. Crisis Response:
a. Example: During a typhoon, recommending optimal evacuation routes and
shelter locations based on real-time flood data, road closures, and shelter
capacity.

4. Policy Optimization:
a. Example: Simulating the impact of different tax incentive policies on local
business growth to recommend the most effective one.

3.2 Sentiment Analysis and its Application to Local Governance

Beyond numerical data, LGUs deal with a vast amount of textual data – citizen
feedback, social media comments, survey responses, news articles. Sentiment
analysis (also known as opinion mining) is a natural language processing (NLP)
technique used to determine the emotional tone (positive, negative, neutral) behind a
piece of text.

How Sentiment Analysis Works (Simplified): It uses algorithms to process text and
classify the sentiment expressed. This can be based on dictionaries of positive/negative
words, or more advanced machine learning models trained on large datasets.

A. Social Weather and Opinion Surveys:

Traditional surveys are a cornerstone of gathering public opinion. Incorporating

sentiment analysis methods into their processing can add depth:

1. How it works:
a. Surveys with open-ended questions (e.g., "What are your suggestions for
improving public services in Tuguegarao?") yield rich text data.
b. Sentiment analysis can then be applied to these responses to identify
overarching positive, negative, or neutral feelings towards specific
services, policies, or LGU initiatives.

13
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 2. Application to Local Governance:
a. Gauging Public Satisfaction: Quickly understand overall public
sentiment towards specific LGU departments (e.g., cleanliness, security,
efficiency).
b. Identifying Pain Points: Pinpoint recurring negative themes or specific
areas of discontent expressed by citizens.
c. Evaluating Policy Impact: Assess public reception to newly implemented
policies or programs.
d. Targeted Communication: Understand public concerns to tailor
communication strategies.

B. Social Media Listening Software/Analytics:

Social media platforms (Facebook, X, etc.) are public forums where citizens discuss
local issues, praise, and complain. Social media listening involves monitoring these
platforms for mentions of the LGU, specific services, or local events, and then analyzing
the sentiment.

1. How it works:
a. Specialized software monitors keywords (e.g., "Tuguegarao LGU,"
"Tuguegarao City Hall," "Traffic in Tuguegarao," "Mayor [Name]") across
social media.
b. It collects relevant posts, comments, and mentions.
c. Sentiment analysis algorithms are then applied to these texts to classify
them as positive, negative, or neutral.
d. Dashboards present trends in sentiment over time, common topics, and
key influencers.

2. Application to Local Governance:

a. Real-time Feedback: Get immediate insights into public reaction to
current events, LGU announcements, or emergencies (e.g., during a
typhoon, quickly gauge public sentiment about evacuation efforts).
b. Issue Identification: Discover emerging problems or complaints before
they escalate (e.g., sudden increase in negative sentiment about power
outages in a specific area).
c. Reputation Management: Monitor public perception of the LGU and its
officials, allowing for proactive response to negative narratives.
d. Citizen Engagement: Identify influential voices or common questions to
guide LGU communication and engagement efforts.
e. Disaster Response: During emergencies, quickly identify calls for help,
specific locations needing assistance, or misinformation spreading.

14
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
Challenges in Sentiment Analysis:

1. Nuance: Sarcasm, irony, and cultural context can be difficult for algorithms to
interpret.
2. Dialect/Slang: Filipino slang or local Tuguegarao dialects might confuse generic
models.
3. Data Volume: Social media data is massive and noisy.
4. Privacy Concerns: While social media posts are public, analyzing them still
requires ethical consideration.

Key takeaway for LGUs: Sentiment analysis provides a powerful way to understand
the qualitative aspects of public opinion, complementing traditional surveys and
providing real-time insights into citizen needs and concerns.

Activity 3.1: Applying Advanced Analytics in Tuguegarao

Instructions: For each of the following scenarios in Tuguegarao City, determine which
type of analytics (Descriptive, Predictive, or Prescriptive) would be most appropriate,
and explain why. Then, suggest how sentiment analysis could be used.

Scenario 1: Optimizing Water Bill Collections

1. Goal: To reduce the number of unpaid water bills and increase revenue from
water services.
2. Questions:
a. What type of analytics would tell you the current average payment rate
and the total amount of outstanding bills?
b. What type of analytics would help identify residents most likely to default
on their bills next month, based on past payment behavior and other
factors?
c. What type of analytics would recommend the best intervention strategy for
each defaulting household (e.g., send a text reminder, dispatch a
collector, offer a payment plan)?
d. How could sentiment analysis be used to understand public satisfaction
with water services and billing processes?

Scenario 2: Improving Disaster Response in Flood-Prone Areas

15
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
 1. Goal: To enhance the LGU's preparedness and response to flooding in flood-
prone barangays.
2. Questions:
a. What type of analytics would show the historical frequency and severity of
floods in different barangays?
b. What type of analytics would forecast which areas are most likely to be
severely affected by an upcoming typhoon, based on rainfall predictions,
historical flood data, and terrain?
c. What type of analytics would recommend the optimal allocation of relief
goods and evacuation routes during a flood event, given real-time data on
road conditions, shelter capacity, and affected populations?
d. How could social media listening (with sentiment analysis) be used by the
CDRRMO during a flood to get real-time feedback and identify urgent
needs?

Write down your answers, focusing on the distinct capabilities of each analytics type.

Instruction:

Submit your output in MS Word file and email them to [email protected]

Save your file as “Your surname_Task3_lesson3_Module4.doc”

Attach and send email as instructed above

16
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo
Module 4 Review & Assessment

You've completed Module 4: Data Analysis. You've delved into the core of data
analytics, distinguishing between descriptive, predictive, and prescriptive approaches.
You've also explored specialized techniques like sentiment analysis and its powerful
applications in local governance. This understanding equips you to ask deeper
questions of your data and derive more actionable insights.

Take some time to review your notes and the activities you've completed. If any
concepts are still unclear, revisit the relevant sections.

Self-Assessment Questions:

1. Differentiate between Descriptive, Predictive, and Prescriptive Analytics using a

single LGU example for each.
2. Name and briefly explain two (2) measures of central tendency and two (2)
measures of dispersion.
3. How can descriptive analysis inform a policy decision regarding public health in
Tuguegarao City?
4. What is the primary goal of predictive analytics in crime analysis (e.g., using a
model like Hitachi's)?
5. Describe one way predictive analytics could help the Tuguegarao City LGU
prepare for a major city event (e.g., Afi Festival).
6. What is sentiment analysis, and how can social media listening contribute to an
LGU's understanding of public opinion?
7. Provide an example of how "social weather and opinion surveys" can leverage
sentiment analysis for better decision-making in local governance.

Moving Forward: You now have a solid understanding of the different types of data
analysis. In the upcoming lessons, we will explore advanced concepts and tools to
further enhance your data analytics capabilities.

17
Fundamentals of Data Science Analytics
Module 4: Data Analysis
Dr Archimedes C Articulo