Epoka University

FACULTY OF ECONOMICS AND ADMINISTRATIVE SCIENCES
DEPARTMENT OF BUSINESS ADMINISTRATION
COURSE SYLLABUS

2025-2026 ACADEMIC YEAR

COURSE INFORMATION

Course Title: PREDICTIVE ANALYTICS

Code	Course Type	Regular Semester	Theory	Practice	Lab	Credits	ECTS
BIDS 404	B	2	3	0	2	4	8

Academic staff member responsible for the design of the course syllabus (name, surname, academic title/scientific degree, email address and signature)	Prof.Dr. Abdurrahman IŞIK aisik@epoka.edu.al
*Main Course Lecturer (name, surname, academic title/scientific degree, email address and signature)* and Office Hours:**	Prof.Dr. Abdurrahman IŞIK aisik@epoka.edu.al , Mondays : 16:00 - 17:00
*Second Course Lecturer(s) (name, surname, academic title/scientific degree, email address and signature)* and Office Hours:**	NA
Language:	English
Compulsory/Elective:	Compulsory
Study program: (the study for which this course is offered)	Master of Science in Business Intelligence and Data Science
Classroom and Meeting Time:	Check timetable
Teaching Assistant(s) and Office Hours:	NA
Code of Ethics:	Code of Ethics of EPOKA University Regulation of EPOKA University "On Student Discipline"
Attendance Requirement:	75%
Course Description:	This course will primarily focus on the practical application of classification models in data mining within the business context. These models are utilized to make predictions that support decision-making processes. Additionally, the course aims to provide a non-technical explanation of the statistical and artificial intelligence-based tools commonly employed in data mining. By expanding upon basic statistical and analytical tools, students will develop a deeper understanding of predictive methods and models specifically tailored for business applications. The ultimate goal is to equip students with the necessary technical knowledge and tools to effectively work with data, databases, and reports derived from business information systems. The course covers essential topics such as data analysis, predictive modeling, linear regression, logistic regression, decision trees, artificial neural networks, and model evaluation concepts.
Course Objectives:	The primary objective of this course is to empower students with the mathematical framework and computational tools necessary to transform historical data into reliable future insights. Students will develop the technical proficiency to build, tune, and deploy predictive models while mastering the critical thinking skills required to evaluate model performance and ethical implications. Ultimately, the course aims to bridge the gap between academic theory and industry practice, ensuring graduates can drive data-informed decision-making in complex, high-stakes environments.

BASIC CONCEPTS OF THE COURSE

1	1. Fundamental Taxonomy Definition: The process of using data analysis, machine learning, and statistical models to identify patterns and forecast future behaviors. The Analytics Hierarchy: Understanding how predictive analytics fits between Descriptive (what happened), Diagnostic (why it happened), and Prescriptive (what should we do) analytics.
2	2. Core Mathematical Pillars Regression Analysis: Estimating relationships between variables to predict continuous outcomes (e.g., sales volume). Classification: Categorizing data into distinct groups or classes (e.g., fraud vs. non-fraud). Time-Series Foundation: Analyzing data points collected or indexed in successive order to identify seasonal trends and cycles. Probability & Risk: Using statistical certainty and confidence intervals to quantify the likelihood of a specific event occurring.
3	3. The Predictive Modeling Workflow Problem Formulation: Translating a business objective into a measurable "target variable". Data Readiness: The critical phase of data cleaning (removing anomalies) and feature engineering (creating new input variables). The Three-Dataset Split: The conceptual requirement of dividing data into Training, Validation, and Testing sets to ensure model reliability.
4	4. Model Integrity Concepts Overfitting vs. Underfitting: Balancing a model's ability to learn from historical data without losing its ability to generalize to new, unseen data. Bias-Variance Tradeoff: Understanding the tension between a model's error due to overly simple assumptions (bias) versus its sensitivity to small fluctuations in the training set (variance). Interpretability vs. Accuracy: Navigating the choice between "black box" models (e.g., Neural Networks) and transparent models (e.g., Decision Trees).
5	. Ethical & Strategic Frameworks Algorithmic Bias: Recognizing how historical data can reinforce existing inequities in automated decisions. Data Governance: Concepts of privacy (GDPR/FERPA), security, and the ethical "right to an explanation" for automated outcomes.

COURSE OUTLINE

Week	Topics
1	Introduction: 1.1 What are data mining and predictive analytics? , 1.2 How good are models at predicting behavior?, 1.3 What are the benefits of predictive models?, 1.4 Applications of predictive analytics, 1.5 Reaping the benefits, avoiding the pitfalls, 1.6 What is Big Data?, 1.7 How much value does Big Data add?
2	Using Predictive Models: 2.1 What are your objectives?, 2.2 Decision making, 2.3 The next challenge, 2.4 Discussion, 2.5 Override rules (business rules)
3	Analytics, Organization and Culture1: 3.1 Embedded analytics, 3.2 Learning from failure, 3.3 A lack of motivation, 3.4 A slight misunderstanding, 3.5 Predictive, but not precise
4	Analytics, Organization and Culture2: 3.6Great expectations, 3.7 Understanding cultural resistance to predictive analytics, 3.8 The impact of predictive analytics, 3.9 Combining model-based predictions and human judgment
5	The Value of Data: 4.1 What type of data is predictive of behavior?, 4.2 Added value is what’s important, 4.3 Where does the data to build predictive models come from?, 4.4 The right data at the right time, 4.5 How much data do I need to build a predictive model?
6	Ethics and Legislation: 5.1 A brief introduction to ethics, 5.2 Ethics in practice, 5.3 The relevance of ethics in a Big Data world, 5.4 Privacy and data ownership, 5.5 Data security, 5.6 Anonymity, 5.7 Decision making
7	Types of Predictive Models1: 6.1 Linear models, 6.2 Decision trees (classification and regression trees), 6.3 (Artificial) neural networks, 6.4 Support vector machines (SVMs), 6.5 Clustering
8	Types of Predictive Models2: 6.6 Expert systems (knowledge-based systems), 6.7 What type of model is best?, 6.8 Ensemble (fusion or combination) systems, 6.9 How much benefit can I expect to get from using an ensemble?, 6.10 The prospects for better types of predictive models in the future
9	The Predictive Analytics Process: 7.1 Project initiation, 7.2 Project requirements, 7.3 Is predictive analytics the right tool for the job?, 7.4 Model building and business evaluation, 7.5 Implementation, 7.6 Monitoring and redevelopment, 7.7 How long should a predictive analytics project take?
10	How to Build a Predictive Model 1: 8.1 Exploring the data landscape, 8.2 Sampling and shaping the development sample, 8.3 Data preparation (data cleaning), 8.4 Creating derived data, 8.5 Understanding the data
11	How to Build a Predictive Model 2: 8.6 Preliminary variable selection (data reduction), 8.7 Pre-processing (data transformation), 8.8 Model construction (modeling), 8.9 Validation, 8.10 Selling models into the business, 8.11 The rise of the regulator
12	Text Mining and Social Network Analysis: 9.1 Text mining, 9.2 Using text analytics to create predictor variables, 9.3 Within document predictors, 9.4 Sentiment analysis, 9.5 Across document predictors, 9.6 Social network analysis, 9.7 Mapping a social network
13	Hardware, Software and All that Jazz: 10.1 Relational databases, 10.2 Hadoop, 10.3 The limitations of Hadoop, 10.4 Do I need a Big Data solution to do predictive analytics?, 10.5 Soft ware for predictive analytics
14	Revision

Prerequisite(s):	Statistics, Econometrics
Textbook(s):	Predictive Analytics in Business :Strategy, methods, technology, Martin Butler , Applied Predictive Analytics: Principles and Techniques for the Professional Data Analyst, Dean Abbott, John Wiley & Sons, 2014, Learning Predictive Analytics with R, Eric Mayor, Packt Publishing, 2015, Using Econometrics: A Practical Guide, Global Edition, A. H. Studenmund, Global Edition (7th ed.). Pearson International Content, 2021.
Additional Literature:
Laboratory Work:
Computer Usage:
Others:	No

COURSE LEARNING OUTCOMES

1	1. Core Technical Mastery Algorithm Implementation: Implement advanced predictive models—including linear and logistic regression, decision trees, random forests, and support vector machines—using industry-standard programming languages like Python or R. Statistical Rigor: Apply deep statistical foundations, such as probability distributions, hypothesis testing, and Bayesian inference, to validate the mathematical integrity of forecasting models. Data Engineering: Execute complex data preprocessing and feature engineering tasks, including cleaning, normalizing, and transforming raw unstructured data into high-quality modeling inputs.
2	2. Model Evaluation & Optimization Critical Benchmarking: Critically compare and contrast different predictive techniques to identify the most suitable model for a specific problem based on accuracy, precision, and recall. Validation Protocols: Design and implement robust validation strategies, such as K-fold cross-validation and hold-out sampling, to ensure model generalizability and prevent overfitting. Performance Tuning: Optimize model performance through hyperparameter tuning and bias-variance tradeoff analysis.
3	3. Strategic Application & Decision Intelligence Problem Reframing: Translate vague business or scientific objectives into precise, analytically solvable research questions and KPIs. Decision Support: Synthesize complex analytical findings into actionable managerial recommendations and strategic guidelines. Industry Specialization: Adapt predictive frameworks to diverse domains such as finance (fraud detection), healthcare (patient outcomes), or manufacturing (predictive maintenance).
4	4. Communication & Professional Ethics Visual Storytelling: Develop interactive dashboards and sophisticated visualizations (using tools like Tableau or Power BI) to communicate data relationships to both expert and non-expert audiences. Ethical Governance: Evaluate the ethical implications of algorithmic bias and ensure compliance with global data privacy standards like GDPR or CCPA. Stakeholder Translation: Articulate complex technical findings as a clear, narrative-driven business case for executive-level decision-makers.

COURSE CONTRIBUTION TO... PROGRAM COMPETENCIES
(Blank : no contribution, 1: least contribution ... 5: highest contribution)

No	Program Competencies	Cont.
Master of Science in Business Intelligence and Data Science Program
1	Demonstrate understanding the value of data driven decision making.	5
2	Graduates will acquire the ability to make informed decisions based on data analysis and interpretation.	5
3	Identify the basic concepts that underpin today’s organizational IT infrastructures, such as concepts of databases, information systems, operations and processes, cloud computing, data warehousing and Big Data, Data Mining and Machine Learning.	2
4	Students will develop advanced skills in data analysis techniques, including statistical analysis, data mining, data visualization, and predictive modeling.	5
5	Apply data mining/analytics (statistical and machine-learning) in order to solve real-world business problems.	5
6	Develop skills related to data analytics pipeline from collection, processing, analysis and interpretation	3
7	Graduates will develop strong communication skills to effectively present complex data analysis findings to diverse stakeholders.	4
8	Effectively communicate to top management the results and implications arising from data analytics, security risk assessments, and emerging technologies.	5
9	Demonstrate professionalism and leadership by taking initiatives within their domain of responsibility while working effectively with other team members.	4
10	The program offers practical training and exposure to industry-standard software and tools used in business intelligence and data analysis.	3

COURSE EVALUATION METHOD

Method	Quantity	Percentage
Presentation	1
Project	1	100
Case Study	10
	Total Percent:	100%

ECTS (ALLOCATED BASED ON STUDENT WORKLOAD)

Activities	Quantity	Duration(Hours)	Total Workload(Hours)
Course Duration (Including the exam week: 16x Total course hours)	16	3	48
Hours for off-the-classroom study (Pre-study, practice)	15	3	45
Mid-terms	3	4	12
Assignments	5	4	20
Final examination	1	0	0
Other	15	5	75
Total Work Load:			200
Total Work Load/25(h):			8
ECTS Credit of the Course:			8

CONCLUDING REMARKS BY THE COURSE LECTURER

In this course, foundational economic principles will be explored, and their application to business and finance will be demonstrated through real-world examples. Student engagement is encouraged, and office hours are made available for further discussion. It is expected that all coursework is completed with academic integrity, and that classroom discussions are conducted with mutual respect, in accordance with the university's Code of Ethics.